Abstract
Background/Aim: To improve patient management, new biomarkers are required that stratify prognosis. Here we focused on glutamic acid decarboxylase 1 (GAD1), which is associated with proliferation of lung cancer cells, and investigated its expression and function in esophageal squamous cell carcinoma (ESCC). Materials and Methods: We evaluated changes in the proliferative potential of ESCC cell lines using small interfering RNA-mediated GAD1 knockdown techniques. We analyzed GAD1 protein expression using a tissue microarray (TMA) and measured GAD1 mRNA expression to evaluate correlations between the expression level of each tissue and postoperative outcomes of two independent cohorts (the TMA and mRNA cohorts) of patients who underwent radical esophagectomy. Results: GAD1 knockdown reduced cell proliferation. In the TMA cohort, high GAD1 expression significantly correlated with lymph node metastasis and advanced stage. Disease-free survival was significantly shorter in the group with high GAD1 expression, as was overall survival. Multivariate analysis of overall survival showed that positivity for GAD1 was an independent prognostic factor for poor survival. In the mRNA cohort, GAD1 mRNA expression in ESCC tissues was significantly up-regulated compared with that in adjacent noncancerous mucosal tissues. When patients were divided into high- and low-expression groups according to the median GAD1 mRNA expression level in ESCC tissues, overall survival was significantly shortened in the high GAD1 expression group. The incidence of initial hematogenous recurrence was significantly higher in the group with high GAD1 expression. Conclusion: GAD1 expression mediates the proliferative potential of ESCC cells, and a high level may serve as a useful prognostic biomarker for patients with ESCC.
The frequent recurrence of esophageal squamous cell carcinoma (ESCC), even after radical resection, contributes to its poor prognosis, as indicated by the dismal survival rates. Serum tumor markers, such as squamous cell carcinoma antigen and carcinoembryonic antigen, are widely used in clinical practice as biomarkers for ESCC (1-3). However, their significance as prognostic factors is low because sensitivity and specificity are insufficient, and accurate risk stratification is impossible using serum levels. Therefore, a new biomarker that can accurately stratify the prognoses of patients with ESCC is required (4, 5). Moreover, we can begin to consider more intensive adjuvant therapy or follow-up plans only when we can identify a group of patients with particularly poor prognosis (6, 7).
Here we focused on the gene encoding glutamic acid decarboxylase 1 (GAD1), which is associated with cancer cell proliferation via the immune response and represents a new candidate biomarker for ESCC (8, 9). Several malignancies express increased levels of GAD1 (10-14). However, the mechanisms that regulate the expression and function of GAD1 in ESCC are unknown. Most studies of cancer biomarkers analyzed data for a single cohort; analysis of multiple cohorts will reduce bias and enhance the reproducibility of the performance of biomarkers. To address these issues, here we analyzed two independent cohorts for the expression of the product of GAD1. Furthermore, we focused on evaluating the relationship between GAD1 expression levels in tissues and clinical data of patients with ESCC to determine whether GAD1 can serve as a potential biomarker to improve risk stratification after curative resection.
Materials and Methods
Ethics. This study, which was approved by the Institutional Review Board of Nagoya University, Japan (approval number 2014-0044) and the Ethics Committee of Akita University School of Medicine (number 1495), conformed to the ethics guidelines of the World Medical Association Declaration of Helsinki Ethical Principles for Medical Research Involving Human Subjects. Written informed consent for the use of clinical samples and data was directly obtained from all patients or through the appropriate database.
Sample collection. The first cohort comprised primary ESCC tissues collected from 177 patients who underwent esophageal resection for esophageal cancer at the Department of Thoracic Surgery, Akita University Hospital between 2000 and 2011 (15). The patients were not administered treatment before curative surgery. The tissue specimens were embedded in paraffin, and a tissue microarray (TMA) was constructed at the Pathology Institute (Toyama, Japan) and stained as described below (16-18). Specimens were histologically classified according to the eighth edition of the Union for International Cancer Control (UICC) classification (19). Relevant clinicopathological parameters were acquired from patients’ medical records.
