p28GANK overexpression is associated with chemotherapy resistance and poor prognosis in ovarian cancer
- Authors:
- Published online on: November 13, 2019 https://doi.org/10.3892/ol.2019.11081
- Pages: 505-512
Abstract
Introduction
Ovarian cancer (OC) has the highest mortality rate of all gynecological malignancies, with ~14,240 OC-associated mortalities reported in the USA in 2016 (1). Due to a lack of methods for early detection, almost 70% of patients present with an advanced stage of OC at diagnosis, resulting in a poor prognosis (2). Therefore, the identification of potential markers and improved understanding regarding the molecular mechanisms underlying OC may facilitate the development of early detection approaches and novel therapeutic strategies for patients with OC.
The 26S proteasome non-ATPase regulatory subunit 10 (p28GANK), which is localized on human chromosome Xq22.3, is a small protein of 226 amino acids containing 7 ankyrin repeats. This protein is a subunit of the 26S proteasome that specifically interacts with the S6b ATPase of the 19S regulator (3–5). p28GANK promotes the hyperphosphorylation and degradation of retinoblastoma (Rb), which releases the transcription factor E2F-1 from the Rb repressor complex. Furthermore, p28GANK binds to cyclin-dependent kinase 4 (CDK4) and prevents CDK4 inhibition by p16NK4a, which accelerates cell cycle progression (5). p28GANK binds to the E3 ubiquitin protein ligase MDM2, which enhances the ubiquitination and degradation of cellular tumor antigen p53 (6). In addition, it has also been demonstrated to regulate the nuclear factor-κB (7), signal transducer and activator of transcription 3 (STAT3) (8) and protein kinase B (9) pathways. p28GANK serves an important role in cell metastasis, proliferation and autophagy in cancer (8,10,11). However, to the best of our knowledge, there is limited understanding regarding its expression and clinical significance in OC, which the present study aimed to investigate.
Materials and methods
Patients
OC tissue samples were obtained from who were surgically treated at the Third Affiliated Hospital of Harbin Medical University (Harbin, China) between January 1999 and 2006. The exclusion criteria were as follows: i) Multifocal carcinoma; ii) loss to follow-up [no overall survival (OS) follow-up and/or progression data]; and iii) previous history of cancer. Inclusion criteria were as follows: i) Presence of ovarian serous carcinoma confirmed by pathological examination; ii) complete basic clinical data; iii) absence of any prior treatment for cancer; iv) no serious complications or any other malignant disease; v) informed consent was provided by the patients and family members prior to treatment; and vi) complete cytoreductive surgery with no peri- or postoperative (within 45 days after surgery) mortality. The cohort for analysis that met the inclusion criteria consisted of 114 patients. The mean age of the patients was 49 years (range, 28–76 years). Additionally, normal ovarian tissue samples were acquired by endoscopy from non-tumor areas from 30 of the patients with OC enrolled in the study. All tissue specimens used in the current study were obtained with written informed consent from the patients. The study protocol was approved by the Ethics Committee of Harbin Medical University.
Immunohistochemistry (IHC)
IHC was performed following standard procedures as described previously (12). All samples were confirmed to contain >80% tumor cells by three pathologists. Formalin-fixed and paraffin-embedded tissue sections (4-µm thick) were dried out at 65°C for 2 h. Samples were deparaffinized in xylene, and hydrated using a graded alcohol series (70%, overnight; 80, 90 and 95% for 1 h each; and two separate 10 min treatments in 100%. Subsequently, the tissue blocks were placed in xylene for 10 min until the specimen were transparent, after which the tissue blocks were placed in xylene + paraffin and at 60°C for 2 h, to ensure the paraffin completely penetrated the tissue. The dissolved paraffin was poured into a metal frame, and the waxed tissue was placed in the center of the metal frame and left to cool. The tissue was sectioned into 5 µM thick slices and placed in a dish with warm water. The unfolded tissue slices were placed on slides and dried at 60°C 4 h. The sections were then treated with 3% H2O2 at room temperature for 10 min and antigen retrieval was performed in citrate buffer (pH 6.0) at 98°C for 10 min. The tissue sections were incubated with anti-p28GANK mouse polyclonal antibodies (cat. no. ab238999; 1:200 dilution; Abcam, Cambridge, UK) at 4°C overnight. Following washing in PBS, the samples were incubated with biotinylated secondary antibody (cat. no. 14709; 1:1,500 dilution; Cell Signaling Technology, Inc., Danvers, MA, USA) at room temperature for 30 min and developed with diaminobenzadine at room temperature for 5 min. As a negative staining control, the primary antibody was replaced with PBS.
