Clinical Cancer Research CTRC-AACR San Antonio Breast Cancer Symposium
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Clinical Cancer Research 13, 2584-2591, May 1, 2007. doi: 10.1158/1078-0432.CCR-06-1785
© 2007 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data, Lee, et al.
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lee, J. H.
Right arrow Articles by Shin, Y. K.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Lee, J. H.
Right arrow Articles by Shin, Y. K.

Human Cancer Biology

Clinical Significance of CD99 Down-Regulation in Gastric Adenocarcinoma

Jung Hyun Lee1, Seok-Hyung Kim5, Li-Hui Wang1,6, Yoon-La Choi4, Young Chul Kim2, Jin Hee Kim3, Tae Sung Park2, Yun-Chul Hong3 and Young Kee Shin1

Authors' Affiliations: 1 Department of Pharmacy, Research Institute of Pharmaceutical Science, Seoul National University College of Pharmacy; 2 Department of Statistics, College of Nature Science, Seoul National University; 3 Department of Preventive Medicine, Seoul National University College of Medicine; 4 Department of Pathology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea; 5 Department of Pathology, Chungbuk National University College of Medicine, Cheongju, Chungbuk, Korea; and 6 Department of Pharmacology, Shenyang Pharmaceutical University College of Pharmacy, Shenyang, China

Requests for reprints: Young Kee Shin, Molecular Pathology Laboratory, Department of Pharmacy, Seoul National University College of Pharmacy, San 56-1, Sillim-dong, Gwanak-gu, Seoul, 151-742, Korea. Phone: 82-2-880-9126; Fax: 82-2-872-1795; E-mail: ykeeshin{at}snu.ac.kr.


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
Purpose: CD99 is a cell adhesion molecule associated with human tumors. The aim of the present study was to characterize its role in the development and progression of human gastric adenocarcinoma.

Experimental Design: The expression of CD99 was investigated in 283 gastric adenocarcinomas and related lesions and 9 gastric carcinoma cell lines. We also analyzed the methylation status of CD99 gene by using methylation-specific PCR and examined loss of heterozygosity (LOH) of this gene locus by using an intragenic marker. Moreover, we assessed whether SP1, a positive transcription factor for CD99, is expressed in these samples.

Results: We found that the decreased expression of CD99 was strongly associated with poor survival and unfavorable clinicopathologic variables. Promoter region methylation (15 of 89, 16.9%) and LOH (21 of 74, 28.4%) were observed and significantly associated with CD99 down-regulation (P < 0.05). In addition, most of the gastric adenocarcinoma cases with CD99 down-regulation had reduced expression of SP1 (47 of 103, 45.6%; P < 0.01). This relationship between CD99 and SP1 was consolidated by using SP1 small interfering RNA transfection experiment and CD99 promoter luciferase assay. Furthermore, we showed that CD99 down-regulation was associated with proliferation and migration in gastric carcinoma cell line.

Conclusion: These observations suggest that CD99 down-regulation is a critical event in the progression of gastric adenocarcinoma, and CD99 promoter methylation, CD99 LOH, and SP1 down-regulation were responsible for the down-regulation of CD99.

Gastric carcinoma is the second leading cause of cancer mortality in the world (1). A number of cell adhesion molecules have been examined as potential clinical biomarkers of gastric adenocarcinoma, including CDH1 (E-cadherin), FXYD5 (dysadherin), TACSTD1 (Ep-CAM), and the CD44 splicing variants CD44v6 and CD44v9 (2–6). A germline mutation in CDH1 has been frequently observed in hereditary gastric carcinoma, and multivariate analyses have revealed that CDH1 down-regulation is an independent prognostic factor for gastric cancer (7). In addition, increased expression of FXYD5, a CDH1 inactivator, has been significantly associated with poor prognosis (8, 9). TACSTD1 down-regulation and CD44v6 up-regulation have been also suggested to be prognostic markers for gastric cancer (5, 6).

Another interesting adhesion molecule is CD99, which is a sialomucin-type glycoprotein that is ubiquitously expressed, albeit to varying degrees, by different cell types (10). Although its function and signaling pathway are not fully understood, it is known that CD99 is involved in a number of cellular events such as cell-cell adhesion (11, 12), maintenance of cellular morphology (13), and cell death (14, 15). CD99 also regulates the adhesion and diapedesis of leukocytes to inflamed vascular endothelium (16). Recently, it was proposed that CD99 down-regulation, induced by the EBV-encoded latent membrane protein-1, in B cells is involved in the pathogenesis of Hodgkin's disease (17). Furthermore, the increased or decreased expression of CD99 has been also suggested to act as a marker of various kinds of tumors, including Ewing's sarcoma/primitive neuroectodermal tumor (18), lymphoblastic lymphoma/leukemia (19), some rhabdomyosarcomas (20), granulose cell tumor and sertoli-leydig cell tumor of the ovary (21), pancreatic endocrine tumors (22), gallbladder carcinoma (23), and stomach carcinoma (24). Based on these previous reports, we speculated that altered CD99 expression could also be a prognostic marker for gastric carcinoma patients.

