INTERNATIONAL JOURNAL OF MOLECULAR MEDICINE 38: 259-266, 2016 Roles of ZIC family genes in human gastric cancer GANG MA1, WEIJIE DAI1, AIYU SANG2, XIAOZHONG YANG1 and QIANJUN LI1 1Department of Gastroenterology, Huai'an First People's Hospital, Nanjing Medical University, Huai'an, Jiangsu 223300; 2Department of Internal Medicine, Lianshui Third People's Hospital, Lianshui, Jiangsu 223411, P.R. China Received October 20, 2015; Accepted April 25, 2016 DOI: 10.3892/ijmm.2016.2587 Abstract. The human zinc finger of the cerebellum (ZIC) Introduction family genes, comprised of 5 members, which are vertebrate homologues of the Drosophila odd-paired gene and encode Gastric cancer (GC), one of the most frequent malignancies, zinc-finger transcription factors, have been shown to be remains the second leading cause of cancer-related mortality involved in various diseases, including cancer. However, the worldwide (the third in males and the fifth in females), despite roles of ZICs in human gastric cancer (GC) have not yet been great advances in therapeutic regimens and improved surgical fully elucidated. This study aimed to investigate the expression outcomes (1). Based on Globocan 2012, GC accounts for nearly patterns of ZICs and determine their clinical significance in 952,000 new cases annually in worldwide (2). Of all GC cases, GC. The mRNA and protein expression levels of ZIC1-5 were 90% are malignant and 95% are comprised of gastric adeno- detected by RT-qPCR and western blot analysis, respectively carcinoma (3). More than 70% of new cases and deaths related using 60 pairs of human GC and matched normal mucosa to GC are occurred in developing countries (4). In China in tissues. The expression pattern and subcellular localization of particular, this cancer has the second highest incidence among ZIC1 in 160 pairs of human GC and matched normal mucosa commonly diagnosed cancers (5). Due to a lack of early specific tissues were verified by immunohistochemistry. Moreover, the symptoms, the diagnosis of GC is often delayed, leading to associations of ZIC1 expression with various clinicopatho- cancer cell invasion into the muscularis propria (6). Consistent logical characteristics and patient prognosis were evaluated. with other types of cancer, the development of GC is a multiple- The mRNA and protein expression levels of ZIC1 were both stage process in which the accumulation of molecular changes found to be significantly decreased in the GC tissues compared lead to malignant phenotypes with aggressive characteris- to matched normal mucosa tissues (GC vs. normal, 2.15±0.69 tics (7). Therefore, the detection of the abnormal expression of vs. 4.28±0.95; P<0.001); however, ZIC2-5 expression exhibited molecular markers may be a promising approach for the early no significant difference between the cancer and normal tissue diagnosis and prognosis of patients with GC. samples. In addition, the downregulation of ZIC1 (ZIC1-low) The human zinc finger of the cerebellum (ZIC) family was more frequently observed in the GC tissues with posi- genes, comprised of 5 members (ZIC1, ZIC2, ZIC3, ZIC4 and tive lymph node metastasis (P=0.006), an advanced TNM ZIC5) which are vertebrate homologues of the Drosophila stage (P<0.001) and a great depth of invasion (P=0.01). Notably, odd-paired (OPA) gene, are structurally similar to each other, a low ZIC1 expression was significantly associated with a poor implying that ZICs share some, but not all, functions (8). disease-free and overall survival. Furthermore, multivariate ZIC genes encode zinc-finger transcription factors, each analysis revealed that ZIC1 expression was an independent composed of 5 C2H2 zinc-finger domains, which have highest prognostic marker for patients with GC. In conclusion, among sequence homology to Drosophila OPA (9). Functionally, the human ZIC family genes, the dysregulation of ZIC1, but not ZIC genes play crucial roles in a wide array of developmental of ZIC2, ZIC3, ZIC4 and ZIC5, may play a crucial role in the systems, including the central nervous system (CNS), muscle progression of GC. ZIC1 may thus serve as a novel molecular and skeletal development (10). The ZIC proteins are mainly marker to predict the progression, survival and relapse of expressed in the developing or mature CNS in a spatiotempo- patients with GC. rally restricted manner, but none contain a canonical nuclear localization signal. As zinc finger transcription factors, these proteins can bind to the GC-rich sequence in target genes (11). With the similar zinc finger domains, the ZIC family members Correspondence to: Dr Xiaozhong Yang, Department of have also been shown to interact with the Gli family proteins Gastroenterology, Huai'an First People's Hospital, Nanjing Medical in both an antagonistic and synergistic manner (12). In University, 6 Beijing Road West, Huai'an, Jiangsu 223300, P.R. China recent years, growing evidence has indicated that ZICs may E-mail: [email protected] be involved