The second cohort comprised 189 primary ESCC tissues and adjacent normal tissues acquired from patients who underwent radical esophageal resection at Nagoya University Hospital between October 2001 and January 2016 (20). The tumors were determined as radically resected when pathologically diagnosed as stages I-IV. All tissue samples that were histologically diagnosed as ESCC were immediately frozen after resection, stored at −80°C, and histologically classified as described above. Fluorouracil combined with platinum-based neoadjuvant chemotherapy has been recommended for patients with clinical stage II-IV, unless contraindicated, since 2006 (21), therefore these patients had been treated with neoadjuvant chemotherapy before sampling. Postoperative follow-up included physical examination, measurement of serum tumor markers every 3 months, and enhanced computed tomography of the chest and abdominal cavity every 6 months.
Immunohistochemistry (IHC). TMA blocks were sectioned and incubated for 1 h at room temperature (25°C) with a rabbit polyclonal antibody to GAD1 (MBS2519172, MyBioSource, San Diego, CA, USA) diluted 1:100 in ChemMate antibody diluent (Dako, Carpinteria, CA, USA). For semiquantification of GAD1 expression, two investigators who were uninformed of the clinical data evaluated the tissue staining. Protein expression was scored as 0: no staining; no or weak cytoplasmic staining in <5% of cells; 1+ (weak staining; weak or moderate cytoplasmic staining in <40% of cells); or 2+ (strong staining; intense cytoplasmic staining in ≥40% of cells) (15). GAD1 positivity was defined as a score of 1+ or 2+.
Cell lines and culture. The human ESCC cell lines (TE1, TE2, TE3, TT and TTn) were obtained from the American Type Culture Collection (Manassas, VA, USA). NUEC2 cell line was established at Nagoya University (22). KYSE510, KYSE590, KYSE890, KYSE1170, KYSE1260, and KYSE1440 were obtained from the Japanese Collection of Research Bioresources Cell Bank (Osaka, Japan) (23). Cells were stored at 80°C in a cell preservative (Cell Banker; LSI Medience Corporation, Tokyo, Japan) and cultured in RPMI-1640 medium (Sigma–Aldrich; Merck KGaA, Darmstadt, Germany) supplemented with 10% fetal bovine serum (Thermo Fisher Scientific, Inc., Waltham, MA, USA) at 37°C in an atmosphere containing 5% CO2.
Small interfering RNA (siRNA)-mediated knockdown of GAD1. We designed three siRNAs specific for GAD1. KYSE590 cells were cultured in six-well plates (2×105 cells/ml). Cells were transiently transfected the next day with 100 nmol/l GAD1 siRNAs (Table I) or a control siRNA (siControl). A NEON electroporation system (Invitrogen, Waltham, MA, USA) was used to introduce the siRNAs into cells. GAD1-expressing cells were subjected to a pulse voltage of 1600 V, with pulse width of 10 ms (three pulses). Knockdown efficiency was determined using qRT-PCR 24 h after transfection. Transfected cells were cultured in RPMI medium without antibiotics for 72 h and then used for functional assays (24).
Sequences of primers and small interfering RNA (siRNA) used in this study.
Cell proliferation assay. Proliferation of transfected and control cells was evaluated using a Cell Counting Kit-8 (Dojindo Molecular Technologies, Inc., Kumamoto, Japan). Cells (5×103 cells/well) were incubated, and the optical density of the solution in each well was measured on days 0, 1, 3, and 5, after the addition of 10 μl of Cell Counting Kit-8 solution (25-27).
Quantitative real-time reverse transcription polymerase chain reaction (qRT-PCR). GAD1 mRNA levels were evaluated in ESCC and adjacent normal tissue, as well as native and GAD1 siRNA-transfected cells using a previously described qRT-PCR assay (20, 28). Three wells were allocated for controls without template for each reaction. Specific primers are presented in Table I. mRNA expression levels of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (TaqMan; GAPDH control reagents; Applied Biosystems; Thermo Fisher Scientific) were quantified, and the data were used to normalize the expression levels of GAD1.
Statistical analysis. Chi-square test was used to analyze the significance of the association between GAD1 expression level and clinicopathological factors. The values of numeric variables between the two groups were compared using the Mann-Whitney test. Overall (OS) and disease-free (DFS) survival rates were analyzed using the Kaplan-Meier method, and a Cox proportional hazards model was employed to calculate hazard ratios and to perform multivariable regression analysis. JMP pro 16 software (SAS Institute Inc., Cary, NC, USA) was used for statistical analysis, and values of p<0.05 indicate a significant difference.
