Evaluation of the expression and clinical value of lncRNA AC010761.9 in human gastric adenocarcinoma

One hundred forty-five fresh GA tissues were collected at Fuzhou General Hospital in Fujian, China, between March 2014 and October 2015. This group of patients included 145 patients, all of whom had a definite diagnosis of primary GA. The tissue samples, including GA and the matched paracancerous tissues (5 cm from the lesion), were obtained immediately after the surgical procedure, which was performed to remove the primary GA, kept in RNAlater (Qiagen, Duesseldorf, Germany), and stored at − 80 °C before use. The histopathologic diagnosis, the staging tumor-node-metastasis (TNM), and the determination of the histologic grade were performed by two pathologists working in the Department of Pathology of Fuzhou General Hospital (Fujian, China) according to the International Union Against Cancer (5th edition) and the guidelines of the National Comprehensive Cancer Network (NCCN) Clinical Practice of Oncology (V.1.2011) [13]. No pre-operative chemotherapy or radiotherapy was administered. The clinical and pathologic data of this group of patients were also collected and were shown in Table 1. (Except patients in the subgroup related with tumor location only had 125 cases with detailed clinical information, other subgroups had 129 patients with detailed clinical information).

The immunohistochemical (IHC) assay used paraffin-embedded tissue samples, which were made using the same fresh tissue samples described above and performed by two pathologists.

The serum samples from the same patients were collected by laboratory technicians prior to surgery and used for measurement of the levels of digestive tract malignant tumor markers (AFP, CEA, and CA19-9).

The results from the Human LncRNA Expression Profile Microarray V3.0 (Arraystar, Rockville, MD, USA) were obtained from the previous work and completed by Kang Chen Bio-tech (Shanghai, China) based on the manufacturer’s instructions [13]. Microarray assay and data analysis were performed, from which the differentially expressed lncRNAs and mRNAs from six GA tissues and the matched paracancerous tissues were identified [13], by KangChen Bio-tech [13].

The three GA cell lines used in the current study were BGC-823, SGC-7901, and MGC-803. BGC-823 and MGC-803 were obtained from the Shanghai Institute of Biochemistry of the Chinese Academy of Sciences in Shanghai, China, and SGC-7901 was obtained from the American Type Culture Collection (Manassas, VA, USA). The normal cell line used in this work was the GES-1 cell line, which was a normal human gastric mucosal epithelial cell line, and was purchased from the Beijing Cancer Institute (Beijing, China). The cells were cultured in RPMI-1640 medium (for BGC-823, SGC-7901, and MGC-803 cells) and Dulbecco’s modified Eagle’s medium (for GES-1 cells; Invitrogen, Grand Island, NY, USA) based on our previously reported methods [13].

Both the total RNA from the tissues and cultured cells were extracted with the TransZol Up Plus RNA Kit (Transgen Biotech Company, Beijing, China) according to the instructions provided by the manufacturer. Reverse transcription of mRNA into cDNA was then performed with a GoScript™ Reverse Transcription System (Promega, Madison, WI, USA) according to the instructions provided by the manufacturer. Amplification of the products with the SYBR Green Mix kit (GoTaq® qPCR Master Mix; Promega) was carried out in a 2720 Thermal Cycler PCR System (ABI, Grand Island, NY, USA). The primers used in the amplification included that for internal reference 18s RNA, lncRNA AC010761.9, and TRAF4 mRNA, which were as follows: forward 5′-GTAACCCGTTGAACCCCATT-3′ and reverse 5′-CCATCCAATCGGTAGTAGCG-3′ (for 18s RNA); forward 5′-GAGGGAACACTTCTTGCGGG-3′ and reverse 5′-CAGGGCCAGTGTCAACCAAA-3′ (for lncRNA AC010761.9); and forward 5′-CCACCGTTTCTGCGATACCT-3′ and reverse 5′-GGGTCTGGGTAGATCTTGGC-3′ for TRAF4 mRNA. The PCR amplification conditions contained the following three steps: step 1, 95.0 °C for 2 min; step 2, 95.0 °C for 15 s and 58 °C for 40 s (total 40 cycles); and step 3 (dissociation stage), 95 °C for 15 s, 60 °C for 15 s, and 95 °C for 15 s. The cycle threshold (Ct) values of 18s RNA (control), lncRNA AC010761.9, and TRAF4 mRNA were automatically recorded by the machine. The level of lncRNA AC010761.9 and TRAF4 mRNA expression was obtained by calculating the ΔCt, which were inversely proportional to the level of expression. Samples with a higher ΔCt meant lower AC010761.9 or TRAF4 expression. The relative fold changes of AC010761.9 or TRAF4 expression in GA tissues versus matched non-GA tissues or in GA cell lines versus normal gastric cell lines (GES-1) were obtained by calculating the formula (2−^^Ct). The results were obtained from three independent experiments and expressed as the means ± SD.