In the present study, we examined the expression of CD99 by normal gastric samples, gastric carcinomas, and other gastric lesions by using immunohistochemistry-based tissue microarray analysis and explored the clinical significance of CD99 down-regulation in gastric adenocarcinoma. In addition, we evaluated the mechanism by which CD99 becomes down-regulated in gastric carcinoma by assessing the methylation status of the CD99 promoter, loss of heterozygosity (LOH) at the CD99 gene locus, and expression of SP1, which is a transcription factor that induces CD99 expression. Furthermore, we showed the pathophysiologic role of CD99 down-regulation in gastric carcinoma.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
Patients, tissue samples, immunohistochemistry, and evaluation of results. Tissues from patients with gastric carcinoma and the other gastric lesions, consisting of 114 gastric carcinomas, 20 high-grade dysplasias, 17 low-grade dysplasias, 48 intestinal metaplasias, 30 chronic atrophic gastritis, and 54 normal gastric epithelium, were used for tissue microarray construction as previously described (25). Ethics committee permission for this study was obtained from the institutional review board of Chungbuk National University Hospital. Immunostaining of CD99 and SP1 was done on consecutive sections, using two anti-CD99 monoclonal antibodies (clones DN16 and YG32; DiNonA) and rabbit polyclonal anti-SP1 antibody (Santa Cruz Biotechnology, Inc.). The evaluation of both the intensity of immunohistochemical staining and the proportion of positively stained epithelial cells was done as previously described (25). A detailed description of the analysis is provided in Supplementary Materials and Methods.

Methylation-specific PCR and LOH assay. Genomic DNA extraction from the tissues microdissected from slides, normal lymphocytes, and gastric cancer cells; bisulfate treatment of DNA; and methylation-specific PCR were done as previously described (25, 26). DNA from normal lymphocytes methylated in vitro with SssI methylase was used as the positive control, and both water blanks and PCR mixtures without template were used as negative controls. All methylated and representative unmethylated PCR fragments were sequenced (for detail, see Supplementary Materials and Methods). Primers and PCR conditions used are also provided in Supplementary Table S1. Briefly, we designed two primer sets [i.e., (A) primers that amplify an area in the promoter region (–273 bp to –57 bp from the transcription initiation site) and (B) primers that amplify an area that is downstream of the transcription initiation site (–105 bp to +106 bp)].

LOH was assessed by subjecting tumor and normal DNA to PCR amplification of DYS403, which is a microsatellite locus that is located between exons 1 and 2 of the CD99 gene on the short arm of chromosomes Xp22.3 and Yp11.3.7 PCR amplification and analysis were done as previously described (for detail, see Supplementary Materials and Methods; ref. 27).

Cell lines, Western blotting, and flow cytometric analysis. Nine human stomach cancer cell lines (i.e., SNU5, SNU16, SNU216, SNU484, SNU638, MKN28, MKN74, KATOIII, and AGS) were maintained. Western blotting was done by using primary antibodies against CD99 and SP1 and ACTB (see Supplementary Materials and Methods), and CD99 and SP1 expression were semiquantified by scanning densitometry with ACTB normalization. For flow cytometric analysis, 106 cells were incubated with FITC-conjugated CD99 (YG32) for 20 min at 4°C, washed three times with cold PBS, fixed with 1% paraformaldehyde, and analyzed on a FACSCalibur (BDIS) by using the CellQuest software (BDIS).

Luciferase assay. CD99 promoter (–1641/+123) reporter and SP1 expression constructs were gifts from Dr. I.S. Lee (Konkuk University, Korea) and Dr. M.W. Hur (Yonsei University School of Medicine, Korea), respectively. Luciferase assay was done at SNU484, SNU638, MKN74, KATOIII, and AGS cells according to standard procedures (see Supplementary Materials and Methods).