Results
Expression of GAD1 in ESCC cell lines. To investigate the expression of GAD1 in ESCC, we first examined GAD1 mRNA levels in 13 ESCC cell lines representing differentiated (n=6) and undifferentiated (n=3) phenotypes, as well as cells with unknown differentiation (n=4). The qRT-PCR analysis revealed no significant difference in GAD1 mRNA levels among them (Figure 1A).
mRNA expression of glutamic acid decarboxylase 1 (GAD1) and the effect of its knockdown on proliferation of esophageal squamous cell carcinoma cells. A: GAD1 mRNA levels in 13 esophageal squamous cell carcinoma cell lines. B: Quantitative real-time reverse-transcription polymerase chain reaction analysis of GAD1 expression in non-transfected and GAD1 siRNA-transfected KYSE590 cells. C: Proliferation of KYSE590 cells transfected with GAD1-specific siRNAs. GAPDH: glyceraldehyde 3-phosphate dehydrogenase. *Significantly different at p<0.05.
Effect of knockdown of GAD1 expression on the proliferation of ESCC cells. We first analyzed cell lines with known phenotype expressing high levels of GAD1 mRNA (TTn, KYSE510, KYSE590, KYSE1260, and KYSE1440). Although we conducted transfection of siRNAs under various conditions using Lipofectamine or electroporation, the transfection efficiencies using TTn, KYSE510, KYSE1260, and KYSE1440 cells were insufficient (data not shown). In contrast, we found that siRNA-mediated knockdown of GAD1 mRNA expression was inhibited with acceptable efficiency when we used KYSE590 cells, which expressed relatively high levels of GAD1 mRNA (Figure 1B). We therefore used siRNA-transfected KYSE590 cells in functional analyses and found that their proliferation was significantly inhibited from 72 h to 120 h compared with controls (Figure 1C).
Prognostic value of GAD1 expression using the TMA cohort. To investigate the expression of GAD1 and its relationship to clinicopathological characteristics, we employed IHC to analyze the TMA cohort in tumor and normal tissue samples collected from 177 patients who underwent curative surgery for ESCC. The population comprised 153 males and 24 females (mean age±standard deviation=66±8.2 years, range=38-82 years) with pathologically diagnosed differentiated (n=120) or undifferentiated (n=57) ESCC. According to the UICC classification (eighth edition), 10, 44, 105, and 18 patients were diagnosed with stage I, II, III and IV ESCC, respectively.
IHC data were expressed using a semiquantitative scoring system that evaluated the localization, intensity and extent of GAD1 expression. Representative photomicrographs of specimens with no staining (score 0), weak staining (score 1+), and strong staining (score 2+) are shown in Figure 2A. Among the 177 samples, 62 (35.0%) scored 0, 84 (47.5%) scored 1+, and 31 (17.5%) scored 2+. To evaluate the prognostic value of GAD1 expression, OS and DFS rates were compared using the Kaplan-Meier method. We found that the 5-year OS rates of the GAD1-positive groups were significantly shorter compared with those of the GAD1-negative group as follows: score 0 vs. 1+: hazard ratio (HR)=3.24, 95% confidence interval (95% CI)=1.60-6.55, p=0.0011; and score 0 vs. 2+: HR=8.24, 95% CI=3.77-18.00, p<0.0001 (Figure 2B). A similar significant outcome was obtained for patients with score 2+ when the 5-year DFS rates were compared: HR (0 vs. 2+)=3.41, 95% CI=1.82-6.40; p=0.0001 (Figure 2C).
Analysis of the association of glutamic acid decarboxylase 1 (GAD1) expression in ESCC with patient survival. A: Representative images of immunohistochemical analysis of GAD1 protein in tumor microarray. Examples of no staining (score 0), weak staining (score 1+), and strong staining (score 2+) (magnification: main image, ×100; magnified inset, ×400). B: Kaplan-Meier analysis of overall survival stratified according to GAD1 expression score. C: Kaplan-Meier analysis of disease-free survival stratified according to GAD1 expression score. CI: Confidence interval; HR: hazard ratio.
The analysis of the relationship between GAD1 expression and patients’ clinicopathological characteristics showed there were no significant differences in age, sex, smoking history, alcohol consumption history, tumor location and depth, differentiation, lymphatic invasion, or venous invasion (Table II). However, here were significant differences in lymph node metastasis and stage. The percentage of patients with lymph node metastasis and advanced stage was significantly higher in the group with high GAD1 expression. Univariate analysis of postoperative OS revealed that lymph node metastasis, poor differentiation, lymphatic invasion, stage III-IV), and positive GAD1 expression (score 1+ or 2+) were significant prognostic factors associated with poor prognosis. Furthermore, multivariate analysis of OS revealed that 1+ GAD1 expression (HR=2.69; 95% CI=1.31-5.52; p=0.0069) and GAD1 expression (2+) (HR=4.93, 95% CI=2.22-10.92; p<0.0001) were independent prognostic factors significantly associated with poor survival (Table III).