Each section was incubated in a 1:50 dilution of anti-p28GANK antibody. An optical microscope (magnification, ×400) was used for assessment of IHC. The IHC reaction was quantified by multiplying the staining intensity by the percentage of positive tumor cells. In the cytoplasm, staining intensity was graded as 0 (no staining), 1 (weak staining), 2 (moderate staining) or 3 (strong staining). The percentage of the extent of reactivity was scored as follows: 0 (no positive tumor cells), 1 (<10%), 2 (10–50%) and 3 (>50%). Each case was scored independently and in a blinded manner by two investigators. Final scores ≤4 were regarded as low expression and the remainder were classified as high expression.
Follow up
All patients were followed until February 2015, unless they succumbed prior to this date. The OS time was defined as the interval from surgery to mortality or the date of the most recent follow-up. Progression-free survival (PFS) time was calculated from the date of surgery to the date of relapse, including proven local recurrence or distant metastasis, or the date of the most recent follow-up.
PrognoScan database analysis
The association between p28GANK expression and survival in OC was analyzed using the PrognoScan database (13) PrognoScan searches for associations between gene expression and patient prognosis, including OS and disease-free survival (DFS), across a large collection of publicly available cancer microarray datasets, were performed. The threshold was adjusted to a Cox P-value <0.05.
GEPIA database analysis
Gene Expression Profiling Interactive Analysis (GEPIA) (14) is a newly developed interactive web server for analyzing RNA sequencing expression data from The Cancer Genome Atlas (TCGA) database (15). The expression and survival analysis of p28GANK was performed using GEPIA online software version 1.0. OC and matched normal TCGA data and Genotype-Tissue Expression (GTEx) data were used for the expression analysis (16), and log2 (TPM+1) was used for the log-scale.
Statistical analysis
Statistical analysis was performed using SPSS (version 18.0; SPSS Inc., Chicago, IL, USA). A Mann Whitney-U test was used to compare two independent non-parametric samples. χ2 tests were used to assess associations between p28GANK expression level and clinicopathological features. Kaplan-Meier survival curves and log-rank tests were used for survival analysis. Prognostic factors were evaluated by univariate and multivariate analyses using Cox proportional hazard regression models. P<0.05 was considered to indicate a statistically significant difference. Kaplan-Meier survival analysis was used on the Gene Expression Omnibus dataset (GSE26712) using PrognoScan (12). Histological grading was based on the World Health Organization Histological grading system (17).
Results
p28GANK expression in OC and normal tissues
The expression level of p28GANK was higher in the cytoplasm of cancer cells compared with normal tissues (Fig. 1). In the 30 normal controls, only 10 samples (33.3%) exhibited a positive p28GANK cytoplasmic staining. By contrast, of the 114 OC specimens, positive cytoplasmic expression of p28GANK was identified in 72 samples (63.2%; P<0.001; Table I).
p28GANK overexpression is associated with clinicopathological features
The association between the protein expression of p28GANK and clinicopathological variables of OC are presented in Table II. p28GANK expression was significantly associated with the International Federation of Gynecology and Obstetrics (FIGO) stage (18) (P=0.042), residual tumor size (P=0.005) and response to chemotherapy (P<0.001). No significant associations were identified between p28GANK expression and the other clinicopathological factors.
 | Table II.Association between p28GANK expression and clinicopathological factors and of samples from patients with ovarian cancer (n=114). |
Association between p28GANK expression and survival time of patients with OC
The 5-year survival curves stratified by p28GANK expression are presented in Fig. 2. p28GANK expression was revealed to be significantly associated with OS (P<0.001) and DFS (P<0.001). A high expression level of p28GANK was associated with worse OS and DFS times for patients with OC. In addition, a FIGO-stratified analysis of all patients was performed according to the level of p28GANK expression. The results revealed that high expression of p28GANK was associated with DFS and OS in patients with FIGO III and IV OC (both P<0.001).
