Plasma tRNA-derived small RNAs signature as a predictive and prognostic biomarker in lung adenocarcinoma

Abstract

Background: The prevalence of lung adenocarcinoma (LUAD) has increased, thus novel biomarkers for its early diagnosis is becoming more important than ever. tRNA-derived small RNA (tsRNA) is a new class of non-coding RNA which has important regulatory roles in cancer biology. This study was designed to identify novel predictive and prognostic tsRNA biomarkers.

Methods: tsRNAs were identified and performed differential expression analysis from 10 plasma samples (6 LUAD and 4 normal, SRP266333) and 96 tissue samples (48 LUAD and 48 normal, SRP133217). Then a tsRNA-mRNA regulatory network was constructed to find hub tsRNAs. Functional enrichment analysis was performed to infer the potential pathways associated with tsRNAs. Afterwards, a Support Vector Machine (SVM) algorithm was used to explore the potential biomarkers for diagnosing LUAD. Lastly, the function of tRF-21-RK9P4P9L0 was explored in A549 and H1299 cell lines.

Results: A significant difference of read distribution was observed between normal people and LUAD patients whether in plasma or tissue. A tsRNA-mRNA regulatory network consisting of 155 DEtsRNAs (differential expression tsRNAs) and 406 DEmRNAs (differential expression mRNAs) was established. Three tsRNAs (tRF-16-L85J3KE, tRF-21-RK9P4P9L0 and tRF-16-PSQP4PE) were identified as hub genes with degree > 100. We found Co-DEmRNAs (intersection of DEtsRNAs target mRNAs and differentially expressed mRNAs in LUAD) were engaged in a number of cancer pathways. The AUC of the three hub tsRNAs' expression for diagnosing LUAD reached 0.92. Furthermore, the qPCR validation of the three hub tsRNAs in 37 paired normal and LUAD tissues was consistent with the RNA-Seq results. In addition, tRF-21-RK9P4P9L0 was negatively associated with LUAD prognosis. Inhibition of tRF-21-RK9P4P9L0 expression reduced the proliferation, migration and invasion ability of A549 and H1299 cell lines.

Conclusion: These findings will help us further understand the molecular mechanisms of LUAD and contribute to novel diagnostic biomarkers and therapeutic target discovery.

Keywords: Diagnostic biomarker; Lung adenocarcinoma; Network; tRFs; tsRNAs.

Fig. 1

Read distribution of small RNAs and tsRNA sub-type in plasma and tissue of normal people and LUAD patients. A Small RNA distribution in plasma of normal people and LUAD patients, the right panel displayed the P value of the difference in the proportion of each type of reads between normal and LUAD samples; B Small RNA distribution in tissues of normal people and LUAD patients, the right panel displayed the P value of the difference in the proportion of each type of reads between normal and LUAD samples; C Expressed tsRNAs sub-type numbers in plasma of normal people and LUAD patients using violin illustration; percentage of tsRNAs sub-type numbers in plasma of normal people (D) and LUAD patients (E) using pie chart; F expressed tsRNAs sub-type numbers in tissues of normal people and LUAD patients using violin illustration; percentage of tsRNAs sub-type numbers in tissues of normal people (G) and LUAD patients (H) using pie chart

Fig. 2

Differential expressed tsRNAs and mRNAs in LUAD. Differential expressed tsRNAs in plasma (A, SRP266333) and tissue (B, SRP133217); C intersection analysis of DEtsRNAs in plasma and tissues; D differential expressed mRNAs in LUAD using TCGA dataset. The grey dots represent genes which under significate differential expression cutoff |log2FC|> 1 and P-value < 0.05, the red plots displayed the up-regulated tsRNAs or mRNAs, the blue plots displayed the down-regulated tsRNAs or mRNAs. LUAD, lung adenocarcinoma; DE, differential expression; Co-DE, common differentially expressed

Fig. 3

The tsRNA-mRNA regulatory network. A Visualized regulatory network of tsRNA-mRNA, nodes were colored to distinguish sub-types of tsRNA and mRNA, edges were colored to identify different action modes, the larger the node, the higher degree; B statistical analysis of the number of mRNAs which tsRNA targeted; C statistical analysis of the number of tsRNA targets among mRNA

Fig. 4

Functional enrichment of Co-DEmRNAs. A Graph of the top ten results from the GO analysis in terms of BP; B graph of the top ten enrichment pathways in KEGG; C graph of the top ten enrichment pathways in Recotome; D graph of the top ten enrichment pathways in Wikipathways; E PPI network of the Co-DEmRNAs, different clusters were marked with colors; F exhibition of 14 closely connected clusters. Co-DE, common differentially expressed; GO, Gene Ontology; BP, biological process; KEGG, Kyoto Encyclopedia of Genes and Genomes; PPI: protein–protein interaction

Fig. 5

Receiver operating characteristic curves for the hub tsRNAs in plasma and tissue to distinguish LUAD patients from normal people. The AUC values obtained by using tRF-16-L85J3KE, tRF-21-RK9P4P9L0 and tRF-16-PSQP4PE individually in plasma (A) and tissue (B); The AUC values obtained in combination of tRF-16-L85J3KE, tRF-21-RK9P4P9L0 and tRF-16-PSQP4PE in plasma (C) and tissue (D). LUAD: lung adenocarcinoma; AUC: area under the receiver operating characteristic curve

Fig. 6

Potential tsRNAs as biomarkers. Dot plot displayed tRF-16-L85J3KE, tRF-21-RK9P4P9L0 and tRF-16-PSQP4PE expression level in plasma (A) and tissues (B); expression levels of tRF-16-L85J3KE (C), tRF-21-RK9P4P9L0 (D) and tRF-16-PSQP4PE (E) in LUAD tissues and paired normal tissues; F survival analysis of tRF-21-RK9P4P9L0 in LUAD; G structure and sequence of tRF-21-RK9P4P9L0

Fig. 7

Functional analysis of tRF-21-RK9P4P9L0 using bioinformatics methods. A Merged network of tRF-21-RK9P4P9L0 and its target mRNAs, grey edge: tsRNA-mRNA interaction, red edge: protein–protein interaction; B tRF-21-RK9P4P9L0’s target gene in PPI network, individual clusters are differently colored; C the core sub-network of merged network

Fig. 8

Functional analysis of tRF-21-RK9P4P9L0 in LUAD cell lines. A tRF-21-RK9P4P9L0 expression level in A549-NC and A549-tRF-21-RK9P4P9L0; B Notch1 expression level in A549-NC and A549-tRF-21-RK9P4P9L0; C proliferation rates of A549-NC and A549 tRF-21-RK9P4P9L0; D migration ability of A549-NC and A549 tRF-21-RK9P4P9L0; E invasion ability of A549-NC and A549 tRF-21-RK9P4P9L0; F tRF-21-RK9P4P9L0 expression level in H1299-NC and H1299-tRF-21-RK9P4P9L0; G Notch1 expression level in H1299-NC and H1299-tRF-21-RK9P4P9L0; H proliferation rates of H1299-NC and H1299 tRF-21-RK9P4P9L0; I migration ability of H1299-NC and H1299 tRF-21-RK9P4P9L0; J invasion ability of H1299-NC and H1299 tRF-21-RK9P4P9L0