Association between the expression of glutamic acid decarboxylase 1 (GAD1) protein and clinicopathological parameters of 177 patients with resected esophageal squamous cell carcinoma.
Prognostic factors for overall survival of 177 patients with esophageal squamous cell carcinoma.
Expression and prognostic value of GAD1 mRNA levels of the mRNA cohort. The median age of the 189 patients in the mRNA cohort was 65 years (range=44-84 years), and the female-to-male ratio was 41:148. According to the UICC staging system (eighth edition), 35, 47, 97, and 10 patients were diagnosed with pathological stages I, II, III, and IV, respectively. The mean normalized GAD1 mRNA expression level was significantly higher in ESCC tissues compared with the corresponding adjacent noncancerous mucosal tissues (p=0.0281) (Figure 3A).
Analysis of mRNA expression of glutamic acid decarboxylase 1 (GAD1) in clinical samples from patients undergoing curative esophagectomy. A: GAD1 mRNA expression levels in 189 resected esophageal squamous cell carcinoma (ESCC) tissues and adjacent noncancerous esophageal mucosa are represented using box plots. Each box represents the interquartile range (from the 25th percentile to the 75th), the bar in the middle of the box is the median, and the bars extend to the minimum and maximum. B: Kaplan-Meier analysis of overall survival as a function of high or low expression of GAD1. C: Kaplan-Meier analysis of disease-free survival as a function of high or low expression of GAD1. D: Frequency of initial recurrence after patients undergoing curative esophagectomy according to GAD1 expression. *Significantly different at p<0.05.
When the groups with high and low GAD1 mRNA expression were separately analyzed according to their median GAD1 mRNA expression level, OS was significantly shorter in the GAD1-high expression group (p=0.0190) (Figure 3B). The pattern was similar for DFS of the GAD1-high expression group (p=0.0603) (Figure 3C). The incidence of first hematogenous recurrence was significantly higher in the group with high GAD1 expression compared with low expression (p=0.0270) (Figure 3D). In contrast, there were no significant differences in overall, lymph node and local recurrence by GAD1 expression.
Discussion
To offer optimal personalized treatments for patients with ESCC, it is mandatory to identify new biomarkers that reflect tumor biology and, ultimately, prognosis (4, 5). Here we focused on GAD1, which is associated with cancer cell proliferation via the immune response (8, 9), and investigated its function and expression. We found that knockdown of the expression of GAD1 reduced the proliferative capacity of ESCC cell lines. To gain insight into the mechanism of this effect, we conducted IHC and qRT-PCR to measure GAD1 expression levels in two independent cohorts of patients who underwent radical esophagectomy. These analyses revealed a significant reproducible association between high GAD1 expression and poor prognosis.
GAD1 encodes the 67 kDa isoform of glutamate decarboxylase that catalyzes the synthesis of γ-aminobutyric acid (GABA) from L-glutamic acid that is required for GABA synthesis (29, 30). Dysregulation of GABA and GAD1 expression is associated with neurological disorders, such as schizophrenia, bipolar disorder, Parkinson’s disease, and cerebellar disorders (10). Recent studies show that GAD1 is highly expressed in lung cancer and oral squamous cell carcinoma, but not in EC (11, 12). In patients who undergo resection of lung adenocarcinomas, GAD1 expression serves as a useful prognostic factor. Furthermore, high levels of GAD1 expression are significantly associated with pleural invasion and lymphatic invasion (12). GAD1-knockdown of oral squamous cell carcinoma cell lines suppresses the nuclear translocation of β-catenin, reduces the expression of matrix metallopeptidase 7, and inhibits cell invasiveness and migration (11). Here we show that GAD1 knockdown in ESCC cell lines reduced cell proliferation. These findings are consistent with those of other studies showing that GABA produced by cancer cells inhibits glycogen synthase kinase 1-3β activity by activating GABABR to enhance β-catenin signaling, leading to cancer cell proliferation (8). In ESCC, GABA-mediated activation of β-catenin by high GAD1 expression in cancer cells may lead to their increased proliferation (8, 9).
