Published ahead of print on October 5, 2006, doi:10.1164/rccm.200602-212OC
© 2007 American Thoracic Society doi: 10.1164/rccm.200602-212OC
Endogenous Secretory Receptor for Advanced Glycation End Products in Non–Small Cell Lung CarcinomaDepartments of Geriatric and Respiratory Medicine, and Department of Pathology, Tohoku University School of Medicine; Department of Thoracic Surgery, Institute of Development, Aging, and Cancer, Tohoku University; Sendai Kousei Hospital, Sendai; and Department of Biochemistry and Molecular Vascular Biology, Kanazawa University Graduate School of Medical Science, Kanazawa, Japan Correspondence and requests for reprints should be addressed to Hiroshi Kubo, M.D., Ph.D., Department of Geriatric and Respiratory Medicine, Tohoku University School of Medicine, 1-1 Seiryoumachi, Aobaku, Sendai 980-8574, Japan. E-mail: hkubo{at}geriat.med.tohoku.ac.jp
Rationale: The receptor for advanced glycation end products is a multiligand receptor that plays an important role in regulating the invasiveness and metastatic potential of cancer cells. A recently discovered novel splice variant, the endogenous secretory receptor for advanced glycation end products, mediates the receptor for advanced glycation end-product–associated cell responses by functioning as a decoy receptor. Objectives: To evaluate the expression pattern of endogenous secretory receptor for advanced glycation end products in non–small cell lung carcinoma, and analyze its impact on prognosis. Methods: We performed immunohistochemical evaluation in 182 non–small cell lung carcinoma surgical specimens. The effect of an overexpressed receptor in cancer cell proliferation was also evaluated. Measurements and Main Results: The endogenous secretory receptor for advanced glycation end-product expression in cytoplasm was reduced or absent in 137 of the 182 (75%) carcinomas in contrast to normal lung tissues. mRNA expression was also suppressed in cancer cells. Overexpression of the secretory receptor in lung cancer cell lines had an inhibitory effect on cell proliferation, suggesting the reduced receptor expression accelerated tumor growth. Among patients with low expression of the cytoplasmic secretory receptor, the overall survival rate was significantly lower than that of patients with normal expression (p = 0.0003). This association was most prominent in TNM stage I patients (p = 0.0001). In a multivariate analysis, endogenous secretory receptor immunoreactivity was an independent prognostic factor with a relative risk of 3.1. Conclusions: The cytoplasmic endogenous secretory receptor for advanced glycation end-product expression has the potential to be a prognostic factor for predicting the outcome of curative surgery in patients with non–small cell lung carcinoma.
Key Words: clinical study • lung cancer • prognostic factor • receptor • surgery
Lung cancer is one of the leading causes of cancer death worldwide. Despite the recent advances in cancer treatment, the relative 5-yr survival rate for all lung cancers is only 15% (1). Lung cancer is histologically classified into non–small cell lung carcinoma (NSCLC) and small cell lung carcinoma. The most effective treatment for early-stage NSCLC is surgical resection; however, even with resection, the 5-yr survival of stage I lung cancer is only 65% (2–4). Currently, stage is the main criterion used to identify surgical candidates. Recently, several randomized trials of adjuvant chemotherapy have demonstrated an improved outcome over surgery alone (5–9). Consequently, it would be beneficial to identify prognostic factors that could be used to identify those patients most likely to benefit from adjuvant chemotherapy.
The receptor for advanced glycation end products (RAGE), a multiligand receptor of the immunoglobulin superfamily (10), recognizes a variety of ligands, including advanced glycation end products (AGEs), amyloid A novel splice variant of RAGE, endogenous secretory RAGE (esRAGE), was recently identified (26). esRAGE is similar to, but more stable than, a previously described form of secretory RAGE (sRAGE) (26, 27). By functioning as a decoy receptor, esRAGE is able to protect vascular cells from injury (28). This suggests that esRAGE is a mediator that controls RAGE-associated cell responses. In this study, we measured the distribution of esRAGE expression in NSCLC and showed that overexpression of RAGE had an inhibitory effect on tumor growth. We also confirmed that esRAGE expression could be used as a prognostic factor to predict a good surgical outcome.