SP1 and CD99 small interfering RNA transfection and real-time reverse transcription-PCR. The sense sequences of SP1 small interfering RNA (siRNA; Santa Cruz Biotechnology), CD99 type I–specific siRNA, and green fluorescent protein siRNA (control) are 5'-AAUGAGAACAGCAACAACUCC-3', 5'-GAAAGGCTGGCCATTATTA-3', and 5'-GGCUACGUCCAGGAGCGCACC-3', respectively. Transfections were done according to standard procedures (see Supplementary Materials and Methods). Real-time quantitative reverse transcription-PCR was carried out using the LightCycler Real-time PCR Detection system (Roche Diagnostics). PCR primers for amplifying CD99 (Type I) and the control gene hypoxanthine-guanine phosphoribosyl transferase (HPRT) are shown in Supplementary Table S1. CD99 (Type I) and HPRT were detected in the same tube at 530 and 705 nm by using FAM-conjugated Universal Probe Library probe 21 and DYXL-labeled HPRT probe (TIB MOLBIOL GmbH) as the Taqman probe, respectively.

Establishment of CD99-overexpressing stable cell line and migration assay. CD99 and empty vector were transfected into SNU216, SNU484, KATOIII, and AGS by using a LipofectAMINE LTX (Invitrogen). Transfected cells were passaged after 1 day and selected with 400 µg/mL G418 for 10 days. AGS and SNU484 were just established into CD99-overexpressing stable lines. Cell migration assay was done using established stable cell lines according to standard procedures (see Supplementary Materials and Methods).

Statistical analysis. Means were compared by the nonparametric methods Wilcoxon's rank sum test and Kruskal-Wallis test. The association of the expression rate with clinicopathologic factors was assessed by Fisher's exact test and Pearson's {chi}2 test. Kaplan-Meier analyses were done for the survival times of gastric adenocarcinoma patients. Log-rank tests were used to compare the survival curves between groups. The contribution of each mechanism to protein down-regulation was determined by using the logistic regression model. Accuracy with which each mechanism predicts protein down-regulation was provided in Supplementary Materials and Methods. P < 0.05 was regarded statistically significant. All statistical analyses were done by using SPSS software (SPSS).


   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
Expression of CD99 by gastric carcinoma and other gastric lesions. The expression of CD99 by gastric adenocarcinoma and other gastric lesions was determined by immunohistochemical analysis. In normal gastric epithelia, CD99 expression was strongest around the mucus neck cells and decreases gradually as these cells differentiate into mucus and glandular cells (Supplementary Figure S1A). Notably, CD99 was constitutively expressed at the basolateral border of epithelial cells (Supplementary Figure S1B). CD99 expression was found to be significantly decreased in the gastric carcinoma and intestinal metaplasia samples in terms of both the average immunoreactive score and the expression rate (Fig. 1A ; Supplementary Fig. S2). Comparing with normal gastric epithelia, there is no significant change in low-grade dysplasia and high-grade dysplasia, but a slightly decease in early gastric cancer and an obvious decrease in advanced gastric cancer (Fig. 1A; Supplementary Table S2). Only 34.5% (39 of 113) of the carcinoma cases showed CD99 expression, whereas almost 90% of the normal gastric mucosa, chronic atrophic gastritis, and dysplasia samples expressed CD99. Of the intestinal metaplasia samples, 47.9% (23 of 48) were positive for CD99 expression. In addition, CD99 expression was completely absent in the complete type of intestinal metaplasia but present in the incomplete type (Supplementary Figure S1C and D).


Figure 1
View larger version (51K):
[in this window]
[in a new window]

  		Fig. 1. Immunohistochemical analysis and survival analysis of CD99 expression by gastric carcinoma samples. A, gastric carcinoma (n = 113), high-grade dysplasia (HD; n = 20), low-grade dysplasia (LD; n = 17), intestinal metaplasia (IM; n = 48), chronic atrophic gastritis (CAG; n = 30), and normal gastric epithelium (NL; n = 54) samples were subjected to immunohistochemical analysis of CD99 expression. Left, average immunoreactive score (IS) for each type of lesion. Right, frequency of samples positive for CD99 expression. A sample is defined as CD99+ if it has an immunoreactive score ≥1. B, relationship between the immunoreactive scores of individual gastric carcinoma sample and its stage. The gastric carcinoma patients were divided according to their stage (I-IV), and the immunoreactive scores of their samples are shown in colors. C, CD99 expression for representative gastric adenocarcinoma samples at each carcinoma stage. Membranous staining of CD99 was frequently observed in the early stages (I and II) but was infrequent in higher-stage (III and IV) samples. D, Kaplan-Meier curves of the overall survival of gastric carcinoma patients in early stages (I and II) and advanced stages (III and IV) whose cancer samples are positive (immunoreactive score ≥1) or negative (immunoreactive score <1) for CD99 expression.

