Purpose: Lymph node metastasis is the most important predictor of prognosis in esophageal squamous cell carcinoma (ESCC). Recently, KiSS-1 was cloned as a human metastasis suppressor gene, and an orphan G-protein-coupled receptor (hOT7T175) was identified as the endogenous receptor of the KiSS-1 product. However, the clinical importance of KiSS-1 and hOT7T175 gene expression in ESCC remains unclear.
Experimental Design: In this study, total RNA was extracted from tumors and noncancerous epithelia of 71 patients with ESCC who underwent surgical esophageal resection. The expression levels of KiSS-1, hOT7T175, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNAs were analyzed quantitatively by real-time reverse transcription-PCR and compared with the clinical findings.
Results: The mean KiSS-1:GAPDH and hOT7T175:GAPDH ratios of the tumors were 1.2 and 0.3 and were at the same levels as those in the noncancerous epithelia. The loss of KiSS-1 and hOT7T175 gene expression was detected in 38% and 61% of tumors. Loss of KiSS-1 and/or hOT7T175 gene expression was not correlated with tumor size or degree of tumor invasion but was found to be a significant predictor of lymph node metastasis.
Conclusions: Loss of KiSS-1 or hOT7T175 gene expression may be an important biomarker for detection of lymph node metastasis in ESCC.
Esophageal squamous cell carcinoma (ESCC) is one of the most difficult tumors to treat. Even at an early stage of the cancer, many patients develop lymph node metastasis soon after surgery (1 , 2) . If we have suitable biomarkers to identify tumors with high metastatic potential, it should be possible to start effective chemo-radiotherapy soon after surgery for such patients.
Recently, the KiSS-1 gene has been reported to be a novel metastasis suppressor gene in human melanoma and breast carcinoma cells (3 , 4) . The KiSS-1 gene encodes a 54-amino acid peptide that is the endogenous ligand of an orphan G-protein-coupled receptor (hOT7T175). The metastasis suppressor function of the KiSS-1 gene product is reported to act after binding with hOT7T175. After binding with hOT7T175, the KiSS-1 gene product inhibits the chemotaxis, invasion, and metastasis of cells (5) . However, the clinical importance of the expression of these genes remains unclear.
In the present study, we analyzed the quantitative expression levels of KiSS-1 and hOT7T175 mRNAs in esophageal tissues (tumors and noncancerous epithelia) using the real-time reverse transcription-PCR (RT-PCR) method. Moreover, we evaluated whether or not the KiSS-1 gene and hOT7T175 gene expressions may be new biological markers for the malignant potential of ESCC.
MATERIALS AND METHODS
Surgical esophageal resection was performed on 71 patients between 1993 and 2001 [67 men and 4 women; age at time of surgery (mean ± SE), 65.1 ± 1 years; range, 45–84 years]. None of the patients received preoperative chemotherapy or radiation therapy. To identify KiSS-1 and hOT7T175 gene expression in the tissues, we enrolled these patients. Informed consent was obtained from all patients for the subsequent use of their resected tissues. The present study conformed to the ethical standards of the World Medical Association Declaration of Helsinki. Tumors were staged according to the tumor-node-metastasis (TNM) system: stage I; tumor invades lamina propria or submucosa with no lymph node metastasis (n = 12); stage II, tumor invades muscularis propria or adventitia with no lymph node metastasis, or tumor invades lamina propria or submucosa with regional lymph node metastasis (n = 23); stage III, tumor invades adventitia with lymph node metastasis, or tumor invades adjacent structures with or without lymph node metastasis (n = 31); and stage IV; tumor with distant metastasis (n = 5). Curative surgery was performed on 57 patients, and noncurative surgery was performed on 14 patients (1 patient had liver metastasis, 5 patients had extended lymph node metastasis, and 8 patients had local invasion). All of the patients were followed until October 2002. The mean follow-up was 29.9 months (range, 1–106 months). Causes of death were determined from clinical findings. Twenty-five patients were alive in October 2002, and a total of 46 patients had died. Eleven patients died from diseases other than esophageal cancer (6 died from operative complications, and the in-hospital mortality rate was 8.5%), whereas 35 died from recurrence or relapse of ESCC. The sites of recurrence were identified by computed tomography and ultrasonography.