in the pathological events of various diseases, including cancer. For example, ZIC1 was found to participate Key words: gastric cancer, zinc finger of the cerebellum family in the progression of human medulloblastoma (13), thyroid genes, zinc finger protein of the cerebellum 1, clinicopathological cancer (14), GC (15-17), colorectal cancer (18), endometrial characteristic, disease-free survival, overall survival cancer (19) and mesenchymal neoplasms (20); ZIC2 may function as an oncogene in small cell lung carcinoma (21), 260 MA et al: ZIC FAMILY GENES IN GASTRIC CANCER pancreatic ductal adenocarcinoma (22), epithelial ovarian 62 cases were stage IV. The detailed information on the clini- cancer (23) and cervical cancer (24); genome-wide analysis of copathological characteristics of all 160 patients with GC are CpG island methylation in bladder cancer identified ZIC4 as a shown in Table I. pTa-specific prognostic marker (25). However, the roles of the All the 160 patients with GC were given a follow-up exam ZIC family members in GC have not yet been fully elucidated. ranging from 3 to 6 years. Patients who died from diseases A comparison of their expression levels and clinical signifi- other than GC or from unexpected events were excluded from cance in GC is required. the case collection in this study. For the analysis of survival These observations led us to investigate the expression and follow-up data, the date of surgery was used to represent profiles of ZIC genes and proteins in GC, and to determine the beginning of the follow-up period. Overall survival was an their clinical implications. We first detected the mRNA and endpoint which was calculated as the amount of time between protein expression levels of ZIC1, ZIC2, ZIC3, ZIC4 and ZIC5 the date of surgery and the date of death, regardless of the by reverse transcription-quantitative polymerase chain reac- cause. Disease-free survival was defined as the time from tion (RT-qPCR) and western blot analysis, respectively, using randomization until recurrence of the tumor or death from any 60 pairs of human GC and matched normal mucosa tissues. cause. Surviving patients were censored on March 31, 2013. The expression pattern and subcellular localization of ZIC1 in 160 pairs of human GC and matched normal mucosa tissues RT-qPCR. Total RNA was isolated using TRIzol reagent were then examined by immunohistochemistry. Moreover, the (Invitrogen, Carlsbad, CA, USA). A total of 2 µg RNA was associations of ZIC1 expression with various clinicopatholog- reverse transcribed using the SuperScript II RNase-Reverse ical characteristics and patient prognosis were also evaluated. Transcriptase system (Invitrogen). GAPDH was used as an internal control. cDNA was then subjected to quantitative (real- Materials and methods time) PCR (qPCR) using primers specific for ZIC1, ZIC2, ZIC3, ZIC4, ZIC5 and GAPDH. PCR primers were designed Ethics statement. This study was approved by the Ethics according to the previous study by Aruga et al (26) as follows: Committee of Huai'an First People's Hospital of Nanjing ZIC1 forward, 5'-GGCCCGGAGCAGAGTAAT-3' and reverse, Medical University and Lianshui Third People's Hospital, 5'-AGCCCTCAAACTCGCACTT-3' (229 bp, 26 cycles); ZIC2 Huai'an, China. Written informed consent was also obtained forward, 5'-CCCTTCAAGG CCAAATACAA-3' and reverse, from all study participants for the use of their samples. All 5'-TGCATGTGCTTCTTCCTGTC-3' (218 bp, 26 cycles); specimens were handled anonymously according to the ethical ZIC3 forward, 5'-GCAAGTCTTTCAAGGCGAAG-3' and and legal standards. reverse, 5'-CATGCATGTGCTTCTTACGG-3' (225 bp, 28 cycles); ZIC4 forward, 5'-GCCCTTCAAAGCCAAAT Patients and tissue samples. For RT-qPCR and western blot ACA-3' and reverse, 5'-GCCCTCGAACTCGCATC-3' (172 bp, analysis, a total of 60 fresh GC and matched normal mucosa 28 cycles); ZIC5 forward, 5'-TCTGCTTCTGGGAGGAC specimens were obtained from 60 patients with GC (48 males TGT-3' and reverse, 5'-GGGAATGTTTCTTCCGATCA-3' and 12 females; median age, 58 years; range, 28-82 years), (252 bp, 28 cycles); and GAPDH forward, 5'-GAAGGTGAA who underwent surgical resection, at the Department of GGTCGGAGT-3' and reverse, 5'-GAAGATGGTGATGGG Gastroenterology of Huai'an First People's Hospital from ATTTC-3' (226 bp, 28 cycles). The PCR cycling conditions January 2009 to December 2010. All specimens were stored at were as follows: 94˚C for 4 min, followed by 40 cycles of 95˚C -80˚C until use to detect the relative expression levels of ZIC1, for 1 min, 60˚C for 1 min and 72˚C for 1 min. The SYBR ZIC2, ZIC3, ZIC4 and ZIC5 genes and proteins. Premix Ex Taq™ kit (Takara Bio, Inc., Otsu, Shiga, Japan) was For immnohistochemistry, a total of 160 paraffin-embedded used to measure the amplified DNA, and qPCR was performed GC and matched normal mucosa specimens, in addition to the using an iQ5 real-time PCR detection system (Bio-Rad, 60 cases mentioned above, were obtained from 160 patients Hercules, CA, USA).