 
Down-regulation of CD99 in gastric carcinoma is associated with unfavorable clinicopathologic variables and poor prognosis. How CD99 expression in gastric carcinoma relates to various clinicopathologic variables was analyzed. Reduced expression of CD99 was significantly associated with all unfavorable variables, including greater tumor size, nodal metastasis, advanced gastric cancer, diffuse type, poor differentiation (high grade), deep tumor invasion, and advanced clinical staging (tumor-node-metastasis staging; Fig. 1B and C; Supplementary Table S2). These findings were observed for both the expression rate and the average immunoreactive score.

Kaplan-Meier survival analysis revealed that, like other unfavorable prognostic factors, reduced CD99 expression was significantly associated with decreased 1-year and 3-year overall survival (P < 0.01; Table 1 ). The mean survival time was 60.1 months for CD99– gastric carcinoma patients (74 cases) and 81.0 months for CD99+ patients (39 cases; P < 0.01). Analysis of the overall survival showed that the presence of CD99 expression was significantly associated with better prognosis (risk ratio, 0.313; P < 0.01). Moreover, Kaplan-Meier survival analysis that was separately done for the early-stage (I and II) and advanced stage (III and IV) groups showed that the absence of CD99 expression was significantly associated with shorter survival times in advanced stage patients (P < 0.05; Fig. 1D).


View this table:
[in this window]
[in a new window]

  		Table 1. Univariate analysis of the overall survival of stomach cancer patients by Kaplan-Meier survival analysis

 
Hypermethylation of the CD99 promoter region and LOH of the CD99 gene locus is associated with CD99 down-regulation. To elucidate the mechanism by which CD99 is down-regulated in gastric carcinoma, we explored the methylation status of the CD99 promoter region by using methylation-specific PCR and LOH of the CD99 gene locus. When the (A) primers were employed, methylation was found in 15 of the 89 cases of gastric cancer (16.9%; Fig. 2A-C ), and a significant correlation between promoter methylation and the absence of CD99 expression was observed (P < 0.05; Tables 1 and 2 ). In contrast, when the (B) primers were used, methylation was rarely detected (4 of 68, 5.9%), and no significant association with CD99 down-regulation was found (Tables 1 and 2). In addition, methylation in the promoter was associated with decreased 1-year/3-year survival (P < 0.01; Table 1), whereas methylation in the downstream region showed no such correlation. Despite these clinicopathologic implications of CD99 promoter methylation, we could not detect similar promoter methylation in nine gastric adenocarcinoma cell lines (data not shown).


Figure 2
View larger version (26K):
[in this window]
[in a new window]

  		Fig. 2. Methylation status of the CD99 promoter region and LOH status of CD99 in gastric carcinoma. A, schematic depiction of the region of the CD99 gene between –400 and +200 bp relative to the transcription start site. Brown arrows, primary PCR primers; black arrows, secondary PCR primers; purple arrows, SP1-binding sites; black circles, CpGs. B, methylation-specific PCR (MSPCR) of the CD99 promoter in representative gastric carcinoma cases. U, nonmethylated; M, methylated; PC, positive control; DW, distilled water. Methylation-specific PCR was done by using sequences upstream (A) and downstream (B) of the transcription start site. C, representative sequences of the methylation-specific PCR (A) and (B) products. Arrows, unmethylated and methylated sites. Although the sequence from the unmethylated DNA sample shows the C-to-T conversion (arrows), the corresponding sequence of the methylated case shows the retention of Cs (arrows). D, schematic depiction of the CD99 gene locus showing the location of the microsatellite marker DYS403. E, representative results after assessing normal gastric muscosa (N) and tumor specimens (T) from two gastric carcinoma patients. Arrow, LOH.

 

View this table:
[in this window]
[in a new window]

  		Table 2. Analysis of the effect of promoter methylation, LOH, and SP1 down-regulation on CD99 expression by using the logistic regression model

 
LOH of the CD99 gene locus was found in 21 of 74 cases (28.4%) and was significantly associated with down-regulation of CD99 expression (P < 0.05; Table 2). In addition, LOH of the CD99 gene locus was associated with decreased 1-year/3-year survival (P < 0.05; Table 1). Analysis of nine cell lines revealed that SNU638 and AGS also showed LOH of the CD99 gene locus (data not shown).

SP1 expression in gastric carcinoma was associated with CD99 expression. Whether CD99 down-regulation in stomach carcinoma is due to altered expression of SP1 was assessed by immunohistochemical analysis because the promoter region of CD99 contains several SP1-binding sites (Fig. 2A; refs. 28, 29). Down-regulation of SP1 in gastric carcinoma was found in 47 of 113 cases (41.6%) and was significantly associated with the down-regulation of CD99 expression (P < 0.05; Table 2). However, decreased expression of SP1 was not associated with decreased 1-year/3-year survival (Table 1).