Patients and Tissue Specimens One hundred eighty-two NSCLC surgical specimens were obtained from patients who were treated from 1993 to 1995 by the Department of Surgery, Sendai Kousei Hospital, Sendai, Japan. The mean age of the patients was 65.3 yr (range, 23 – 82 yr). Patients included in this study did not receive radiation therapy or chemotherapy before surgery. The mean follow-up time was 4.4 yr (range, 17–3,695 d). Patients were classified according to the size of the primary tumor (T), nodal involvement (N), and the presence of distant metastasis (M) into TNM stages I–IV (29). Tumor histology was classified according to the World Health Organization classification system (30). Research protocols for this study were approved by the ethics committee at Tohoku University School of Medicine and Sendai Kousei Hospital, Sendai, Japan.
Immunohistochemistry
Laser Capture Microdissection and Reverse Transcriptase–Polymerase Chain Reaction To localize esRAGE gene expression, microdissection was conducted using the Laser Scissors CRI-337 (Cell Robotics, Inc., Albuquerque, NM) as previously described (34). Approximately 100 carcinoma or intratumoral stromal cells were collected separately under the microscope from 36 specimens of frozen NSCLC tissue. Total RNA was extracted from laser-transferred cells, and polymerase chain reaction (PCR) amplification was performed. Real-time quantitative reverse transcriptase (RT)–PCR was performed using a LightCycler (Roche Diagnostics, Basel, Switzerland) and QuantiTect SYBR Green PCR kit (Qiagen K.K., Tokyo, Japan), according to the procedure provided by the manufacturers. The primer sequences used to analyze human esRAGE (AB061668) mRNA expression were forward 5'-AGG AAA GCC GTG CTG TCA-3' and reverse 5'-ACA TGT GTT GGG GGC TAT-3'; primers for full-length RAGE (AB036432) mRNA expression were forward 5'-CAT CAG CAT CAT CGA ACC AG-3' and reverse 5'-AAG ATG ACC CCA ATG AGC AG-3'. The transcript amount for the esRAGE or full-length RAGE gene was estimated from the respective standard curves and normalized to the human glyceraldehyde-3-phosphate dehydrogenase (GAPDH; Hs_GAPDH_1_SG QuantiTect Primer Assay kit; Qiagen) transcript amount determined in corresponding samples.
Transfection, Cell Proliferation Assay, and Cell Cycle Analysis
Statistical Analysis
Immunohistochemistry for esRAGE in NSCLC Consistent with a previous study (31), esRAGE immunoreactivity was readily detectable in the cytoplasm of bronchial epithelial cells in normal lung tissues (Figure 1A). esRAGE immunoreactivity was also positive in the stromal cells (Figure 1A). In contrast, esRAGE immunoreactivity was only variably detected in the cytoplasm of NSCLC cells (Figure 1C). The number of cases expressing immunoreactive esRAGE in each group are summarized as follows: +, n = 45 (24.7%); +/–, n = 78 (42.9%); and –, n = 59 (32.4%). Intratumoral stromal cells were positive for esRAGE. Cytoplasmic esRAGE was frequently negative in male patients (p = 0.0007) and in squamous cell carcinomas (p = 0.0001), and was inversely associated with the pathologic T factor (p = 0.0063), histologic differentiation (p = 0.0061), and Ki-67 labeling index (p < 0.0001) (Table 1). The number of cases expressing immunoreactive full-length RAGE in each group are summarized as follows: +, n = 53 (29.1%); +/–, n = 55 (30.2%); and –, n = 74 (40.7%). However, expression of esRAGE and full-length RAGE were not parallel (Figures 1C–1F), and there was no association between esRAGE and full-length RAGE immunoreactivity in NSCLCs (p = 0.199; Table 2).
esRAGE mRNA Expression To evaluate the localization of esRAGE mRNA expression in NSCLC, we performed laser capture microdissection followed by quantitative RT-PCR for the esRAGE and full-length RAGE mRNA. All samples from intratumoral stromal cells expressed both esRAGE and full-length RAGE mRNA. In contrast to the intratumoral stromal cells, either esRAGE or full-length RAGE mRNA were absent in 81% (29 of 36) of NSCLCs. Expression of esRAGE and full-length RAGE mRNAs were not parallel, and there was no association between esRAGE and full-length RAGE expression in NSCLCs (p = 0.468; Table 3).