Quantitative examinations of serum carcinoembryonic antigen (CEA), carbohydrate antigen 19–9 (CA19-9), and alpha-fetoprotein (AFP) concentrations in 129 patients with GA were part of the routine work-up for GA patients, which were performed by technicians before surgery with the Quantitative Kit for Tumor Marker (Protein Chip-Chemiluminescence; HealthDigit, Huzhou, China) with the HD-2001A ChipReader System (HealthDigit) [13]. Individuals with < 5.0 ng/ml for CEA, < 35.0 ng/ml for CA19-9, and < 20.0 ng/ml for AFP were regarded as having low (normal) expression of the three digestive tract tumor markers.

There were ten immunohistochemical markers to be routinely performed for patients with GA in this hospital (vascular endothelial growth factor [VEGF], human epidermal growth factor receptor 2 [C-erbB-2 or HER2], thymidylate synthase [TS], breast cancer 1 [BRCA1], excision repair cross-complementation group 1 [ERCC1], Ki67 antigen [Ki67], ribonucleotide reductase subunit M1 [RRM1], synaptophysin [Syn], neuronal cell adhesion molecule 1 [CD56], and chromogranin A [CgA]). The data were collected from the medical records. The methods, reagents used for IHC assays, and the result determination were the same as previously described [13]. The performances of the IHC assay and result analysis were conducted by pathologists working in this hospital.

SPSS 17 was used for statistical analysis. Data are expressed in the form of mean ± standard deviation. The comparison between the groups was performed using a t test and one-way ANOVA analysis. The relationship between the levels of lncRNA AC010761.9 and TRAF4 mRNA expression was analyzed using Pearson correlation analysis. Multivariate logistic regression analysis was further used to assess the relationships between lncRNA AC010761.9 expression and patient clinical pathologic factors, and factors used in the multivariate analysis included age, gender, tumor location, diameter, differentiation grade, invasion, lymphatic metastasis, venous invasion, and perineural invasion. Multivariate models did not adjust for that factor which was being tested. A p < 0.05 was taken as statistically significant.

The lncRNA microarray chip assays were performed in six GA tissues and six matched paracancerous tissues. The results showed that the level of lncRNA AC010761.9 expression was higher in six GA tissues than the matched non-GA tissues (mean increased fold was 2.01, p < 0.05) (Fig. 1). LncRNA AC010761.9 may be a dysregulated lncRNA in GA, thus it was considered to warrant further investigation.

As depicted in Fig. 2, the results of qRT-PCR further showed that the level of lncRNA AC010761.9 expression was higher in 99 of 145 GA tissues than the matched non-GA tissues, with an over-expressed rate of 68.3% (p < 0.01). When the mean level of expression of this lncRNA in the matched non-GA tissues was taken as 1, an elevated fold up to 35.14 in GA tissues was observed. The results of qRT-PCR also showed that the level of lncRNA AC010761.9 expression was higher in the three GA cell lines (BGC-823, SGC-7901, and MGC-803) compared to normal gastric cells (GES-1 cells; p < 0.05) (Fig. 3). All of these findings indicated that lncRNA AC010761.9 is over-expressed in GA tissues and cell lines.

Among the 145 patients investigated in the present study, all were diagnosed with GA, and 129 patients had detailed clinical pathologic data. The clinical value of the expression of lncRNA AC010761.9 was then evaluated according to the level of expression and clinical pathologic factors. As depicted in Table 1, the elevated expression of this lncRNA was related with a tumor size ≥ 4 versus < 4 cm (p = 0.028) and degree of differentiation (p = 0.047), but not related with other clinical or pathologic factors, including age, gender, tumor location, lymphatic metastasis, and venous and perineural invasion (Table 1). Data from multivariate analysis further exhibited that lncRNA AC010761.9 expression was related with patient age (p = 0.043) and the degree of tumor differentiation (p = 0.015; Table 2). These results indicated that lncRNA AC010761.9 may be GA-associated lncRNA.

As depicted in Table 1 and Additional file 1: Table S1, the relationships between the level of lncRNA AC010761.9 expression and serum CEA, CA19-9, and AFP concentrations and the ten IHC markers showed that lncRNA AC010761.9 expression was related with the serum carbohydrate antigen (CA19-9) and carcinoembryonic antigen (CEA) concentrations (p = 0.026 and p = 0.037, respectively), but not related to serum AFP concentrations and ten IHC markers (Table 1 and Additional file 1: Table S1).

TRAF4 has been observed to participate in several human cancer developments [11, 12]. The TRAF4 gene is speculated to be the potential sense strand gene of lncRNA AC010761.9, thus the level of its expression in 78 GA tissues and matched non-GA tissues was also investigated and related to the level of lncRNA AC010761.9 expression in the same tissues. As depicted in Fig. 4, the expression of TRAF4 mRNA in 54 of 78 GA tissues was higher than the matched non-GA tissues with a mean over-expressed rate of 69.2% (p < 0.05) and an elevated fold up to 10.04. The results from Pearson correlation analyses showed that the level of lncRNA AC010761.9 expression was to some degree a positive correlation to the expression of the potentially associated gene (TRAF4) mRNA in GA tissues (r = 0.385, p < 0.01)(Fig. 5). Additionally, the expression levels of lncRNA AC010761.9 and TRAF4 mRNA among the three GA cell lines (BGC-823, SGC-7901, and MGC-803) also exhibited similarly over-expressed trend compared with those in GES-1 cells.