When the protein expression of CD99 and SP1 were semiquantified, the SNU5, SNU16, MKN28, and MKN74 cell lines showed high expression of both CD99 and SP1 (Fig. 3A and B ). However, the amount of CD99 transcript in SNU484, SNU216, KATOIII, and AGS did not correlate with the protein expression of Sp1 (Fig. 3B). When CD99 promoter assay was done, the discrepancy of SP1 and CD99 observed in SNU484 and KATOIII could be explained. In the case of SNU484, CD99 promoter was activated by itself, and SP1 inhibited CD99 promoter activity (Fig. 3C). On the other hand, CD99 promoter was not activated by Sp1 in KATOIII (Fig. 3C). In SNU638, MKN74, and AGS, SP1 activated CD99 promoter (Fig. 3C). Furthermore, when SP1 siRNA was introduced into in the high CD99- and SP1-expressing line MKN74, SP1 siRNA dramatically down-regulated CD99 transcription (Fig. 3D-F).


Figure 3
View larger version (45K):
[in this window]
[in a new window]

  		Fig. 3. Transfection of gastric cell lines with SP1 siRNA decreases their CD99 expression. A, CD99 and SP1 protein expression profiles of nine gastric carcinoma cell lines. The Western blot bands were analyzed by scanning densitometry. B, comparison of CD99 mRNA and SP1 protein levels in nine gastric carcinoma cell lines. CD99 mRNA levels were analyzed by real-time PCR with HPRT1 normalization. SP1 protein levels were estimated as described in (A). C, effect of SP1 on the promoter activity of CD99. D, effect of SP1 siRNA on CD99 expression by MKN74. E, fluorescence-activated cell sorting analysis of the CD99 expression of MKN74 transfected with green fluorescent protein (GFP) or SP1 siRNA. Gray line, anti-CD99 antibody–stained cells; black profile, control antibody–stained cells; R.I., relative intensity. F, immunoblot analysis of the CD99 and SP1 expression of MKN74 cells transfected with green fluorescent protein or SP1 siRNA. G, fluorescence-activated cell sorting analysis of the CD99 expression in CD99-overexpressing stable AGS. Gray line, CD99; black profile, control. H, migration assay of CD99-overexpressing stable AGS. I, effect of CD99 in proliferation rate. J, quantitative real-time PCR analysis of CD99 expression. K, effect of CD99 knockdown on migration by MKN74.

 
CD99 down-regulation was associated with proliferation and migration in gastric carcinoma cell line. CD99-overexpressing AGS stable cell line showed reduced migration and cell proliferation (Fig. 3G-I). In accordance with, CD99 knockdown in MKN74 increased migration (Fig. 3J and K). However, apoptosis was not induced in transiently CD99-transfected KATOIII (data not shown). These results showed that CD99 may have essential role in migration and proliferation in gastric carcinoma.

Contribution of SP1 down-regulation, LOH, and promoter methylation to CD99 down-regulation in gastric adenocarcinoma. To assess the effect of promoter methylation, LOH, and SP1 down-regulation on CD99 expression, we did multivariate logistic regression analysis using these variables. We found that all variables were significantly associated with CD99 expression (Table 2). Notably, promoter methylation and CD99 LOH showed high odds ratio for CD99 down-regulation (Table 2). These changes were not observed in most of cases without showing CD99 down-regulation (1 of 28, specificity = 96.4% for promoter methylation and 2 of 22, specificity = 96.4% for CD99 LOH), suggesting that these mechanisms are specifically related to CD99 down-regulation (Supplementary Materials and Methods). In addition, SP1 down-regulation were responsible for more cases showing CD99 down-regulation than methylation and LOH (Supplementary Materials and Methods). Furthermore, decreased expression of SP1, CD99 LOH, and CD99 promoter methylation can together predict about 81% of the gastric carcinoma cases with CD99 down-regulation, suggesting that these constitute the major mechanism of inducing it (Supplementary Materials and Methods).


   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
Here, we studied the expression of CD99 in gastric carcinoma samples and determined its clinical significance, particularly with regard to the prognosis of the patients. We found that the expression of CD99 was clearly down-regulated in gastric carcinoma, particularly for advanced stage (P < 0.001). In addition, the loss of CD99 expression was strongly associated with a poor prognosis (P < 0.001). We also analyzed the potential mechanisms that could be responsible for CD99 down-regulation in gastric carcinoma and found that methylation of the CD99 promoter and CD99 LOH can account for the cases with CD99 down-regulation. One of supporting evidence for this is our observation that the SNU638 cell line, which shows CD99 LOH, also expresses CD99 at low levels. Finally, we found that loss of SP1 expression, which is a potent positive regulator of CD99, accounts for many of the CD99– gastric carcinoma cases. This was revealed by immunohistochemical analysis, CD99 promoter assay, and an in vitro siRNA transfection study.

