1. INTRODUCTION
When delving into the realms of codon optimality, we must first take a step back to observe our definitions and origins. At the most fundamental level, our genome is constructed from four basic building blocks, the nucleotides adenine (A), cytosine (C), guanine (G) and thymine (T). When a transcript (utilising a uridine (U) instead of a thymine) is being decoded by the ribosome, it is read and decoded in three-nucleotide units, called codons. When these two basic precepts are taken into account, this leads to an ultimate combination of 64 different codons (in the more mathematical sense, 4 to the power of 3 equals 64). 61 of these codons ultimately code for one of the 20 canonical amino acids, which get incorporated into the peptide chain as a mRNA transcript is translated, while 3 function as stop codons, terminating translation. As there is a certain discrepancy with these numbers (64 codons vs 20 amino acids a), there is a certain amount of degeneracy seen, with some amino acids being coded for by multiple codons (e.g., leucine).
But what is optimality, and why does it matter? Optimality typically refers to the efficiency/speed with which a specific codon is able to pair with the tRNA anticodon with which it matches. This is largely governed by the environment within which the codon inhabits. This variation can encompass different cell types within the same organism, different species, or completely different domains of life (bacteria/viruses). Optimality may even be altered in artificial in vitro translational systems. Highly expressed genes have higher codon optimality. It is believed that the higher optimality has only a small effect on the rate of protein synthesis because initiation, and not elongation, is the rate limiting step. The higher optimality, however, allows to reduce the number of ribosomes per mRNA, explaining the optimality of highly expressed genes. Nonetheless several recent reports points to existence of a relationship between codon optimality and mRNA stability as well as the rate of initiation. It has been suggested that the speed with which ribosome synthesize the protein could also affect protein folding, as certain domains require delays in order to be folded properly.
But what determines this variation across environments? An important factor is the tRNAs. Within the human genome, there are estimated to be between 400-500 different tRNA genes, a particularly large number considering the 61 sense codons. Debate does, however, exists as to the functionality of these tRNAs, with some theorised to play no role in mRNA decoding. It is even hypothesised that nearly half of tRNA genes have little to no expression at all. Nonetheless, a significant variety are present. While some tRNAs recognise unique codons, most recognise more than one codon, albeit having different affinity. For example, in humans, Lys-tRNAs with anticodon CUU are believed to recognise only AAG codon, Lys-tRNAs with anticodon UUU can recognise both AAG and AAA, this is because of a lack of requirement for base pairing with perfect Whatson-Crick geometry in the third position of a codon. Of note the affinity of tRNAs to codons is largely influenced by RNA modifications and structural features of tRNA that may be distant from the codon. The speed and accuracy of a specific codon decoding is largely determined by concentrations and affinities of tRNAs decoding that codon (so-called isodecoders).
2. MECHANISTIC ROLE
But what does this feature mean for a cell, and how can it take advantage of it for its benefit? In this section, we will take a look at some of the processes for which codon optimality has been associated with, with a particular focus on mechanisms by which they operate.
2.1 Ribosome Availability
One striking feature of codon optimality relates to the increased presence of optimal codons within genes that are particularly highly expressed. In one line of thinking, such optimality is linked to the idea whereby increasing elongation efficiency results in increase protein yield. While logical, the observation for most genes is that it is the rate of initiation that is by far the strongest determinator of protein yield (1). As such, optimality of codons would be relegated to a peripheral role in this respect. Instead, the optimality of codons in such highly expressed genes appears to be linked to preserve the availability of ribosomes in the cell. If we consider the opposite, wherein highly expressed genes display poor codon optimality, the highly expressed nature of these transcripts would result in a significant usage of available translational machinery, limiting the translation of other transcripts in the cell. Instead, with enhanced codon optimality on these transcripts, ribosomes can complete the translation of these transcripts within a shorter timeframe, subsequently releasing them back to free association in the cytosol to act upon other transcripts.
