This article is from
Journal of Creation 35(1):104–110, April 2021

Browse our latest digital issue Subscribe

The surprisingly complex tRNA subsystem: part 4—tRNA fragments regulate processes

by Royal Truman

In part 1¹, part 2² and part 3³ of this series we examined the complex enzymatic processing necessary to generate functional tRNAs. Dozens of protein-based molecular machines were therefore necessary from the very beginning for the genetic code to work. In part 5,⁴ we will consider whether evolution is a plausible explanation for these findings.

Here in part 4, we will discuss additional biological services provided by tRNAs and also by degradation fragments generated from tRNAs. Instead of being worthless RNA fragments, these are used for important regulatory purposes. The intention of this paper is to help gain a more comprehensive picture of the vast number of interacting processes involved in a cell and how to interpret this fact. Evolutionists assume that life must have started very simple and complexity was added incrementally through a vast number of individual fortuitous accidents upon which natural selection could then act. An understanding of what needs to be explained in a reasonable amount of detail is necessary before blindly assuming that natural, unguided chemistry is causally sufficient to provide a feasible pathway from an abiogenetic state to functioning cells.

tRNA cleavage products had been considered biologically irrelevant for more than three decades. This is the kind of error which results from the wrong mindset, that cells should be as simple as possible to make evolution credible. Given the vast number of kinds of tRNA fragments and very low concentrations of most of them, many scientists were blinded by their presuppositions to thinking these did not merit serious research.

The tRNA world is now so vast that all the processes affected and regulated are almost beyond comprehension.⁵ The potential for utter cellular chaos is overwhelming without an initial master plan as the basis for carefully implemented controlled activities.

We will point out next that tRNAs do more than only serve as linker molecules to translate codons.

tRNAs regulate protein translation

It is common for individual cellular components to be used for multiple purposes, analogous to how a metal gear or software subroutine is reused in different contexts. Modular designs based on standardized elements are characteristic of well-engineered systems. It is now known that tRNAs collaborate in many cellular activities in eukaryotes, bacteria, and archaea beyond serving only as adaptor molecules. Generally, the concentration of uncharged tRNAs in cells is low, so when translation slows down for any reason free tRNA serves as a signal to regulate other linked processes. For example, free tRNA can interact with the protein kinase⁵ MEK2 and thereby inhibit the latter’s activity, slowing down cell-cycle progression until translation can resume at full capacity.⁶

As another example of translation regulation, under stress or nutrient limitation the amount of uncharged tRNA increases, activating the kinase GCN2. This then increases phosphorylation of the eukaryotic initiation factor eIF2α thereby hindering delivery of the initiator tRNA^Met to the ribosome. The result is also a slowdown of translation.⁶

tRNAs involved in non-translation processes

Many tRNAs are also involved in other pathways which are not related to genetic code translation,⁷ for example in the first step of heme and chlorophyll biosynthesis, as reverse transcription primers, and for strand transfer during retroviral replication.⁸ tRNAs can interact with the cytochrome C protein preventing it from binding to the caspase activator Apaf-1, thus helping to regulate apoptosis (cell suicide for the good of an entire organism).⁶ Eukaryotic tRNAs participate in targeting proteins for degradation via the N-end rule pathway,⁸ which helps regulate protein concentrations to the necessary levels.⁹

Aminoacylated tRNAs involved in non-translation processes

Aminoacylated tRNAs (aa-tRNAs) are also involved in non-ribosomal peptide bond formation, post-translational protein labelling, modification of phospholipids in the cell membrane, and antibiotic biosynthesis.¹⁰

Regulation with tRNA-derived small RNA

So far, we have focused on tRNA molecules, and now we will examine RNA degradation fragments derived from them. tRNA-derived small RNAs (tsRNAs) have been identified in all domains of life, and aberrant tsRNA expression is linked to several diseases such as cancer and neurological disorders.^11-14 Since this is still a new area of research, a consistent nomenclature to describe these specially crafted fragments has not been established yet.¹⁵ Online databases of tsRNAs are available,¹⁶ which also include those derived from tRNAs encoded by organelles such as mitochondria and chloroplasts.¹¹

Even though the same kinds of tsRNAs are formed, many of the enzymes used (ribonucleases) by eukaryotes are different than those found in bacteria and archaea, and these ribonucleases are often tRNA-type specific.¹⁵ We propose that this is best explained as examples of different designs used in order to solve similar requirements in an optimal manner.