Although CD99 has been suggested to participate in various physiologic processes, its exact functions, particularly those in epithelial cells, remain unclear. Unlike lymphoid cells, epithelial cells have been rarely reported to express CD99. Here, we found that in normal gastric epithelia, CD99 is constitutively expressed at their basolateral border. The location and expression pattern of CD99 and its down-regulation in gastric carcinoma progression suggest that this molecule could act as an intercellular adhesion molecule. The role of CD99 as an intercellular adhesion molecule has already been previously documented in T cells (11, 12). Moreover, upon CD99 stimulation, an Ewing's sarcoma cell line forms cell junctions that resemble adherens junctions (30). It is known that CDH1 down-regulation is critical for the formation of adherens junctions and gastric carcinogenesis (7). Thus, it is possible that CD99 signaling is involved in epithelial cell-epithelial cell adhesion, similarly to CDH1.

With regard to gastric carcinoma progression, the loss of CD99 expression can be considered as a later event, supporting the suggestion that gastric carcinoma cells arise not from metaplastic cells but from gastric cells and then lose their identity as tumor progression (31). First, the loss of CD99 expression was also observed in intestinal metaplasias of complete type, in which the gastric cell identity was lost, through the gastritis-dysplasia-carcinoma sequence, suggesting that the intestinal metaplasia should be recognized as a paraneoplastic phenomenon (31). Second, this late event may be responsible for the loss of intercellular adhesion capability. CD99 down-regulation was clearly associated with increased direct tumor invasion and nodal metastasis. This agrees well with our previous clinicopathologic study for the loss of CD99 expression in gallbladder carcinomas (23). Notably, the CD99-overexpressing AGS cell line showed reduced migration and proliferation, and CD99 knockdown in MKN74 increased migration. In accordance with, the forced expression of CD99 in the osteosarcoma cell lines significantly inhibited cell migration and abrogated their metastatic ability (32). Third, it is possible that CD99 down-regulation is related to the loss of cellular identity or differentiation. We found that loss of CD99 expression by tumor cells was clearly associated with their loss of cellular polarity and their failure to maintain a glandular structure in gastric carcinoma cases. It has also been shown that the CD99 down-regulation leads to the loss of normal morphology in Hodgkin's disease (13, 17). Fourth, because the mouse CD99 orthologue has been found to specifically activate natural killer cells and dendritic cells by binding to PILRB (33, 34), the loss of CD99 in gastric carcinoma may serve as an immune escape mechanism.

When the contributions of CD99 promoter methylation, CD99 LOH, and SP1 down-regulation to CD99 down-regulation in gastric carcinoma were analyzed by using a logistic regression model, it was found that the CD99 down-regulation can be explained much better by all three potential causes together than by just one. It was shown that CD99 promoter methylation and LOH specifically occur when CD99 is down-regulated. In contrast, whereas over half of the cancer cases with CD99 down-regulation had decreased expression of SP1, SP1 down-regulation was also observed in some CD99+ cases. To confirm the clinicopathologic study about the relationship between CD99 and SP1, we did CD99 promoter assay and verified that SP1 may be critical for CD99 promoter activity in several cancer cell lines. Furthermore, the combination of CD99 promoter methylation, CD99 LOH, and SP1 down-regulation can account for nearly 81% of the gastric cancer cases with CD99 down-regulation. It is likely that the remaining 19% of cases may be explained by additional mechanisms that lead to the loss of CD99 expression, such as mutations, posttranslational modifications, and additional derangements of other transcription factors. We also suggest that CD99 mutations could be responsible to some degree for CD99 down-regulation as the KATOIII cell line contains a point mutation or an unknown single nucleotide polymorphism in its 5' untranslated region and expresses little CD99 mRNA. It is possible that such CD99 alteration may decrease mRNA stability (Supplementary Materials and Methods; Supplementary Table S3).

In conclusion, this study revealed that loss of CD99 expression in gastric carcinoma is clearly associated with poor prognosis and unfavorable clinicopathologic variables. Most of the cases with loss of CD99 expression can be explained by a combination of loss of SP1 expression, LOH of the CD99 gene locus, and methylation of the CD99 promoter region. We also showed that CD99 down-regulation was associated with proliferation and migration in gastric carcinoma cell line. These observations suggest that the loss of CD99 expression is a critical event in gastric carcinoma progression.


