A codon is a triplet of three nucleotides in DNA. Genes are read in these triplet codons, each one standing for an amino acid or a “punctuation” mark as the gene gets translated (61 of the 64 possible triplets actually code for amino acids; the others work as “start” and “stop” codons). This much we’ve known since the 1960s. Now, however, two scientists from the University of Utah want to complicate matters further.
An article at Phys.org explains:
The so-called central dogma of molecular biology states the process for turning genetic information into proteins that cells can use. “DNA makes RNA,” the dogma says, “and RNA makes protein.” Each protein is made of a series of amino acids, and each amino acid is coded for by sets of “triplets,” which are sets of three informational DNA units, in the genetic code.
University of Utah biologists now suggest that connecting amino acids to make proteins in ribosomes, the cell’s protein factories, may in fact be influenced by sets of three triplets – a “triplet of triplets” that provide crucial context for the ribosome. [Emphasis added.]
It sounds like a wild idea, but it was just published in the Proceedings of the National Academy of Sciences. What could be the impact of this “Case for the genetic code as a triplet of triplets”?
Neo-Darwinian evolution is supposed to work by mutating DNA codons, either making them “synonymous” with the prior codon (i.e., yielding the same amino acid), or “non-synonymous” (i.e., putting a different amino acid in its place, potentially affecting the resulting protein). If codons could be treated as independent entities acted on by natural selection, Darwinians at least understood the challenge before them. If these researchers are correct, the stakes just skyrocketed.
We had already learned that synonymous codons are not truly identical, even if the same amino acid gets into the final protein. Like English synonyms with different shades of meaning, synonymous codons can affect the genetic language. They can do this by affecting the rate of transcription, the modifications needed on the transfer RNA (tRNA), or the rate of translation (see “A Genetic Snooze Button” here). That finding provided a reason for the redundancy in the genetic code, where one amino acid can be represented by one to six different triplet codons. This new paper agrees:
Synonymous changes will lead to differences in translation rates that, especially when different tRNAs are used, have different binding efficiencies, abundances, and charging rates and result in differential mRNA stabilities. In addition, the same tRNA reads different codons with different efficiencies, as was determined in an in vivo translational speedometer assay system.
Finding this extra functionality in this redundancy in the genetic code (called “degeneracy”) was interesting enough, but now evolutionists are faced with a new challenge. The authors of the PNAS paper, Hughes and Chevance, describe what drove them to examine the context for each triplet codon. They were playing with the genes for a component of the bacterial flagellum named FlgM when they noticed something interesting:
Changing the codon on one side of the defective codon resulted in a 10-fold increase in FlgM protein activity. Changing the codon on the other side resulted in a 20-fold decrease. And the two changes together produced a 35-fold increase. “We realized that these two codons, although separated by a codon, were talking to each other,” Hughes says. “The effective code might be a triplet of triplets.”
To understand the genetic code, if this is true, geneticists are going to have to understand the context of each triplet. Each codon could be affected by its flanking codons, creating vastly different outcomes in terms of gene activity.
We conclude that codon recognition is initiated by codon–anticodon hydrogen bonding between the first and second bases of the translated codon followed by sensing the correct fit at the wobble base and base-stacking interactions contributed by the preceding two codons and bound tRNAs.
The “triplet of triplets” problem helps explain why you can’t easily get the same expression pattern by substituting a plant or animal protein in a bacterium, a lab procedure called heterologous expression. One doesn’t just tinker with a particular codon and expect to get the same result in a different organism that has a different expression context. The particular codon used affects downstream factors, including tRNA modifications, which the authors say are extensive in every organism.
Modification of tRNA species in E. coli also has been shown to vary with the growth phase of the cell. Specific codon-context effects could represent translation domains of life based on tRNA modifications.
We begin to see what this could do to neo-Darwinian theory. The authors don’t go into detail, but they issue an ominous warning in the paper’s final paragraph:
The difficulty for natural selection would be in finding codon optimization for a given gene. If the speed through a codon is dependent on the 5′ and 3′ flanking codons, and the flanking codons are dependent on their 5′ and 3′ flanking codons, then selection pressure on a single codon is exerted over five successive codons, which represent 615 or 844,596,301 codon combinations. If modified tRNAs interact with bases in a codon context-dependent manner that differs among species depending on differences in tRNA modifications, ribosome sequences, and ribosomal and translation factor proteins, it is easy to understand why many genes are poorly expressed in heterologous expression systems in which codon use is the primary factor in the design of coding sequences for foreign protein expression. The potential impact of differences in tRNA modifications represents a significant challenge in designing genes for maximal expression whether by natural selection or in the laboratory.
The paragraph on the “significance” of the hypothesis states the challenge succinctly:
“Data presented here support a model in which the evolutionary selection pressure on a single codon is over five successive codons, including synonymous codons.
The more that natural selection has to “think” about (if you’ll pardon the expression), the less able it will be to get things right. More accurately, it’s going to take a lot more of what David Berlinski calls “sheer dumb luck” to find a beneficial change. If there are 845,596,301 codon combinations to worry about, it’s like having to get many more numbers right in Powerball than you thought when you bought your lottery ticket. This is what they imply:
The tRNA modifications vary throughout the three kingdoms of life and could affect codon–anticodon pairing. The differences in tRNA modifications could account for differences in synonymous codon biases and for the effects of codon context (the ability to translate specific triplet bases relative to specific neighboring codons) on translation among different species. Here, using in vivo genetic systems of Salmonella, we demonstrate that the translation of a specific codon depends on the nature of the codons flanking both the 5′ and 3′ sides of the translated codon, thus generating higher-order genetic codes for proteins that can include codon pairs and codon triplets.
It will be interesting to see how this hypothesis plays out. One immediate impact will be on research concerning genetic diseases. The triplet-of-triplets coding scheme might explain why mouse models of disease treatments don’t always translate well into human trials: the context is different.
“Higher-order genetic codes” — what a concept! Actually, ID advocates like Jonathan Wells have been talking about this for years.
Image credit: lisichik via Pixabay.