We obtained tumors and noncancerous esophageal epithelium samples, sufficiently distant from the tumors, from 71 patients immediately after resection of the esophagus. Each tissue sample was stored at −80°C until use.
Real-Time RT-PCR Assay.
Total RNA was isolated from the tissues using RNeasy Mini Kits (Qiagen, Hilden, Germany), and cDNA was synthesized from 1 μg of total RNA with Ready-to-Go You-Prime First-Strand Beads (Amersham Pharmacia Biotech Inc., Piscataway, NJ). Primers and the TaqMan probes for KiSS-1, hOT7T175, and glyceraldehyde-3- phosphate dehydrogenase (GAPDH) were synthesized (5) : (a) KiSS-1, 5′-ACTCACTGGTTTCTTGGCAGC-3′ (forward primer), 5′-ACCTTTTCTAATGGCTCCCCA-3′ (reverse primer), and 5′-6-carboxy-fluorescein (FAM)-ACTGCTTTCCTCTGTGCCACCCACT-6-carboxy-tetramethyl-rhodamine (TAMRA)-3′ (probe); (b) hOT7T175, 5′-CGACTTCATGTGCAAGTTCGTC-3′ (forward primer), 5′-CACACTCAATGGCGGTCAGAG-3′ (reverse primer), and 5′-FAM-ACTACATCCAGCAGGTCTCGGTGCAGG-TAMRA-3′ (probe); and (c) GAPDH, 5′-GAAGGTGAAGGTCGGAGTC-3′ (forward primer), 5′-GAAGATGGTGATGGGATTTC-3′ (reverse primer), and 5′-FAM-CAAGCTTCCCGTTCTCAGCC-TAMRA-3′ (probe). The TaqMan probes were labeled with a fluorescent reporter dye (FAM) at the 5′ end and a fluorescent quencher dye (TAMRA) at the 3′ end.
Quantification of gene expression was performed by real-time quantitative RT-PCR [Gene Amp 5700 Sequence Detection System (Perkin-Elmer Applied Biosystems, Foster City, CA)], which uses the 5′ nuclease activity of Taq polymerase to detect PCR amplicons (6 , 7) . The threshold line was set at an Rn of 0.05 (6) . The point at which the amplification plot crossed this threshold was defined as Ct, which represented the cycle number at this point. Standard curves for KiSS-1, hOT7T175, and GAPDH were generated using serial dilutions (containing 160, 80, 40, 20, and 10 ng) of total RNA derived from the normal esophageal epithelium from one case with ESCC. The plots represent the log of the input amount (log ng of total starting RNA) as the X axis and Ct as the Y axis. Equations were derived from the calibration curves (6) . The formulas for KiSS-1, hOT7T175, and GAPDH were as follows: KiSS-1, y = 36.5 − 4.6x (r2 = 0.997); hOT7T175, y = 42.2 − 3.6x (r2 = 0.999); GAPDH, y = 36.4 − 5.8x (r2 = 0.999). The amounts of KiSS-1, hOT7T175, and GAPDH mRNAs of samples were determined from the standard curves. The amounts of KiSS-1 and hOT7T175 mRNAs were determined by dividing the amount of KiSS-1 and hOT7T175 mRNAs by the amount of GAPDH mRNA, and the KiSS-1:GAPDH ratio and hOT7T175:GAPDH ratio were evaluated.
The χ2 and Fisher’s exact probability tests were used to compare the distribution of individual variables among the patient groups. Differences in the numerical data between two groups were evaluated using the Mann-Whitney U test. Survival rates were calculated using the Kaplan-Meier method. The log-rank test was used for comparisons of two survival curves. A multivariate survival analysis was performed using Cox’s proportional-hazard model. The influence of each variable on the mode of lymph node metastasis was assessed by multivariate logistic regression analysis. A P of <0.05 was considered statistically significant.
The KiSS-1:GAPDH and hOT7T175:GAPDH Ratios of Tumors and Noncancerous Esophageal Tissues.