Three key enzymes are involved in the generation and maturation of tsRNAs, namely RNase Z, Dicer, and angiogenin¹⁴ and the non-coding RNA fragments generated from cleavage of both tRNA and pre-tRNA participate in regulating several cellular processes,¹⁷ including the cellular RNA interference (RNAi) system.¹⁰

Studies on tsRNAs have revealed commonalities with the better-known miRNAs (micro-RNAs).¹⁸ Like miRNAs, tsRNAs are also known to base-pair with complementary sites of mRNA (messenger-RNA) to prevent their translation.¹⁷

The presence of virus can stimulate the production of tsRNAs which are able to interfere with reverse transcription of RNA of virus origin. The abnormal expression of tsRNA has been shown to be strongly associated with various diseases, such as tumours, the cardiovascular system, epigenetics,¹⁹ and neurological diseases,¹⁴ but properly expressed help hinder cancerous growth.¹⁷

There are two families of tsRNAs (figure 1): tiRNAs and tRFs, which we will discuss next.

tiRNAs (tRNA-derived stress-induced RNAs)

tiRNAs, also called tRNA-halves, are cleaved specifically in the anticodon loop to produce two fragments: 30–35 nucleotide (nt) long 5’-tRNA halves and 40–50 nt 3’-tRNA halves²⁰ (figure 2). The enzymes responsible for this cleavage are Rny1¹² and angiogenin (ANG), a member of the pancreatic RNase superfamily,¹³ but the mechanisms have not been worked out in detail yet. ANG targets preferentially the dinucleotide ‘CA’ pattern in single-stranded RNA, with ~20-fold higher preference for CA over UA and 3-fold higher for CA over CG.¹³

tiRNAs are produced almost exclusively under stress conditions, for example, amino acid deficiency, hypoxia (oxygen deprivation), UV radiation, oxidative damage, heat shock, phosphate starvation, presence of arsenite, and viral infection,^12,17 promoting assembly of stress granules²¹. Studies indicate that tiRNAs might also react to environmental challenges via mechanisms that do not involve the well-known integrated stress response.¹³ For example, tRNA cleavage was not observed in other forms of stress investigated, like γ-irradiated, etoposide-treated or caffeine-treated human cells.¹³

The concentration of mature tRNA is not significantly altered by formation of tiRNAs. Less than 5% of the tRNA pool is cleaved, and only 5’-tiRNAs, but not 3’-tiRNAs inhibit translation in cultured cells. Moreover, only tiRNAs derived from tRNA^Ala and tRNA^Cys have been shown to be responsible for the inhibition,²⁰ and the nature of cleavage is regulated according to type of tissue. The level and composition of tiRNAs have also been shown to change with age and calorie intake.¹³

tiRNAs are believed to help suppress metastasis and tumour formation.²² Fragments introduced into breast cancer cells decreased cancer growth, whereas inhibition of these fragments increased the cancer phenotype.¹³ ANG-induced tiRNAs are also involved in the cellular response to virus infection¹³ and can act as signalling molecules that participate in cell-to-cell communication and in immune signalling.¹³ Thus, tiRNAs can serve as biomarkers to detect stress-induced tissue damage, which is important since uncontrolled tissue damage is a common underlying cause of cancer (figure 2).¹³

Several studies have shown that various diseases result if tiRNA expression is dysregulated under stress conditions,¹⁴ and modification of specific nucleosides seems to help regulate their expression by providing resistance to ANG-induced cleavage in tRNAs during stress responses.¹³

There is also a group of special tiRNAs known as sex-hormone-dependent tRNA-derived RNAs, found in breast and prostate cancer cell lines, which are not produced under stress conditions.¹⁴

