The diagnostic/prognostic potential and molecular functions of long non-coding RNAs in the exosomes derived from the bile of human cholangiocarcinoma

Cholangiocarcinoma (CCA) is an aggressive malignancy associated with unfavorable prognosis, and it’s difficult to diagnose and no effective treatments are available. Long non-coding RNAs (lncRNAs) play important roles in tumorigenesis and metastasis. Intact lncRNAs from exosomes have sparked much interest as potential biomarker for the non-invasive analysis of disease. Here, via exosome sequencing on lncRNAs, GO analysis, KEGG pathway and co-expression analysis, receiver operating characteristic curve and survival analyses, we found that, compared with control group, lncRNAs of ENST00000588480.1 and ENST00000517758.1 showed significantly increased expressions in CCA group. Moreover, area under the curve (AUC) was increased to 0.709 when combined the two lncRNAs, they had a sensitivity and specificity of 82.9% and 58.9% respectively. Further, the higher levels of the two lncRNAs showed a significantly increasing trend with the advancement of cancer TNM stages, and prognosticated a poor survival. In addition, KEGG pathway analysis showed that the most significant difference term was “p53 signaling pathway” (KEGG ID: hsa04115, p: 0.001). The altered lncRNAs and their target genes were included to reconstruct a co-expression network. These altered lncRNAs were mainly related to cellular processes, environmental information processing and organismal systems, etc. Collectively, our findings provided the potential roles of lncRNAs of ENST00000588480.1 and ENST00000517758.1 in CCA, and implicated these lncRNAs as potential diagnostic and therapeutic targets for CCA.


SUPPLEMENTARY MATERIALS Exosome identification
To observe the size and shape of exosomes, transmission electron microscopy (TEM) and multiparameter nanoparticle optical analysis were performed. Exosome-specific membrane proteins such as Alix, TSG101, CD81, CD82, CD63, CD9, CD24, EpCAM, Hsc70 are commonly used. They can be ditected by flow cytometry analysis and western blotting. In our study, we performed Flow cytometry analysis for CD63 and CD81.

TEM
Resuspend extracellular vesicles in 100 μl of PBS. Adsorb 20 μl to 400 mesh carbon-coated Parlodion copper grids for 2 min and allow to dry at RT. Fix the exosome in 1% (v/v) glutaraldehyde (EM-grade) for 5 min at RT by placing drops carefully on the dried preparation. Wash the grids twice with water for 5 min and then contrast stain EVs with 1% phosphotungstic acid (PTA) for 30 sec. Acquire images with an electron microscope with an acceleration voltage of 80 kilovolts starting at magnifications of 20,000X and increasing to 100,000X when determining the size of the particles. Acquire the images at 200 kV with magnifications of Multi-parameter nanoparticle optical analysis. Resuspend the isolated exosomes in 1ml aseptic cold PBS. Slowly poured into a disposable clean sample pool, avoiding bubbles, then sealed the sample pool with a lid. Visualize the exosomes by ZETASIZER Nano series-Nano-ZS (Malvern, U.K.) according to manufacturer's protocol.

Real-time reverse transcription-PCR
We used the bile RNA concentration as a quantitative standard. Total RNA was isolated from cells using the Trizol Total RNA Isolation kit (Invitrogen, Carlsbad, CA) according to the manufacturer's protocol. RNA was eluted with RNasefree water. Reverse transcription (RT)-PCR was performed using the Transcriptor First Strand cDNA Synthesis kit (Roche Molecular Biochemicals, Indianapolis, IN) according to the manufacturer's protocol. Then the mRNAs were amplified by qRT-PCR with SYBR green (Invitrogen) under the following conditions: 95 °C for 30 sec; 40 cycles of 95 °C for 5 sec; 60 °C for 30 sec; and melting at 95 °C for 5 sec, followed by 60 °C for 1 min, and cooling at 50 °C for 30sec. The mRNAs expression was normalized to that of PLA ( details of chosing method was described in the next paragraph).
Fold changes in expression of each gene were calculated by a comparative threshold cycle (Ct) method using the formula 2 -(ΔΔCt) . The primers used in this study were listed in the Supplementary Table 2.

Reference gene identification
To systematically select reference genes which are applicable in exosome RNA of bile (25 paired CCA and biliary stricture samples), 13 common reference candidates were chosen. Of our reference candidates, β-actin, L13, and Tub are structure related. HPRT, PBGD, GAPDH, G6PDH, and phospholipase A2 (PLA) are metabolism related. TBP and RNA polymerase II (RPII) are transcription related, whereas albumin (Alb), β2M, and PPIA cannot be clearly put into above categories. The sequences of primers of these reference genes were listed in Supplementary Table 3. The most stable reference genes in exosome RNA were determined using Bestkeeper which was applied in order to visualize the centrifuged pellet. PLA ranked the most suitable reference gene in bile exosome RNA.

RNA sequencing data analyses
After sequencing, the raw reads should be filtered according to criterias: ribosomal RNA sequences, low quality reads and reads with sequence adaptors were removed. After the filtration, the clean reads were aligned to the human genome by Tophat-2 software. Both the value of "read-mismatches" and "read-gap-length" were set as 2, and the coverage as well as depth of the sequencing data were assessed. The resulting data were assembled into transcripts by the Cufflinks. The assembled transcripts were filtered using NCBI, RNA central, and GENCODE to select the known lncRNA and the known protein-coding mRNA. The relative expression levels of lncRNAs and mRNAs were measured as expected number of reads per kilobase of transcript sequence per million base pairs sequenced (RPKM) by HTSeq. DEGseq software was used to identify differentially expressed genes based on the following criteria: |log2 (fold change)| > 1 and p < 0.05. Target genes were predicted according to databases such as ncbi, RNAcentral, genecode and lncRNAdb. Pathway analysis (based on Kyoto encyclopedia of genes and genomes, KEGG, http://www.genome.jp) and gene ontology (GO) analysis were performed on the target genes of differentially expressed lncRNAs to predict the biological roles of them, and p < 0.05 used as threshold for defining significantly enriched GO terms and pathways. The co-expression network was constructed based on