Using tRNA halves as novel biomarkers for the diagnosis of gastric cancer

2.1 Specimens and clinical data collection

Gastric cancer tissues and paired adjacent non-tumorous tissues were obtained from the surgical patients at Yinzhou People’s Hospital Affiliated to Medical School of Ningbo University, China, from September 2012 to April 2016. Healthy peoples’ plasma and plasma from patients with gastric cancer were also obtained on the day before surgery and seven days after surgery, at Yinzhou People’s Hospital Affiliated to Medical School of Ningbo University and Ningbo No. 1 Hospital Affiliated to Medical School of Ningbo University, China. The diagnosis was confirmed by histopathology and the tumor staging was based upon the International Union Against Cancer’s Tumor-Node-Metastasis (TNM) staging system [15]. The histological grades were assessed according to the National Comprehensive Cancer Networks (NCCNs) guidelines for clinical practice of oncology (V.1.2011) [16]. All of the clinical data were collected by experienced physicians. The experimental operators were unaware of the clinical data. Written informed consent was obtained from all patients. The ethics committee of human studies at Ningbo University approved all aspects of this study.

2.2 Cell culture

Human gastric epithelial cell line GES-1 and gastric cancer cell lines HGC-27, AGS, BGC-823, SGC-7901, and MGC-803 were purchased from Shanghai Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences. AGS cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) High Glucose (HyClone, Logan, UT, USA) containing 1% penicillin/streptomycin (Life Technologies, Carlsbad, CA, USA) and 10% fetal bovine serum (FBS) (PAN-Biotech, Aidenbach, Germany) at 37 ∘ C and 5% CO 2 . Other cell lines were cultured in Roswell Park Memorial Institute (RPMI) medium 1640 (HyClone) with 10% FBS (PAN-Biotech), 1% penicillin/streptomycin (Life Technologies) at 37 ∘ C and 5% CO 2 [17].

2.3 Total RNA extraction

Total RNA in tissues/cells and plasma samples was first extracted with TRIzol reagent (Invitrogen, Karlsruhe, Germany) and TRIzol LS reagent (Invitrogen), respectively. Next, the purity and concentration were measured by SmartSpec Plus Spectrophotometer (Bio-Rad, Hercules, CA, USA) Finally, RNA was stored at - 80 ∘ C till use.

Assess RNA purity was based upon A260/A280 values, within 1.8–2.1. In addition, the quality of total RNA was further evaluated by 1% agarose gel electrophoresis (Fig. S1).

2.4 RNA pretreatment

tRFs and tiRNAs are heavily decorated by RNA modifications. rtStar™ tRF&tiRNA Pretreatment Kit (Arraystar, Rockville, MD, USA) is designed to remove the modifications that interfere with the quantitative polymerase chain reaction (qPCR). These modifications include terminal modifications that block adaptor ligation to the RNA ends and internal methylations that hinder reverse transcription for cDNA synthesis. Pretreatment (de-modification) of RNA was performed following the manufacturer’s protocol (Fig. S2A–E).

2.5 Reverse transcription

rtStar™ First-Strand cDNA Synthesis Kit (Arraystar) is designed to prepare cDNA for qPCR detection of tRFs and tiRNAs. The method sequentially ligates 3’ adaptor and 5’ adaptor (Fig. S2F and G). The 3’ adaptor contains a priming site for reverse transcription (RT) primer (Fig. S2H and I). The 3’ adaptor and 5’ adaptor were provided by Arraystar. RT was performed according to the manufacturer’s protocol.

2.6 Real-time qPCR

Real-time qPCR was performed with a GoTaq qPCR Master Mix (Promega, Madison, WI, USA) in the Mx3005P real-time PCR system (Stratagene, LaJolla, CA, USA) according to the manufacturer’s protocol (Fig. S2J). The full-length sequence of tiRNA-5034-GluTTC-2 is 5’-UCCCAUAUGGUCUAGCGGUUAG GAUUCCUGGUUU-3’. The sequences of PCR pri- mers for tiRNA-5034-GluTTC-2 were 5’-GTCCGAC GATCTCCCATA-3’ and 5’-TGTGCTCTTCCGATC TAAAC-3’. The primers include part sequences of tiRNA-5034-GluTTC-2 and adaptor specific. This allows for the specific amplification of tRFs and tiRNAs instead of mature tRNA. Therefore, an abnormal E value will occur when performing BLAST search with any of the primers. The primers’ sequences of snRNA (RNU6-2), as a control, were 5’-GCTTCGGCAGCACATATACTAAAAT-3’ and 5’-CGCTTCACGAATTTGCGTGTCAT-3’. These pri- mers were synthesized by Sangon Biotech (Shanghai, China). The Δ ⁢ C q method was used to analyze their expression levels, with a lower Δ ⁢ C q value indicating a higher expression [18]. The relative expression was calculated using the method of 2 - Δ ⁢ Δ ⁢ C ⁢ q [18] and all results were expressed as means ± SD of three independent experiments.

2.7 Cloning and sequencing of qRT-PCR products

The qRT-PCR products of tiRNA-5034-GluTTC-2 were first purified using the UNIQ-10 PCR product purification kit and then cloned into the pUCm-T vector (Thermo Fisher Scientific, Hudson, NH, USA), following the manufacturer’s instructions. DNA sequencing was performed by Thermo Fisher Scientific.