A comprehensive study involving several institutes from India and Singapore examined how specific tiRNAs are involved in cell-state transitions using mouse embryonic stem cells. They found differential enrichment in a broad range of cell states.²³

tRFs (tRNA-derived fragments)

The second family of tsRNAs consist of tRNA-derived fragments (tRFs) which are 18–22 nt in length (figure 1). These tRFs have been found in all domains of life, but the processing details are different. For example, eukaryote tRFs which are produced inside the nucleus must be exported into the cytoplasm.²⁴

tRF are not irrelevant degradation products

The evidence includes several observations: (a) tRFs are not always derived from the more abundant tRNAs, nor do the numbers of tRFs correlate with the gene of the parental tRNAs; (b) fragmentation patterns depend on the specific parental tRNA; (c) fragmentation patterns change according to developmental stage or cellular conditions; and (d) some tRFs are bound to Argonaute/Piwi proteins, well-known components of the RNA-induced silencing complex.²⁵

Compared with randomly degraded fragments, tRFs possess at least the following three characteristics: (1) remarkable site-specificity; (2) defined lengths; and (3) significantly higher abundance. This allows them to be reliably identified using computer tools.²⁶

In one detailed study²⁵ the expression levels of mitochondrial and nuclear tRFs were found to differ significantly during the six stages (the egg, 1^st−4^th instar larvae, and adult) of the tadpole shrimp Triops cancriformis, a ‘living fossil’ whose morphological form is claimed not to have changed in almost 200 million years. The total read number of nuclear tRFs was 412 times larger than the number of mitochondrial tRFs and the amount of tRFs deriving from each kind of parental tRNA varied considerably. For example, among the mitochondrial tRFs, tRF^Ser(GCU) was most abundant (30.1 %).²⁵ In the case of nuclear tRF, 72.9% derived from only one kind of tRNA, tRNA^Gly(GCC).²⁵

Aberrant tRNAs are a source of RFs

tRFs are extracted from tRNAs and tRNA precursors by nucleases Dicer and RNase Z.¹⁷ In particular aberrant and misfolded tRNAs are recognized and cleaved,¹⁵ serving a quality control function. But tRFs also play many useful cellular functions directly. They form in response to various cellular stresses (e.g. oxidative stress, UV irradiation, and nutrient starvation) and those correctly formed are often protected from degradation by addition of various ligands.²⁴

There are four types of tRFs

Many researchers distinguish between tRF-5s, tRF-3s, tRF-1s, and tRF-2s, classified based on the part of the mature tRNA or pre-tRNA from which they are derived (figure 1)¹⁷ but the abundances of tRF-1s and tRF-2s are usually considerably lower than of tRF-3s and tRF-5s.¹⁴ The tRF-5s are further classified by their lengths: tRF-5a (14–16 nts), tRF-5b (22–24 nts), and tRF-5c (28–30 nts) and the lengths of these tRF-5 series follow a normal distribution.¹⁴

Generally, tRF-1s vary in length depending on the location of a termination signal within each precursor tRNA.³⁴ Since tRNA genes can come in multiple variants, this leads to a variety of tRFs which then provide a means to help regulate different processes in more precise manners. This design-centric interpretation views the existence of multiple tRNA gene copies not as the result of meaningless accidental gene duplication events but planned variety and adaptability.

tRFs are involved in translation-related cellular processes

Evidence is accumulating that tRFs can regulate translation, by displacing initiation factors or by displacing mRNA from the initiation complex. Studies revealed that 18-nt 5’ tRFs having the terminal oligoguanine (TOG) motif modified to pseudouridine (Ψ) at the eighth uridine position can inhibit global translation, whereas the normal uridine-bearing 18-nt 5’ tsRNAs cannot.^12,27 It requires much credulity to believe that countless examples like these of extreme specificity executed by complex enzymes to perform key cellular processes, arose by random mutations.

Another study found that a valine tRNA-derived fragment (Val-tRF) produced under certain stress conditions in eukaryotes and archaea can bind to the small ribosomal subunit. This then displaces mRNA from the initiation complex, decreasing the amount of protein formed.²⁸ In S. cerevisiae both 5’- and 3’-parts of tRNA fragments can bind directly to ribosomes,²⁹ but to different sites. These interactions are stress-dependent and inhibit protein synthesis.²⁹ In other words, there exists a prepared solution able to react almost instantly when the problem arises.