The Kaplan-Meier survival analysis demonstrated that patients with high expression of p28GANK had significantly shorter DFS (P=0.038) and OS (P=0.037) times using the Gene Expression Omnibus dataset (GSE26712) from PrognoScan (12) (Fig. 3). It was identified that p28GANK had no significant impact on and DFS (P=0.18) and OS time (P=0.31; Fig. 4).
Univariate and multivariate analyses of prognostic variables in patients with OC
Using univariate analysis, OS and DFS times were revealed to be significantly associated with high p28GANK expression (both P<0.001), omental metastasis (P=0.011 and P=0.001, respectively), FIGO stage (P=0.007 and P=0.005, respectively), residual tumor size (both P<0.001) and response to chemotherapy (both P<0.001) (Table III).
The multivariate analysis revealed that high expression of p28GANK (P=0.013), residual tumor size (P<0.010) and response to chemotherapy (P=0.015) were independent prognostic factors for OS in patients with OC (Table IV). Furthermore, it indicated that high expression of p28GANK (P=0.001), histological grade (P=0.010), residual tumor size (P=0.009) and response to chemotherapy (P<0.001) were independent prognostic factors for DFS time in these patients (Table IV).
Discussion
The present study first analyzed the expression levels of p28GANK by IHC, revealing that it was higher in OC tissues compared with normal tissues. This finding is consistent with the results of previous studies (19).
A high p28GANK expression level has also been reported in several other types of cancer, including lung cancer, pancreatic cancer, colorectal cancer, esophageal squamous cell carcinoma, mammary carcinoma, endometrial carcinoma and hepatocellular carcinoma (10,12,20–24)
To further characterize the clinical prognostic implications of p28GANK, the present study analyzed the associations between p28GANK expression and the clinicopathological features and prognosis of patients with OC. The IHC results indicated significant associations between the expression of p28GANK and FIGO stage, residual disease and response to cisplatin-based chemotherapy. In addition, a significant association was revealed between the expression of p28GANK and poor prognosis for OC.
Chen et al (19) identified that a high expression level of p28GANK was positively associated with clinical stage and serum cancer antigen 125 levels, and negatively associated with tumor grade. Furthermore, high levels of p28GANK expression were associated with a poor prognosis and early relapse. In addition, a high expression level of p28GANK has been demonstrated to be associated with primary tumor, lymph node metastasis, clinical stage and poor prognosis in esophageal squamous cell carcinoma (22). In breast cancer, p28GANK overexpression has been associated with lymph node metastasis. Knockdown of p28GANK has been reported to inhibit tumor metastases to the lungs in animal models (23). Furthermore, p28GANK expression was revealed to be significantly higher in cases of hepatocellular carcinoma with increased tumor size and distant metastases (8). A similar observation has also been reported for colorectal cancer (21). In summary, these findings indicate a significant role of p28GANK in tumor metastasis and progression.
Higashitsuji et al (4) first identified p28GANK as an oncogenic protein that is overexpressed in hepatocellular carcinoma. This protein controls the phosphorylation of Rb by CDK4 and promotes the ubiquitination of p53 by MDM2 (3,6). Man et al (10) revealed that p28GANK serves an essential role in Ras-initiated tumorigenesis in human lung cancer. It may decrease cancer cell focal adhesions by regulating the activity of Ras-related C3 botulinum toxin substrate 1, resulting in metastasis (23). p28GANK promotes tumor growth and metastasis in hepatocarcinogenesis via the phosphoinositide 3-kinase (9), STAT3 (8) and β-catenin (11) signaling pathways. Further investigation has confirmed that overexpression of p28GANK enhances the epithelial-mesenchymal transition, which is defined as the switching of polarized epithelial cells to a migratory fibroblastoid phenotype, strengthening matrix metalloproteinase 2 activity, and increasing vascular endothelial growth factor expression (8,9). Therefore, p28GANK may promote cancer metastasis in numerous ways.