2.8 Northern blot

After dividing the total RNA molecules by size in a denaturing polyacrylamide gel, they were transferred to a nylon membrane (Amersham Biosciences, Piscataway, NJ, USA) and hybridized with immobilized RNA using a 32 P-labeled RNA probe. The probe binds to specific RNA, only by the pairing sequence, and the size and amount of the hybrid RNA can be displayed by the brightness and the size of the strip. The probe sequence of tiRNA-5034-GluTTC-2 for Northern blotting is 5’-AAACCAGGAATCCTAACCGCTAGACC ATATGGGA-3’.

2.9 Statistical analysis

All statistical analyses were performed by GraphPad Prism 6.0 (GraphPad Software, LaJolla, CA, USA) and Statistical Program for Social Sciences (SPSS) 20.0 software (SPSS, Chicago, IL, USA). The comparison of the relative expression levels of tiRNA-5034-GluTTC-2 between gastric cancer tissues and the paired adjacent normal tissues was analyzed by paired t -test. A comparison of the relative expression levels of tiRNA 5034-GluTTC-2 in the plasma between healthy people and patients with gastric cancer, on the day before surgery and seven days after surgery, was analyzed by one-way analysis of variance (ANOVA). The correlation between tiRNA-5034-GluTTC-2 levels and clinicopathological factors of patients with gastric cancer was further analyzed by χ 2 test. The diagnostic value of tiRNA-5034-GluTTC-2 was conducted by the receiver operating characteristic (ROC) curve. The optimal cut-off threshold was achieved at the highest Youden index. A value of 0.05 was considered to have a significant difference.

OPEN IN VIEWER

OPEN IN VIEWER

OPEN IN VIEWER

OPEN IN VIEWER

3.1 Amplification of tiRNA-5034-GluTTC-2

Since tiRNAs are small non-coding RNAs, its amplification by qRT-PCR needs adaptors (Fig. S2F–J). To analyze the specificity of the amplified tiRNA-5034-GluTTC-2 product, the melting curve analysis was first used, and the result revealed the amplification product as a single peak (Fig. S3), indicating no non-specific amplification or primer-dimers were produced. The Tm value is slightly increased due to the addition of the adaptors. To confirm the correction of qRT-PCR, the PCR products were further analyzed by T-A clone sequencing. The results showed that the sequence of the amplified tiRNA-5034-GluTTC-2 product (Fig. S4) was consistent with that in mintBase (https://cm.jefferson.edu/MINTbase/). Expression of tiRNA-5034-GluTTC-2 in tissues can also be detected by Northern blotting (Fig. 1). These results showed that tiRNA-5034-GluTTC-2 existed within the tissues and could be amplified by qRT-PCR.

OPEN IN VIEWER

We first detected the expression levels of tiRNA-5034-GluTTC-2 in 86 gastric cancer tissues and their paired adjacent non-cancerous tissues; as shown in Fig. 2A, tiRNA-5034-GluTTC-2 was significantly downregulated in cancer tissues. Among them, the lower expression of samples was accounted for 73.3% (63/86) (Fig. 2B). Since bloods are the most commonly used samples in clinical practice, we next collected paired plasma from patients with gastric cancer on the day before surgery and seven days after surgery, along with the plasma of healthy people. The results indicated that the levels of tiRNA-5034-GluTTC-2 in preoperative and postoperative gastric patients was reduced when compared with the healthy control (Fig. 2C).

Next, we performed an analysis to identify whether tiRNA-5034-GluTTC-2 expression levels were associated with clinicopathological features of patients with gastric cancer. We divided the expression levels of tiRNA-5034-GluTTC-2 into high and low groups and then analyzed the relationship between the expression levels of tiRNA-5034-GluTTC-2 and the clinicopathological factors. The results showed that tiRNA-5034-GluTTC-2 expression levels were associated with tumor size ( 0.001; Table 1).

The area under the ROC curve (AUC) of tiRNA-5034-GluTTC-2 was 0.779 and 0.835 in tissues and plasmas, respectively. In combined use, the AUC was up to 0.915, with 84.7% sensitivity and 92.8% specificity (Fig. 3). The cut-off was 9.7 and 8.9 in tissues and plasmas, respectively.

Further tiRNA-5034-GluTTC-2 levels in gastric cancer tissues from patients before the operation were categorized into two groups; the higher or lower group, according to the expression levels. The results of survival curves showed that patients with lower tiRNA-5034-GluTTC-2 expression group had a much shorter overall survival than those in the higher expression group ( P = 0.0158; Fig. 4). Multivariate analysis indicated that lymphatic metastasis (95% CI 1.635–10.797; P = 0.003) was an independent predictor in the prognosis of gastric cancer (Table 2).

To obtain more information about tiRNA-5034-GluTTC-2 expression in gastric cancers, we further analyzed tiRNA-5034-GluTTC-2 expression in human gastric cancer cell lines. As expected, the results showed that tiRNA-5034-GluTTC-2 was significantly downregulated in five gastric cancer cell lines, AGS, MGC-803, HGC-2, BGC-823, and SGC-7901, compared with the human normal gastric epithelial cell line GES-1 (Fig. 5).