The tRF interactions with key proteins of the translation machinery, whether ribosomal proteins or translation initiation factors, rely on sequence-specific binding.¹¹ Using data from several Drosophila genomes, the authors of one study reported that the 7-nt target sites associated with a reduced translation efficiency were perfectly complementary to the respective 5’tsRNAs.¹¹ Intriguingly, examples are also known of increases in translation for at least two ribosomal proteins by binding to sequence-complementary target sites that presumably lie within the coding sequence of their mRNA.¹¹

Some tRFs assemble into intermolecular G-quadruplexes (RG4s), structures consisting of four guanine molecules which interact with each other through Hoogsteen hydrogen bonding and are further stabilized by cations like Na⁺ and K⁺.^30,31 Selected 5’tiRNAs (e.g. tiRNA^Ala and tiRNA^Cys) bearing the terminal oligoguanine (TOG) motif (four to five guanine residues) at their 5’-termini displace translation initiation factor eIF4A/F/G from mRNAs.¹²

Some tRFs can also bind to the exact aminoacyl-tRNA synthetases involved in producing their parent tRNA.^12,14 In yeast, they were reported to bind to ribosome-associated aminoacyl-tRNA synthetases and thereby at least in vitro inhibit translation by regulating tRNA aminoacylation.¹² This is an example of feedback self-regulation, which causes an overproduced biomolecule to slow down its own creation. Robustness and regulation to ideal concentrations are typical of intelligently designed engineered systems.

tRFs can mimic the principle used by miRNAs to fine-tune translation rates by interacting with mRNAs. A statistical analysis of a dataset generated from The Cancer Genome Atlas representing 32 types of cancer revealed a multitude of statistically significant and context-dependent associations between various tRFs and mRNAs.³² The tRF-mRNA circuitry was found to depend on a patient’s gender and the expression levels of tRFs are also known to change with oncogene³³ activation and cancer progression regulating the occurrence and development of tumours.³⁴

One particular tRF fragment, tRF5^Glu(CTC) (tFR5-GluCTC) is predicted to be able to bind to at least 44 mRNAs with a minimum free energy of ≤ −30 kcal/mol,³⁵ which is typical of miRNAs. This is of interest and known through medical research because infection by the Respiratory Syncytial Virus (RSV) interferes with these 44 mRNAs and RSV can deactivate the host immune defences combating it. The mRNA of a specific target of tFR5-GluCTC, apolipoprotein E receptor 2 (APOER2) was predicted to have the best binding and was examined in depth³⁵ (figure 3). APOER2 was shown to inhibit RSV replication by interacting with key RSV protein(s),³⁵ so interfering with APOER2’s mRNA would enhance the virus survival. This is exactly what happens upon infection by RSV: this induces production of tRF5-GluCTC, leading to a lower level of APOER2.³⁵ We will speculate on the significance of this from a design point of view below.

Several variants of tFR5-GluCTC were synthesized to provide fewer nt pairing with APOER2, and the effect of exposure to RSV was examined. Surprisingly, the fewer and dispersed pairings at the 3’ end of tFR5-GluCTC were found to affect concentration of APOER2 most strongly.

Two facts may point to a design purpose behind a virus with RSV-like properties. First, it seems noteworthy how many genes can be regulated by tFR5-GluCTC, around 44. In addition, APOER2 belongs to a class of lipoprotein receptors which promote the replication of some Flaviviruses via endocytosis.³⁶ Perhaps the original design and intention was to use some viruses to regulate genes, endowed also with autoregulation. One can envision that the correct amount of APOER2 could control how much of a virus would survive and thereby regulate dozens of mRNAs. It is well established that there are diseases caused by malfunctioning regulation. The concept is shown in figure 4.