   Acknowledgments
 
We thank Jung Sun Lee and Mee Young Sim for their technical assistance, including slide cutting and immunohistochemistry.


   Footnotes
 
Grant support: Seoul R&BD Program, Korea Health 21 R&D Project of the Ministry of Health and Welfare grant A050260 Korea, and National Research Laboratory Program of Korea Science and Engineering Foundation grant M10500000126 (T.S. Park).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/).

J.H. Lee and S.H. Kim contributed equally to this work.

7 Genome Database (http://www.ensembl.org/). Back

Received 7/21/06; revised 1/25/07; accepted 2/19/07.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

  1. Crew KD, Neugut AI. Epidemiology of upper gastrointestinal malignancies. Semin Oncol 2004;31:450–64.[CrossRef][Medline]
  2. Yasui W, Oue N, Aung PP, Matsumura S, Shutoh M, Nakayama H. Molecular-pathological prognostic factors of gastric cancer: a review. Gastric Cancer 2005;8:86–94.[CrossRef][Medline]
  3. Mayer B, Jauch KW, Gunthert U, et al. De-novo expression of CD44 and survival in gastric cancer. Lancet 1993;342:1019–22.[CrossRef][Medline]
  4. Saito H, Tsujitani S, Katano K, Ikeguchi M, Maeta M, Kaibara N. Serum concentration of CD44 variant 6 and its relation to prognosis in patients with gastric carcinoma. Cancer 1998;83:1094–101.[CrossRef][Medline]
  5. Songun I, Litvinov SV, van de Velde CJ, Pals ST, Hermans J, van Krieken JH. Loss of Ep-CAM (CO17-1A) expression predicts survival in patients with gastric cancer. Br J Cancer 2005;92:1767–72.[CrossRef][Medline]
  6. Yamamichi K, Uehara Y, Kitamura N, Nakane Y, Hioki K. Increased expression of CD44v6 mRNA significantly correlates with distant metastasis and poor prognosis in gastric cancer. Int J Cancer 1998;79:256–62.[CrossRef][Medline]
  7. Wang HD, Ren J, Zhang L. CDH1 germline mutation in hereditary gastric carcinoma. World J Gastroenterol 2004;10:3088–93.[Medline]
  8. Ino Y, Gotoh M, Sakamoto M, Tsukagoshi K, Hirohashi S. Dysadherin, a cancer-associated cell membrane glycoprotein, down-regulates E-cadherin and promotes metastasis. Proc Natl Acad Sci U S A 2002;99:365–70.[Abstract/Free Full Text]
  9. Shimada Y, Yamasaki S, Hashimoto Y, et al. Clinical significance of dysadherin expression in gastric cancer patients. Clin Cancer Res 2004;10:2818–23.[Abstract/Free Full Text]
  10. Gelin C, Aubrit F, Phalipon A, et al. The E2 antigen, a 32 kd glycoprotein involved in T-cell adhesion processes, is the MIC2 gene product. EMBO J 1989;8:3253–9.[Medline]
  11. Bernard G, Zoccola D, Deckert M, Breittmayer JP, Aussel C, Bernard A. The E2 molecule (CD99) specifically triggers homotypic aggregation of CD4+ CD8+ thymocytes. J Immunol 1995;154:26–32.[Abstract]
  12. Cassuto-Viguier E, Ticchioni M, Bernard-Pomier G, et al. Differential effects of antilymphocyte globulins on lymphocytic molecules (E2 and CD2) implicated in the rosette phenomenon. Transplant Proc 1993;25:2322–4.[Medline]
  13. Kim SH, Choi EY, Shin YK, et al. Generation of cells with Hodgkin's and Reed-Sternberg phenotype through downregulation of CD99 (Mic2). Blood 1998;92:4287–95.[Abstract/Free Full Text]
  14. Bernard G, Breittmayer JP, de Matteis M, et al. Apoptosis of immature thymocytes mediated by E2/CD99. J Immunol 1997;158:2543–50.[Abstract]
  15. Pettersen RD, Bernard G, Olafsen MK, Pourtein M, Lie SO. CD99 signals caspase-independent T cell death. J Immunol 2001;166:4931–42.[Abstract/Free Full Text]
  16. Schenkel AR, Mamdouh Z, Chen X, Liebman RM, Muller WA. CD99 plays a major role in the migration of monocytes through endothelial junctions. Nat Immunol 2002;3:143–50.[CrossRef][Medline]
  17. Kim SH, Shin YK, Lee IS, et al. Viral latent membrane protein 1 (LMP-1)-induced CD99 down-regulation in B cells leads to the generation of cells with Hodgkin's and Reed-Sternberg phenotype. Blood 2000;95:294–300.[Abstract/Free Full Text]