The KiSS-1:GAPDH ratio of 71 carcinomas (mean ± SE, 1.2 ± 0.3; range, 0–17.5; median, 0.3) was not different from that of 71 noncancerous esophageal epithelium samples sufficiently distant from the tumors (mean ± SE, 1.2 ± 0.4; range, 0–24.4; P = 0.383). In addition, the hOT7T175:GAPDH ratio of carcinomas (mean ± SE, 0.3 ± 0.1; range, 0–3.9) was not different from that of noncancerous tissues (mean ± SE, 0.3 ± 0.1; range, 0–6.7; P = 0.705). Fig. 1⇓ shows the relative expression levels of KiSS-1 and hOT7T175 genes in noncancerous epithelia and tumors. The cases with loss of KiSS-1 or hOT7T175 gene expression in tumors were decided based on the levels of the KiSS-1:GAPDH and hOT7T175:GAPDH ratios in 71 noncancerous tissues (mean − 2×SE; Ref. 8 ). The calculated cutoff values of the KiSS-1:GAPDH and hOT7T175:GAPDH ratios of tumors were 0.4 and 0.1, respectively. Loss of KiSS-1 expression (tumor KiSS-1:GAPDH < 0.4) was detected in 27 cases (38%), and loss of hOT7T175 expression (tumor hOT7T175:GAPDH < 0.1) was detected in 44 cases (61%).
Loss of KiSS-1 and hOT7T175 Gene Expression and the Clinicopathological Findings of Patients.
The loss of both KiSS-1 and hOT7T175 gene expression was found in 16 cases. Preservation of both gene expressions was found in 18 (25%) cases. Table 1⇓ shows the correlation between the expression patterns of the KiSS-1 and hOT7T175 genes and the clinicopathological findings of the tumors. Loss of KiSS-1 or hOT7T175 gene expression did not correlate with depth of tumor invasion or lymphatic invasion of cancer cells. However, cases with preserved expression of both genes showed lower incidence of lymph node metastasis than cases with loss of KiSS-1 or hOT7T175 gene expression. Loss of KiSS-1 or hOT7T175 gene expression was found by multivariate logistic regression analysis to be a significantly important factor for lymph node metastasis among factors that were thought to be affecting lymph node metastasis (Table 2)⇓ . Correlation among depth of tumor invasion, lymph node metastasis, and loss of expression of one or both genes (KiSS-1 and hOT7T175) is indicated in Table 3⇓ . Even in T1 cancer, 20% of cases showed lymph node metastasis. Despite the depth of tumor invasion, loss of KiSS-1 and/or hOT7T175 gene expression was detected in 86–100% of primary tumors in cases with lymph node metastasis.
Prognostic Significance of the Loss of KiSS-1 and/or hOT7T175 Gene Expression in ESCC.
In the 65 patients surviving surgery (excluding the 6 patients who died from surgical complications), disease-specific survivals were compared among three groups (group A, 17 patients with tumors with preserved expression of both KiSS-1 and hOT7T175 genes; group B, 34 patients with tumors with loss of one of the two genes; and group C, 14 patients with tumors with loss of both genes). The disease-specific 5-year survival rates were as follows: group A, 68%; group B, 31%; and group C, 32%. The disease-specific survival curves of groups B and C were same, thus we compared the difference in survival curves between group A and group B + C (Fig. 2)⇓ . Depth of tumor invasion and lymph node metastasis were detected as strong prognostic factors in multivariate survival analysis (P = 0.005 and P = 0.044). On the other hand, tumor size and gene expressions of KiSS-1 and hOT7T175 were not detected as prognostic factors (P = 0.217 and P = 0.687) independent from the depth of tumor invasion or lymph node metastasis.