Some scientists, such as Dr Peter Borger from Wort und Wissen in Germany, have suggested that viruses seem to have originally been encoded on genomes for useful purposes, reminiscent of transposable elements, but mutated and became transmissible and harmful as a consequence of the Fall.³⁷

tRFs are also involved in non-translation-related cellular processes

tRNA fragments are important for a wide range of biological processes, and millions of small tRNA-derived molecules can interact with cellular proteins and mRNAs.⁶

Torres and colleagues at the Barcelona Institute of Science and Technology detected several examples of significant differences in expression level among genes in the same isodecoder³⁸ affecting tRFs levels but not that of mature tRNAs.³⁹ Variations of individual isodecoder genes should not affect translation timing or amount, so their experiments reinforce the new understanding that tRNA genes are often also involved in regulating functions which are not related to serving as adaptor molecules for mRNA translation.

Indeed, tRFs are an emerging class of regulatory RNAs which play a role in producing various noncoding RNAs—snoRNAs, scaRNAs, and snRNAs¹⁷—and participate in many processes, a few of which we will mention next.

Modify chromatin organization

tRFs can modify chromatin organization processes.⁴⁰

Regulate apoptosis and cell growth

Some tRFs are involved in initiating and inhibiting apoptosis, ²⁴ and cell viability after cell damage.¹⁴ tRFs have also been reported to participate in cell growth.¹⁴

Epigenetic inheritance

tRFs can also be directly involved in epigenetic inheritance in germ-line cells, affecting the metabolism of offspring. Key studies showing this were performed on low-protein-diet mice whose sperm showed markedly different tiRNAs concentrations.^19,41,42

Suppress movement of transposable elements

tRFs can suppress reverse transcription and retrotransposon mobility and can silence long terminal repeat (LTR) retrotransposons.¹⁷

Interact with miRNA-based mRNA silencing process

As research in tRFs progressed it became apparent that several miRNAs had been incorrectly annotated and are actually derived from tRNAs.¹⁵ Alternative gene sources of mature miRNAs are now known, so that miRNAs are not always generated from hairpin pri-miRNA (primary microRNA) structures. This poses no problem if cellular processes were designed in advance, but evolutionary attempts occurring through random mutations would create an endless potential for interference with mRNAs.

Some tRFs take part in globally controlling small RNA silencing through competitively combining with one of the key Ago⁴³ family proteins.¹⁷ Although these tRFs can deactivate mRNAs, unlike miRNAs, the target sites of mRNAs are distributed through all regions, including the 5’-UTR (untranslated region), CDS (coding sequence), and 3’-UTR.¹² Once again, the potential for interference is overwhelming, with the wrong mRNA targets being affected without an initial top-down design and implementation.

On-going research is uncovering ever more layers of regulation. A novel class of tRF-2s derived from tRNA(Asp), tRNA(Tyr), tRNA(Gly), and tRNA(Glu) was discovered recently, able to inhibit various oncogenic³³ mRNAs from stabilization thus helping to fight cancerous growth.¹⁷

Interact with the immune system

During the acute inflammatory stage, tRFs are significantly upregulated in blood circulation, implying a function in immune regulation. Investigations have demonstrated that tsRNAs can directly interact with a key receptor to activate immune responses in some lymphocytes.¹⁴

Discussion and conclusions

We began this series with the assumption that the collection of tRNA linker molecules seemed like the simplest subsystem of the genetic system since presumably only RNA is involved. This assumption was shown to be wrong in parts 1–3, since extensive biochemical processing is necessary for tRNAs to work, carried out by protein-constructed molecular machines. Now we see that the tRNAs and their degradation fragments play many additional important biological roles and communicate with various processes, some indirectly related to translations and others completely unrelated.

tRNAs are integrated into auto-regulation schemes such that translation is slowed down when substantial amounts of non-charged species are present as signals, and tRNAs are also integrated with cell-cycle progression. However, evolution’s lack of foresight prevents any planning of translation regulation before code translation even existed (especially when the same parts will serve multiple purposes). The examples presented involve precisely crafted interaction between tRNAs and protein complexes. Thus, our major conclusion is that the ensemble of tRNAs also includes several parts which are irreducibly complex. Evolution processes cannot begin with just RNA and somehow evolve an entire genetic system. This will be further explored in part 5.⁴