On the expanding roles of tRNA fragments in modulating cell behavior

Abstract

The fragments that derive from transfer RNAs (tRNAs) are an emerging category of regulatory RNAs. Known as tRFs, these fragments were reported for the first time only a decade ago, making them a relatively recent addition to the ever-expanding pantheon of non-coding RNAs. tRFs are short, 16-35 nucleotides (nts) in length, and produced through cleavage of mature and precursor tRNAs at various positions. Both cleavage positions and relative tRF abundance depend strongly on context, including the tissue type, tissue state, and disease, as well as the sex, population of origin, and race/ethnicity of an individual. These dependencies increase the urgency to understand the regulatory roles of tRFs. Such efforts are gaining momentum, and comprise experimental and computational approaches. System-level studies across many tissues and thousands of samples have produced strong evidence that tRFs have important and multi-faceted roles. Here, we review the relevant literature on tRF biology in higher organisms, single cell eukaryotes, and prokaryotes.

Figure 1.

The structural types of the various tRFs. tRFs can be produced from either the precursor tRNA or from the mature tRNA. For more detail, refer to the description in the text.

Figure 2.

Visual summary of tRF mechanisms in the literature during the last 10 years. These illustrations represent tRF mechanisms and behaviors that involve either direct or indirect effects on protein or mRNA and have been confirmed with experimentation. Citations are provided, including first author last name and year of publication. Note the use of single letter codes to denote the organism in which the work was done: A – A. thaliana; B – B. mori; D – D. melanogaster; E – A. aegypti; H – H. sapiens; M – M. musculus; P – T. thermophilus; R – R. norvegicus; T – T. brucei or T. cruzi; V – H. volcanii; and, Y – S. cerevisiae. Red shading indicates processes that are downregulated by tRFs whereas green shading indicates activities that are upregulated by tRFs.

Figure 3.

Visual summary of the findings that emerged from the analyses of all TCGA cancers previously described by Rigoutsos and colleagues (4,37,43,52–57,59). The analyses are based on computing positive and negative correlations between tRFs and mRNAs (‘co-expression networks’). These correlations capture direct molecular couplings as well as indirect interactions (e.g. decoying events and propagated regulatory effects). For those mRNAs that participate in the correlations with tRFs, their intronic and exonic lengths, their respective genomic spans, the repetitive content of those genomic spans, and the cellular localization of the proteins that are produced by these mRNAs were also examined. The picture that emerges is complex, yet remarkably consistent across the 32 TCGA cancer types. Summarily, numerous tRFs of nuclear and mitochondrial origin are correlated with mRNAs. These mRNAs belong to key processes including development, receptor tyrosine kinase signaling, the proteasome, and metabolic pathways. The mRNAs that participate in positive correlations are generally shorter and enriched in repetitive elements. On the other hand, the mRNAs that participate in negative correlations are longer and depleted in repetitive elements. Perhaps the most striking finding that emerges from this analysis is the prominent participation of the mitochondrial tRFs in correlations with mRNAs that belong to processes that are not mitochondria-specific. This raises the possibility of the participation of mitochondrial tRFs in an ‘information exchange’ that could be implemented in one of two ways: either the mitochondrial tRFs exit the mitochondrion and are shuttled to the cytoplasm and the nucleus, or they are produced by transcription of the mitochondrial ‘tRNA-lookalikes’ encoded in the nuclear genome and subsequent processing. The picture that emerges suggests that older and younger categories of repetitive elements as well as gene architecture are tightly-coupled with tRFs of nuclear and mitochondrial origin to a wider information exchange framework. The use of dashed gray lines in this figure is meant to show relationships that arise from the analysis of TCGA cancers and await independent experimental validation.