Biomarkers for esophageal cancer include serum levels of carcinoembryonic antigen and squamous cell carcinoma antigen. Furthermore, elevated levels of serum tumor markers precede imaging diagnosis, and may serve as adjunct diagnostic markers to predict postoperative recurrence (31, 32). However, preoperative values do not accurately stratify prognosis. In the present study, we demonstrate the possibility of establishing a definitive three-tiered risk classification for OS, DFS and postoperative recurrence.
Here we found that analysis of a TMA did not reveal a relationship between cytoplasmic or nuclear GAD1 protein staining and clinicopathological factors. Furthermore, there was not a significant association between GAD1 expression and typical predictors of poor prognosis, such as tumor depth, nor lymphatic or venous invasion. However, there was a significant association between GAD1 expression and lymph node metastasis and stage. Multivariate analysis of overall postoperative survival revealed that high GAD1 expression was an independent poor prognostic factor. In particular, in the present study, a higher GAD1 IHC score correlated with increasingly poorer prognosis, suggesting the possibility of risk stratification of patients with ESCC according to the expression level of GAD1. For example, by administering postoperative chemotherapy and follow-up according to the intensity of GAD1 staining in resection specimens, it may be possible to identify cases with poor prognosis that cannot be classified using the existing TNM classification alone (33, 34). The clinical application of these findings will likely lead to lower recurrence rates, with early detection and treatment of recurrence, which will contribute to improved treatment outcomes.
Moreover, GAD1 expression can be measured using IHC in diagnostic biopsy specimens during endoscopy, which may be applicable to the selection of a preoperative treatment, such as chemotherapy. For example, preoperative chemotherapy may be considered for patients with cStage 0/I with high GAD1 expression; or preoperative chemotherapy may be considered for patients with cStage II/III with low GAD1 expression, particularly for older patients or those with insufficient organ function. Furthermore, our present analysis of the site of first recurrence reveals that hematogenous recurrence was significantly more frequent in the group with high GAD1 expression. Although computed tomography and upper gastrointestinal endoscopy are the main follow-up methods after radical resection of esophageal cancer, it may be possible to add bone scintigraphy and fluorodeoxyglucose positron-emission tomography to detect bone metastasis, and ultrasound imaging for liver metastasis in the group with high GAD1 expression, which shorten the postoperative follow-up interval.
The relationship between high GAD1 expression at the transcriptional and translational levels with poor prognosis after radical ESCC resection was demonstrated here using two independent cohorts. We therefore believe that the significant correlations discovered here are highly plausible because of reduced bias and enhanced reproducibility. Furthermore, the versatility of GAD1 expression as a biomarker represents another advantage because it can be evaluated using IHC on formalin-fixed and paraffin-embedded specimens, which are routinely used in the clinic.
There are several limitations to the present study. Firstly, its retrospective nature of and the long duration of case accumulation may have introduced bias because, over this time, accumulation of evidence may lead to changes in treatment interventions, such as surgical procedures and preoperative or postoperative chemotherapy, depending on the timing of treatment. Secondly, the mechanism of GAD1-mediated ESCC cell proliferation was not identified in detail. Recently reported evidence indicates a relationship between GAD1 expression and immune responses, such as those mediated by CD8+ T-cells (8, 9). Further studies are required to gain a better understanding of the biological pathways associated with GAD1 expression and function. Such studies should include detailed characterization of the nature of the immune response.
Conclusion
We conclude that GAD1 expression contributes to the proliferative potential of ESCC cells, and high levels of its expression in tissues may serve as a useful prognostic biomarker after patients with ESCC undergo curative resection.
Acknowledgements
The Authors thank Edanz (https://jp.edanz.com/ac) for editing a draft of this article.
Footnotes
Conflicts of Interest
The Authors declare no competing financial interests.
Authors’ Contributions
T. Kishida conceived and designed the research, performed the experimental procedure and analysis, and wrote the article. M. Kanda, and Y. Sato collected clinical samples and data. D. Shimizu, Y. Inokawa, N. Hattori, M. Hayashi, C. Tanaka and G. Nakayama proofread the article. M. Kanda and Y. Kodera supervised the study.
- Received May 28, 2023.
- Revision received July 15, 2023.
- Accepted July 24, 2023.
- Copyright © 2023, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.