Global analysis of tRNA and translation factor expression reveals a dynamic landscape of translational regulation in human cancers

Zhao Zhang,Leng Han,Lixia Diao,An-Yuan Guo,Hang Ruan,Jiqiang Ling,Chunyan Cai,Jing Gong,Chun-Jie Liu,John N Weinstein,Yu Xiang,Youqiong Ye

doi:10.1038/s42003-018-0239-8

Abstract

The protein translational system, including transfer RNAs (tRNAs) and several categories of enzymes, plays a key role in regulating cell proliferation. Translation dysregulation also contributes to cancer development, though relatively little is known about the changes that occur to the translational system in cancer. Here, we present global analyses of tRNAs and three categories of enzymes involved in translational regulation in ~10,000 cancer patients across 31 cancer types from The Cancer Genome Atlas. By analyzing the expression levels of tRNAs at the gene, codon, and amino acid levels, we identified unequal alterations in tRNA expression, likely due to the uneven distribution of tRNAs decoding different codons. We find that overexpression of tRNAs recognizing codons with a low observed-over-expected ratio may overcome the translational bottleneck in tumorigenesis. We further observed overall overexpression and amplification of tRNA modification enzymes, aminoacyl-tRNA synthetases, and translation factors, which may play synergistic roles with overexpression of tRNAs to activate the translational systems across multiple cancer types.

Highlights

The protein translational system, including transfer RNAs and several categories of enzymes, plays a key role in regulating cell proliferation
The transfer RNAs (tRNAs) expression data were merged to the codon level and amino acid level according to the anticodon and amino acid information (Supplementary Figure 1A)
Our analysis showed a high correlation at the tRNA level (Spearman's correlation Rs = 0.73, p < 2 × 10−16), codon level (Rs = 0.61, p = 7.2 × 10−7), and amino acid level (Rs = 0.59, p = 4.1 × 10−3, Supplementary Figure 1B) between both data types, indicating the reliability of our computational pipeline to infer the relative expression levels of tRNAs

Summary

Introduction

The protein translational system, including transfer RNAs (tRNAs) and several categories of enzymes, plays a key role in regulating cell proliferation. Despite the essential functions of tRNAs in the cell, it is still challenging to perform high-throughput quantification of tRNAs, mainly due to the presence of post-transcriptional modifications and secondary structures[16] To address these challenges, several methods have been designed to quantify tRNA expression level, including tRNA microarrays, which can only achieve codon level resolution by recognizing the tRNA’s anticodon loop[17,18,19], and tRNA-sequencing methods, such as demethylase-tRNA-seq (DMtRNA-seq)[16], which has been applied in a few cell lines[16,20,21]. Multiple categories of enzymes are involved in translational regulation, including the tRNA modification enzymes, aminoacyl tRNA synthetases (ARSs), and translation factors. We performed a comprehensive analysis elucidating a dynamic landscape of translational regulation, including tRNAs, tRNA modification enzymes, ARSs, and translation factors, across multiple cancer types in TCGA. Our results highlight a synergistic activation of the translational system in cancer

Methods

Results

Conclusion