2.2 Protein Folding
Early on, protein structure was thought to be purely related to amino acid sequence. These residues, by virtue of inherit characteristics, such as electrical charge or degree of hydrophobicity, can determine comparative orientation, thus determining directionality and, ultimately, structure. However, research into the realm of codon optimality threw up a particular quirk. Translationally optimal codons have a greater tendency to associate with structural sensitive sites (2), that is, sites where mutations are much more likely to result in alternate protein structure, often to the detriment of function. Further research determined that much of these impacts are likely conveyed through the alteration of translation elongation speed associated with optimal and non-optimal codons (4). Although codons may be referred to as non-optimal in this context, this does not necessarily mean detrimental, as specific delays in elongation may allow the time for a protein to fold into its correct state. However, a corresponding theory exists, wherein the protein folding within the exit tunnel slows down ribosome movement. In such cases, optimal codons may be unlikely to have a strong impact on protein synthesis, and thus, have no evolutionary pressure selecting for them, leading to the appearance of selected non-optimality.
2.3 mRNA Decay
While the importance of codon optimality in protein folding continues to be debated, a larger body of work has delved into the role optimality plays in the areas of mRNA stability and decay. When a ribosome stalls too long on a codon without any corresponding tRNA incorporation, this is often read as a signal to initiate such decay. As such, the presence of a comparatively significant amount of rare or inefficient codons within an individual transcript can contribute to mRNA stability, which is often measured through the half-life of a specific mRNA transcript. This is a metric that is multi-factorial, which also takes into account poly A tail length and mRNA modifications. However, codon optimality also plays a strong role in this instance, and especially so within certain biological situations, such as the maternal-to-zygotic transition (5). In 2015, Presnyak et al. provided evidence suggesting that codon optimality is a major factor determining mRNA stability. Through their elegant substitution of optimal codons with their synonymous non-optimal counterparts, they drastically reduced mRNA stability in the process (6). Later research elucidated the trans-acting factors that contribute to codon optimality mediated decay. It was eventually determined that the Ccr4-Not complex, taking advantage of a ribosomal conformation change caused by an empty A-site (itself due to a non-optimal codon), engages at the E-site (7), thereby recruiting the DEAD-Box protein Dhh1 (8), previously known to couple codon optimality to mRNA decay by recruiting mRNA decapping and deadenylation factors.
3. FUNCTIONAL ROLE
While we have a clearer idea of how the optimality of codons can impact on such processes such as protein folding and mRNA stability, what are some real-world examples of these processes in action, or perhaps more specifically, what instances are there of when these processes go awry? Much of these more “hands-on” discoveries have been made in the last 2-3 years, and so wide-ranging research is comparatively lacking. But what is there makes for interesting reading.
3.1 Neuronal Function
Take Fragile X syndrome, for example. This condition is caused by a loss of production of the Fragile X Mental Retardation Protein, or FMRP, and leads to intellectual disability. While this protein is typically associated with translational repression, recent evidence has suggested a further role in mRNA stabilisation. In an oddly specific quirk, it was found that that transcripts with a typically higher codon optimality score are preferentially degraded in its absence, which would suggest it can help to stabilise already stable mRNAs. Meanwhile, transcripts that score lower with regard to their codon optimality remain unaffected, together indicating that this protein acts as an enhancer of codon optimality-associated mRNA stability (9).
3.2 Cellular Differentiation
Staying with the brain, but switching out of neurons, we arrive in the world of oligodendrocytes, the producers of the myelin sheath within the CNS. These are curious cells, in that they go through a number of stages of differentiation, from the proliferative oligodendrocyte precursor cells, to premyelinating oligodendrocytes, to mature myelinating oligodendrocytes, with each stage requiring particular proteins. Relevantly, a significant cohort of patients with leukodystrophies, which are essentially disorders of myelin production, have known dysregulation of tRNA and its associated machinery. Picking up on this quirk, a team from John Hopkins University discovered markedly reduced expression of the enzymes required for the mcm5U/mcm5s2U modification in mature oligodendrocytes when compared to their precursors. Similarly, specific tRNAs (namely, Phe-GAA and Lys-UUU isodecoders) are hypomodified in mature oligodendrocytes, while the reverse is true in their precursors. This hypomodification alters their optimality for their corresponding codons, resulting in those codons now becoming mRNA destabilising factors in mature oligodendrocytes. Together, this indicates the optimality of the tRNA itself changes concomitant with cellular differentiation. Even more importantly, there is a marked reduction of the now corresponding non-optimal codons within genes essential for the function of the mature oligodendrocytes, namely in genes related to components of the membrane, i.e., in this case, myelin production (10).