Patients with OC have been treated with carboplatin since 1989 (25). To the best of our knowledge, the present study is the first to report that p28GANK overexpression was associated with the response to platinum-based chemotherapy and the OS time of patients with OC. Accumulating data suggest that cancer stem cells (CSCs) exhibit a higher capacity for self-renewal and chemoresistance. Sun et al (26) identified that p28GANK mediates the dedifferentiation of hepatocytes via a hepatocyte nuclear factor 4 α-dependent mechanism. The decrease in p28GANK levels leads to a significant decrease in the proportion of CSCs and the degradation of octamer-binding transcription factor 4 in hepatoma cells (27). p28GANK has also been reported to be significantly associated with prominin-1 and Nanog in colorectal cancer (28). These data suggest that p28GANK may stimulate cancer cell stemness, resulting in resistance to cisplatin chemotherapy.
A previous study suggested that p28GANK inhibits apoptosis in hepatocellular carcinoma cells by enhancing the adaptive response and endoplasmic reticulum chaperone BiP expression under endoplasmic reticulum stress conditions (29). Furthermore, it increases resistance to apoptosis and enhances autophagy with sorafenib treatment (30). In summary, these findings indicate that p28GANK may be a therapeutic target for OC and a p28GANK inhibitor is likely to enhance the effects of cisplatin chemotherapy.
Chen et al (19) identified that p28GANK can promote OC cell proliferation. Consistent with this study, the present findings revealed that high expression of p28GANK was associated with FIGO stage and drug resistance. Therefore, p28GANK may promote tumor progression by enhancing resistance to treatment and may be a valuable therapeutic target.
In conclusion, the current study demonstrated high expression levels of p28GANK in OC. This expression was associated with FIGO stage, residual tumor size, response to chemotherapy, and poor OS and DFS. The present results highlight the importance of p28GANK in the progression of OC, and suggest that it may be a potential prognostic marker and therapeutic target for the improvement of OC clinical management.
Acknowledgements
Not applicable.
Funding
The present study was part of the Program on the Clinical Study of Chinese Medicine funded by the National Natural Science Foundation of China (grant no. 81373673). The present study was also supported by Heilongjiang Health and Family Planning Commission Foundation (grant no. 2016-085), Heilongjiang Postdoctoral Foundation, China (grant no. LBH-Z16246) and Innovation Scientific Research Fund of Harbin Medical University (grant no. 2017LCZX81).
Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Authors' contributions
XF supervised the entire study and participated in study design and coordination. GY, NL and WW performed the experiments and statistical analysis, and drafted the manuscript. MN performed the statistical analysis and article revision. All authors read and approved the final manuscript.
Ethics approval and consent to participate
The present study was approved by the Research Ethics Committee of Harbin Medical University Cancer Hospital (Harbin, China). Written informed consent was obtained from all patients involved in the study.
Patient consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
References
|
DeSantis CE, Siegel RL, Sauer AG, Miller KD, Fedewa SA, Alcaraz KI and Jemal A: Cancer statistics for African Americans, 2016: Progress and opportunities in reducing racial disparities. CA Cancer J Clin. 66:290–308. 2016. View Article : Google Scholar : PubMed/NCBI | |
|