Lee et al. (3 , 4) transfected full-length human KiSS-1 cDNA into human melanoma and breast cancer cells with high metastasis. After i.v. injection of the cells (with or without KiSS-1 transfection) into the tail vein of athymic nude mice, they found that the number of lung metastases was significantly suppressed in the KiSS-1-transfected mice. This result strongly suggests that KiSS-1 is a novel metastasis suppressor. Furthermore, a recent investigation has shown that KiSS-1 encodes a peptide that is a ligand of a novel human G-protein-coupled receptor [GPCR; named hOT7T175, AXOR12, or GPR54, respectively (5)] . Ohtaki et al. (5) named this peptide encoded from the KiSS-1 gene “metastin.” Recently, Stafford et al. (9) found that activation of the receptor by the KiSS-1 peptide leads to activation of G-protein-activated phospholipase C, which suggests a direct coupling of the KiSS-1 peptide to the Gαq-mediated phospholipase C-Ca2+ signaling pathway. Stimulation of this pathway induces the inhibition of cell proliferation and cell migration. These findings indicate that the metastasis suppressor function of the KiSS-1 gene product may act after binding with hOT7T175.
High expression levels of KiSS-1 mRNA and hOT7T175 mRNA were observed in placenta and in pancreatic tissues (5 , 10 , 11) . In many normal tissues, the expression level of KiSS-1 mRNA was recognized to be 2- to 10,000-fold higher than that of hOT7T175 mRNA (5 , 11) . However, significantly high hOT7T175 mRNA expression levels with low KiSS-1 mRNA expression levels were detected in some ovarian cancers (5) . Ohtaki et al. (5) proposed metastin treatment for patients who have tumors with a high expression level of hOT7T175 mRNA. The expression levels of KiSS-1 and hOT7T175 genes in esophageal epithelia and esophageal cancers have not been reported previously. In the present study, we found that the average expression level of KiSS-1 mRNA was 4-fold higher than that of hOT7T175 mRNA in both normal epithelia and ESCCs. However, in 10 cases, high tumor hOT7T175 mRNA and low tumor KiSS-1 mRNA expression were observed. In such cases, metastin treatment may have some prognostic benefit.
Lymph node metastasis is the principal negative prognostic factor in patients with ESCC. It is difficult to determine whether lymph node metastasis has occurred preoperatively, despite new imaging techniques such as positron emission tomography (12) . Moreover, good biological markers for the detection of lymph node metastasis have not been reported previously in ESCC. If we can obtain information about this metastasis from small preoperative bioptic samples, we can perform radical lymphadenectomy even in patients with early-stage ESCC and start chemo-radiotherapy soon after surgery in patients with a high risk of lymph node metastasis. From our study, we found that loss of KiSS-1 or hOT7T175 gene expression in tumors showed significant and close correlation with high incidence of lymph node metastasis and unfavorable patient prognosis. This phenomenon was detected in cases with loss of one or both of the genes. Furthermore, tumors with preserved expression of both genes suppressed lymph node metastasis regardless of the depth of tumor invasion. These findings strongly suggest that expression of both the KiSS-1 and hOT7T175 genes is essential for inhibition of tumor metastasis and indicate that the loss of expression of one or both genes (KiSS-1 or hOT7T175) may play an important role in ESCC metastasis. Shirasaki et al. (13) noted the importance of KiSS-1 down-regulation in the progression of melanoma. Sanchez-Carbayo et al. (14) demonstrated that loss of KiSS-1 gene expression correlated with poor patient survival in bladder cancer. Thus, in other malignant tumors, investigation of the clinical importance of KiSS-1 and hOT7T175 gene expressions has been started.
In the present study, we used real-time RT-PCR to evaluate the quantitative expression levels of the target genes in esophageal tissues. The recent development of real-time RT-PCR technology has made reliable and accurate PCR quantification possible (15) . This technology can monitor the entire PCR reaction from a small volume of tissues, such as bioptic specimens. Thus, analysis of the expression levels of the KiSS-1 and hOT7T175 genes in preoperative bioptic samples by real-time RT-PCR may provide useful information about the existence of lymph node metastasis in patients with ESCC.
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.
Requests for reprints: Masahide Ikeguchi, Division of Operating Room, Faculty of Medicine, Tottori University, 36-1 Nishi-cho, Yonago 683-8504, Japan. Phone: 81-859-34-8111; Fax: 81-859-34-8095; E-mail:
- Received December 12, 2002.
- Revision received December 2, 2003.
- Accepted December 10, 2003.