Association between Cytoplasmic esRAGE Status and Clinical Outcome of Patients with NSCLC A reduction of esRAGE expression in cytoplasm was significantly associated with a worse clinical outcome in the 182 patients with NSCLC (p = 0.0003) (Figure 2A). This association was most apparent in the patients with TNM stage I NSCLC (p = 0.0001) (Figure 2B). In advanced stage NSCLC (TNM stage II and III), however, cytoplasmic esRAGE expression status was not predictive of clinical outcome (p = 0.2735), although patients with positive (+) esRAGE tended to show a better prognosis (Figure 2C). In contrast, full-length RAGE expression was not associated with overall survival (Figure 3). In a univariate analysis, pathologic TNM stage (p < 0.0001) and esRAGE immunoreactivity (– and +/– vs. +; p < 0.001) were both significant prognostic factors for overall survival (Table 4). In a multivariate analysis, both TNM stage (p < 0.0001) and esRAGE immunoreactivity (p = 0.0016) were independent prognostic factors, with relative risks over 1.0 (Table 4). Impact of the pathological T (pT) stage on survival was not significant (p = 0.0510) by univariate analysis (Table 4). In contrast, the pathological N (pN) stage was a significant prognostic factor for overall survival (p < 0.0001). When the pT and pN stages were incorporated in the multivariate analysis instead of the pTNM stage, the p value of esRAGE immunoreactivity changed to 0.0137 and the relative risk was 2.452 (1.202–5.004).
Role of Cytoplasmic Expression of esRAGE in Cell Proliferation To investigate the role of cytoplasmic esRAGE in cell proliferation, esRAGE, full-length RAGE, or dnRAGE cDNA were cloned into the pCI expression vector (Promega, Fitchburg, WI) and transfected to A549 or LK-1 cells. esRAGE was overexpressed within the cytoplasm of the transfectants (Figure 4B) and resulted in significantly attenuated tumor growth in all cell lines (Figures 4C and 4D). In contrast, full-length RAGE expression had no effect on tumor growth. Finally, neither dnRAGE nor an empty vector affected tumor growth.
As demonstrated in Figure 4E, overexpression of esRAGE resulted in an increment of G0/G1 fractions in both A549 and LK-1 cells. As previously reported (25), overexpression of full-length RAGE also increased the G0/G1 fraction. However, the effect of esRAGE on G0/G1 fraction was greater than that of full-length RAGE.
In this study, we demonstrated that the lack of expression of esRAGE in the cytoplasm of NSCLC cells was associated with an increased patient mortality after complete surgical resection. This association was particularly prominent in patients with stage I lung cancer (Figure 1). Although complete tumor resection in the patients with stage I disease carries a relatively good prognosis, the 5-yr survival is only 60 to 72%, with death being due to the development of distant metastases (2–4). In this study, the 5-yr survival of patients with stage I cancer with an absence of esRAGE expression was less than 20% (Figure 2B). A multivariate analysis confirmed esRAGE immunoreactivity to be an independent factor for predicting the clinical outcome of patients with NSCLC after surgery (Table 4). Consistent with a previous study, cytoplasmic esRAGE was ubiquitously present in the normal lung samples (31). In contrast, esRAGE expression disappeared or was reduced in most NSCLC tissues. Expression of esRAGE mRNA was also decreased in cancer cells isolated by laser capture microdissection. Full-length RAGE and esRAGE were differentially expressed in the same patients (Figures 1C–1D; Tables 2 and 3), suggesting that esRAGE expression is regulated independently of full-length RAGE expression. Cytoplasmic esRAGE immunoreactivity was inversely correlated with the pathologic T factor, histologic differentiation, and Ki-67, but was not correlated with the pathologic N factor (Table 1). The T factor is determined by a combination of tumor size and invasiveness. Because esRAGE immunoreactivity was not correlated with the tumor size (Table 1), invasiveness is another key factor influenced by esRAGE. These findings suggest that cytoplasmic esRAGE plays a role in local proliferation of carcinoma and invasion. Transfection of full-length RAGE has an antiproliferative effect in cancer cell lines (20), and engagement of RAGE with its ligand amphoterin is the main pathway through which RAGE contributes to cancer growth (19, 20). In our study, however, overexpression of esRAGE in lung cancer cells attenuated the proliferation of cancer cell lines through increasing