Tørring ML, Falborg AZ, Jensen H, Neal RD, Weller D, Reguilon I; ICBP Working Group, ; Menon U and Vedsted P: Advanced-stage cancer and time to diagnosis: An international cancer benchmarking partnership (ICBP) cross-sectional study. Eur J Cancer Care (Engl). May 22–2019.(Epub ahead of print). View Article : Google Scholar | |
|
Dawson S, Apcher S, Mee M, Higashitsuji H, Baker R, Uhle S, Dubiel W, Fujita J and Mayer RJ: Gankyrin is an ankyrin-repeat oncoprotein that interacts with CDK4 kinase and the S6 ATPase of the 26 S proteasome. J Biol Chem. 277:10893–10902. 2002. View Article : Google Scholar : PubMed/NCBI | |
|
Higashitsuji H, Itoh K, Nagao T, Dawson S, Nonoguchi K, Kido T, Mayer RJ, Arii S and Fujita J: Reduced stability of retinoblastoma protein by gankyrin, an oncogenicankyrin-repeat protein overexpressed in hepatomas. Nat Med. 6:96–99. 2000. View Article : Google Scholar : PubMed/NCBI | |
|
Krzywda S, Brzozowski AM, Higashitsuji H, Fujita J, Welchman R, Dawson S, Mayer RJ and Wilkinson AJ: The crystal structure of gankyrin, an oncoprotein found in complexes with cyclin-dependent kinase 4, a 19 S proteasomal ATPase regulator, and the tumor suppressors Rb and p53. J Biol Chem. 279:1541–1545. 2004. View Article : Google Scholar : PubMed/NCBI | |
|
Higashitsuji H, Higashitsuji H, Itoh K, Sakurai T, Nagao T, Sumitomo Y, Masuda T, Dawson S, Shimada Y, Mayer RJ and Fujita J: The oncoprotein gankyrin binds to MDM2/HDM2, enhancing ubiquitylation and degradation of p53. Cancer Cell. 8:75–87. 2005. View Article : Google Scholar : PubMed/NCBI | |
|
Chen Y, Li HH, Fu J, Wang XF, Ren YB, Dong LW, Tang SH, Liu SQ, Wu MC and Wang HY: Oncoprotein p28 GANK binds to RelA and retains NF-kappaB in the cytoplasm through nuclear export. Cell Res. 17:1020–1029. 2007. View Article : Google Scholar : PubMed/NCBI | |
|
Zheng T, Hong X, Wang J, Pei T, Liang Y, Yin D, Song R, Song X, Lu Z, Qi S, et al: Gankyrin promotes tumor growth and metastasis through activation of IL-6/STAT3 signaling in human cholangiocarcinoma. Hepatology. 59:935–946. 2014. View Article : Google Scholar : PubMed/NCBI | |
|
Fu J, Chen Y, Cao J, Luo T, Qian YW, Yang W, Ren YB, Su B, Cao GW, Yang Y, et al: p28GANK overexpression accelerates hepatocellular carcinoma invasiveness and metastasis via phosphoinositol 3-kinase/AKT/hypoxia-inducible factor-1α pathways. Hepatology. 53:181–192. 2011. View Article : Google Scholar : PubMed/NCBI | |
|
Man JH, Liang B, Gu YX, Zhou T, Li AL, Li T, Jin BF, Bai B, Zhang HY, Zhang WN, et al: Gankyrin plays an essential role in Ras-induced tumorigenesis through regulation of the RhoA/ROCK pathway in mammalian cells. J Clin Invest. 120:2829–2841. 2010. View Article : Google Scholar : PubMed/NCBI | |
|
Dong LW, Yang GZ, Pan YF, Chen Y, Tan YX, Dai RY, Ren YB, Fu J and Wang HY: The oncoprotein p28GANK establishes a positive feedback loop in β-catenin signaling. Cell Res. 21:1248–1261. 2011. View Article : Google Scholar : PubMed/NCBI | |
|
Zhang J, Yang Y, Zhang Z, He Y, Liu Z, Yu Y, Wu S, Cai B and Feng Y: Gankyrin plays an essential role in estrogen-driven and GPR30-mediated endometrial carcinoma cell proliferation via the PTEN/PI3K/AKT signaling pathway. Cancer Lett. 339:279–287. 2013. View Article : Google Scholar : PubMed/NCBI | |
|
Mizuno H, Kitada K, Nakai K and Sarai A: PrognoScan: A new database for meta-analysis of the prognostic value of genes. BMC Med Genomics. 2:182009. View Article : Google Scholar : PubMed/NCBI | |
|
Tang Z, Li C, Kang B, Gao G, Li C and Zhang Z: GEPIA: A web server for cancer and normal gene expression profiling and interactive analyses. Nucleic Acids Res. 45:W98–W102. 2017. View Article : Google Scholar : PubMed/NCBI | |
|
Guo Q, He Y, Sun L, Kong C, Cheng Y and Zhang G: In silico detection of potential prognostic circRNAs through a re-annotation strategy in ovarian cancer. Oncol Lett. 17:3677–3686. 2019.PubMed/NCBI | |
|