  18. Stevenson AJ, Chatten J, Bertoni F, Miettinen M. CD-99 (p30/p32 MIC-2) neuroectodermal/Ewing's sarcoma antigen as an immunohistochemical marker. Appl Immunohistochem 1994;2:231–40.
  19. Dworzak MN, Fritsch G, Fleischer C, et al. CD99 (MIC2) expression in paediatric B-lineage leukaemia/lymphoma reflects maturation-associated patterns of normal B-lymphopoiesis. Br J Haematol 1999;105:690–5.[CrossRef][Medline]
  20. Folpe AL, McKenney JK, Bridge JA, Weiss SW. Sclerosing rhabdomyosarcoma in adults: report of four cases of a hyalinizing, matrix-rich variant of rhabdomyosarcoma that may be confused with osteosarcoma, chondrosarcoma, or angiosarcoma. Am J Surg Pathol 2002;26:1175–83.[CrossRef][Medline]
  21. Choi YL, Kim HS, Ahn G. Immunoexpression of inhibin alpha subunit, inhibin/activin betaA subunit and CD99 in ovarian tumors. Arch Pathol Lab Med 2000;124:563–9.[Medline]
  22. Goto A, Niki T, Terado Y, Fukushima J, Fukayama M. Prevalence of CD99 protein expression in pancreatic endocrine tumours (PETs). Histopathology 2004;45:384–92.[CrossRef][Medline]
  23. Choi YL, Xuan YH, Shin YK, et al. An immunohistochemical study of the expression of adhesion molecules in gallbladder lesions. J Histochem Cytochem 2004;52:591–601.[Abstract/Free Full Text]
  24. Jung KC, Park WS, Bae YM, et al. Immunoreactivity of CD99 in stomach cancer. J Korean Med Sci 2002;17:483–9.[Medline]
  25. Wang LH, Kim SH, Lee JH, et al. Inactivation of SMAD4 tumor suppressor gene during gastric carcinoma progression. Clin Cancer Res 2007;13:102–110.[Abstract/Free Full Text]
  26. Wang LH, Choi YL, Hua XY, et al. Increased expression of sonic hedgehog and altered methylation of its promoter region in gastric cancer and its related lesions. Mod Pathol 2006;19:675–83.[CrossRef][Medline]
  27. Wang C, Horiuchi A, Imai T, et al. Expression of BRCA1 protein in benign, borderline, and malignant epithelial ovarian neoplasms and its relationship to methylation and allelic loss of the BRCA1 gene. J Pathol 2004;202:215–23.[CrossRef][Medline]
  28. Lee I, Kim MK, Choi EY, et al. CD99 expression is positively regulated by Sp1 and is negatively regulated by Epstein-Barr virus latent membrane protein 1 through nuclear factor-kappaB. Blood 2001;97:3596–604.[Abstract/Free Full Text]
  29. Lee IS, Shin YK, Chung DH, Park SH. LMP1-induced downregulation of CD99 molecules in Hodgkin and Reed-Sternberg cells. Leuk Lymphoma 2001;42:587–94.[Medline]
  30. Cerisano V, Aalto Y, Perdichizzi S, et al. Molecular mechanisms of CD99-induced caspase-independent cell death and cell-cell adhesion in Ewing's sarcoma cells: actin and zyxin as key intracellular mediators. Oncogene 2004;23:5664–74.[CrossRef][Medline]
  31. Tatematsu M, Tsukamoto T, Inada K. Stem cells and gastric cancer: role of gastric and intestinal mixed intestinal metaplasia. Cancer Sci 2003;94:135–41.[CrossRef][Medline]
  32. Manara MC, Bernard G, Lollini PL, et al. CD99 acts as an oncosuppressor in osteosarcoma. Mol Biol Cell 2006;17:1910–21.[Abstract/Free Full Text]
  33. Shiratori I, Ogasawara K, Saito T, Lanier LL, Arase H. Activation of natural killer cells and dendritic cells upon recognition of a novel CD99-like ligand by paired immunoglobulin-like type 2 receptor. J Exp Med 2004;199:525–33.[Abstract/Free Full Text]
  34. Park SH, Shin YK, Suh YH, et al. Rapid divergency of rodent CD99 orthologs: implications for the evolution of the pseudoautosomal region. Gene 2005;353:177–88.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data, Lee, et al.
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lee, J. H.
Right arrow Articles by Shin, Y. K.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Lee, J. H.
Right arrow Articles by Shin, Y. K.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online