G0/G1 fraction. This inhibitory effect was much greater than what was produced by full-length RAGE, and occurred in the absence of amphoterin (Figure 4). We inferred from these findings that esRAGE independently controls cell proliferation, and that loss of esRAGE in cytoplasm accelerates the growth of the carcinoma. Bartling and colleagues reported a similar inhibitory effect of full-length RAGE in tumor growth in NSCLCs, and they found that transfection of full-length RAGE demonstrated the decrease of S/G2 fraction in the cancer cell line (25). Because full-length RAGE and esRAGE expression seemed to be regulated independently, the role of full-length RAGE and esRAGE on lung cancer might be different. In addition, down-regulation of full-length RAGE had no impact on patient mortality after complete surgical resection (Figure 3). The effect of full-length RAGE in different cancers is inconsistent. In gastric (16), colorectal (17), and prostate cancer (18), RAGE up-regulation is closely associated with tumor invasion and metastasis (14). In contrast, a decrease in RAGE signaling inhibits tumor growth and metastasis in neuroblastoma, melanoma (20), and lung cancer (25). The reason for the diametrically opposite effects of RAGE in different tumors is still under debate; however, it may in part be related to the expression pattern of esRAGE. esRAGE staining patterns can be considered to fall into one of four categories: pattern A is diffuse cytoplasmic staining, pattern B is supranuclear dotlike staining, pattern C is staining in the stromal area, and pattern D is diffuse staining in the secreted material. Interestingly, pneumocytes and neuronal cells, the progenitor cells for lung cancer and neuroblastoma, respectively, show pattern A staining. In contrast, pattern B staining is demonstrated in the progenitors of gastrointestinal and prostate cancer (31). This suggests that the pattern of esRAGE expression in the progenitor may contribute to the later effects of RAGE on tumor growth. RAGE belongs to the immunoglobulin superfamily (10). Recently, the associations of other immunoglobulin superfamily molecule expression in the cytosol with tumor growth were reported (35–37). For example, the activated leukocyte cell adhesion molecule (ALCAM/CD166) was up-regulated in colorectal cancer and breast cancer, and overexpression of ALCAM was reported as an independent prognostic factor of these carcinomas (35, 37). Our results and these reports suggest that some of the cytosolic immunoglobulin superfamily molecules play a role in tumor proliferation, and have potency as new prognostic factors. In summary, the expression of cytoplasmic esRAGE, a novel splice variant of RAGE, decreased in lung cancer. Loss of esRAGE expression was associated with an increased risk of mortality after complete surgical resection, and was an independent predictor of the clinical outcome in patients with NSCLC. Our findings suggest that cytoplasmic esRAGE could be used in patients with stage I NSCLC to predict the outcome of curative surgery. Given the central role of esRAGE in tumor biology, it has the potential to predict the patients who will benefit from additional interventions such as adjuvant chemotherapy.
The authors thank Dr. Koichi Tsuneyama (Toyama Medical and Pharmaceutical University) for his technical advice, and Dr. Masaru Yanai (Ishinomaki Red Cross Hospital) for his extensive encouragement.
Supported by a grant from the Japanese Society for the Promotion of Science No. 13670589 to H.K. This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Originally Published in Press as DOI: 10.1164/rccm.200602-212OC on October 5, 2006 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form February 13, 2006; accepted in final form October 5, 2006
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
AT A GLANCE COMMENTARY
-peptides, S100/calgranulines, and high-mobility group protein (HMGB)-1/amphoterin (
2 test. Overall and disease-free survival curves were generated according to the Kaplan-Meier method. Univariate and multivariate analyses were performed with a proportional hazard model (Cox) using PROC PHREG in the SAS software (SAS Institute, Inc., Cary, NC). Fisher's exact test was performed using the SAS software. p values less than 0.05 were considered significant. 
p < 0.05 versus full-length RAGE. (E) Flow cytometric analysis showed an increment of G0/G1 fraction rate in esRAGE transfected A549 and LK-1 cells. Columns, mean (n = 3). dn, dominant negative RAGE; full, full-length RAGE; es, esRAGE.