Keen JC and Moore HM: The genotype-tissue expression (GTEx) project: Linking clinical data with molecular analysis to advance personalized medicine. J Pers Med. 5:22–29. 2015. View Article : Google Scholar : PubMed/NCBI | |
|
Silverberg SG: Histopathologic grading of ovarian carcinoma: A review and proposal. Int J Gynecol Pathol. 19:7–15. 2000. View Article : Google Scholar : PubMed/NCBI | |
|
Pecorelli S, Benedet JL, Creasman WT and Shepherd JH: FIGO staging of gynecologic cancer. 1994–1997 FIGO committee on gynecologic oncology. international federation of gynecology and obstetrics. Int J Gynaecol Obstet. 65:243–249. 1999. View Article : Google Scholar : PubMed/NCBI | |
|
Chen J, Bai M, Ning C, Xie B, Zhang J, Liao H, Xiong J, Tao X, Yan D, Xi X, et al: Gankyrin facilitates follicle-stimulating hormone-driven ovarian cancer cell proliferation through the PI3K/AKT/HIF-1 α/cyclin D1 pathway. Oncogene. 35:2506–2517. 2015. View Article : Google Scholar : PubMed/NCBI | |
|
Meng Y, He L, Guo X, Tang S, Zhao X, Du R, Jin J, Bi Q, Li H, Nie Y, et al: Gankyrin promotes the proliferation of human pancreatic cancer. Cancer Lett. 297:9–17. 2010. View Article : Google Scholar : PubMed/NCBI | |
|
Bai Z, Tai Y, Li W, Zhen C, Gu W, Jian Z, Wang Q, Lin JE, Zhao Q, Gong W, et al: Gankyrin activates IL-8 to promote hepatic metastasis of colorectal cancer. Cancer Res. 73:4548–4558. 2013. View Article : Google Scholar : PubMed/NCBI | |
|
Ortiz CM, Ito T, Tanaka E, Tsunoda S, Nagayama S, Sakai Y, Higashitsuji H, Fujita J and Shimada Y: Gankyrin oncoprotein overexpression as a critical factor for tumor growth in human esophageal squamous cell carcinoma and its clinical significance. Int J Cancer. 122:325–332. 2010. View Article : Google Scholar | |
|
Zhen C, Chen L, Zhao Q, Liang B, Gu YX, Bai ZF, Wang K, Xu X, Han QY, Fang DF, et al: Gankyrin promotes breast cancer cell metastasis by regulating Rac1 activity. Oncogene. 32:3452–3460. 2013. View Article : Google Scholar : PubMed/NCBI | |
|
Qin X, Wang X, Liu F, Morris LE, Wang X, Jiang B and Zhang Y: Gankyrin activates mTORC1 signaling by accelerating TSC2 degradation in colorectal cancer. Cancer Lett. 376:83–94. 2016. View Article : Google Scholar : PubMed/NCBI | |
|
Lloyd K: The resurgence of platinum-based cancer chemotherapy. Nat Rev Cancer. 7:573–584. 2007. View Article : Google Scholar : PubMed/NCBI | |
|
Sun W, Ding J, Wu K, Ning BF, Wen W, Sun HY, Han T, Huang L, Dong LW, Yang W, et al: Gankyrin-mediated dedifferentiation facilitates the tumorigenicity of rat hepatocytes and hepatoma cells. Hepatology. 54:1259–1272. 2011. View Article : Google Scholar : PubMed/NCBI | |
|
Qian YW, Chen Y, Yang W, Fu J, Cao J, Ren YB, Zhu JJ, Su B, Luo T, Zhao XF, et al: p28 GANK prevents degradation of oct4 and promotes expansion of tumor-initiating cells in hepatocarcinogenesis. Gastroenterology. 142:1547–1558. 2012. View Article : Google Scholar : PubMed/NCBI | |
|
Mine H, Sakurai T, Kashida H, Matsui S, Nishida N, Nagai T, Hagiwara S, Watanabe T and Kudo M: Association of Gankyrin and Stemness factor expression in human; Colorectal cancer. Dig Dis Sci. 58:2337–2344. 2013. View Article : Google Scholar : PubMed/NCBI | |
|
Dai RY, Chen Y, Fu J, Dong LW, Ren YB, Yang GZ, Qian YW, Cao J, Tang SH, Yang SL and Wang HY: p28GANK inhibits endoplasmic reticulum stress-induced cell death via enhancement of the endoplasmic reticulum adaptive capacity. Cell Res. 19:1243–1257. 2009. View Article : Google Scholar : PubMed/NCBI | |
|
Luo T, Fu J, Xu A, Su B, Ren Y, Li N, Zhu J, Zhao X, Dai R, Cao J, et al: PSMD10/gankyrin induces autophagy to promote tumor progression through cytoplasmic interaction with ATG7 and nuclear transactivation of ATG7 expression. Autophagy. 12:1355–1371. 2016. View Article : Google Scholar : PubMed/NCBI |