3.3 Viral Infection
An even curiouser case arises following infection from the Chikungunya virus (CHIKV). This is positive-sense RNA virus, endemic to tropic regions where mosquitoes, its vector, are present. Such RNA viruses tend to have an enrichment of rare codons within their genetic code (codons typically ending in A/U, whereas more common codons within the human genome have a bias towards ending in G/C). As such, they would naturally encounter some difficulty in the realm of codon optimality entering a human cell, with its tRNA pools directed towards the translation of human genome-contained codons. Yet here, in a first, this virus was specifically seen to target a particular gene, namely KIAA1456. Here, while its transcriptional levels were decreased akin to the majority of mRNAs, its translational activation was increased 12-fold, resulting in an overall increase in protein production. Intriguingly, the function of this gene is related to the modification of uridine in the wobble position of tRNA anticodons, and knockdown of this gene in CHIKV-infected cells dramatically reduced CHIKV protein production. Upregulation improved translational efficiency of viral transcripts over host transcripts. Later research by this group uncovered that viruses enriched for particular codons (such as GAA, AAA, CAA, AGA, and CGA, all codons which have a corresponding U in the wobble position of their anticodon) also demonstrate upregulation of KIAA1456 (11), suggesting a commonly preserved mechanism for influencing codon optimality in such species, favouring their propagation.
4. CONCLUSION
While the evolutionary theory of codon optimality is not new, comparatively novel translational techniques, such as ribosome profiling, have recently been able to give more specific functional insights into its consequences. With such novelty, this area has an especially high potential for future discoveries in more distinct situations. It has already been shown that there is a degree of importance of this feature in differentiation of oligodendrocytes. The idea that such a phenomenon is limited to a singular cell type alone is highly unlikely. Furthermore, the evidence of Chikungunya virus on optimality upon infection is new, leaving the door open for further analysis of the plethora of other pathogens of interest. It also may open avenues for some degree of optimisation of industrial protein production, given many human proteins are synthesised in non-human vectors, such as Chinese hamster ovary cells or E. coli. Time will tell.
References
1. Gebauer F, Hentze MW. Molecular mechanisms of translational control. Nature Reviews Molecular Cell Biology. 2004;5(10):827-35.
2. Zhou T, Weems M, Wilke CO. Translationally optimal codons associate with structurally sensitive sites in proteins. Mol Biol Evol. 2009;26(7):1571-80.
3. Pechmann S, Frydman J. Evolutionary conservation of codon optimality reveals hidden signatures of cotranslational folding. Nat Struct Mol Biol. 2013;20(2):237-43.
4. Zhao F, Yu CH, Liu Y. Codon usage regulates protein structure and function by affecting translation elongation speed in Drosophila cells. Nucleic Acids Res. 2017;45(14):8484-92.
5. Bazzini AA, Del Viso F, Moreno-Mateos MA, Johnstone TG, Vejnar CE, Qin Y, et al. Codon identity regulates mRNA stability and translation efficiency during the maternal-to-zygotic transition. Embo j. 2016;35(19):2087-103.
6. Presnyak V, Alhusaini N, Chen YH, Martin S, Morris N, Kline N, et al. Codon optimality is a major determinant of mRNA stability. Cell. 2015;160(6):1111-24.
7. Buschauer R, Matsuo Y, Sugiyama T, Chen YH, Alhusaini N, Sweet T, et al. The Ccr4-Not complex monitors the translating ribosome for codon optimality. Science. 2020;368(6488).
8. Radhakrishnan A, Chen YH, Martin S, Alhusaini N, Green R, Coller J. The DEAD-Box Protein Dhh1p Couples mRNA Decay and Translation by Monitoring Codon Optimality. Cell. 2016;167(1):122-32.e9.
9. Shu H, Donnard E, Liu B, Jung S, Wang R, Richter JD. FMRP links optimal codons to mRNA stability in neurons. Proc Natl Acad Sci U S A. 2020;117(48):30400-11.
10. Martin S, Allan KC, Pinkard O, Sweet T, Tesar PJ, Coller J. Oligodendrocyte differentiation alters tRNA modifications and codon optimality-mediated mRNA decay. Nature Communications. 2022;13(1):5003.
11. Jungfleisch J, Böttcher R, Talló-Parra M, Pérez-Vilaró G, Merits A, Novoa EM, et al. CHIKV infection reprograms codon optimality to favor viral RNA translation by altering the tRNA epitranscriptome. Nature Communications. 2022;13(1):4725.