Swapping Iron For Magnesium in the Ribosome

A daring experiment corroborates translational system’s place at earliest foundations of life on Earth

This experiment had a good chance of crashing. Instead, it delivered whopping evidence to corroborate the earliest evolution of the translational system, the mechanisms which make life out of our genes. The study swapped out all its magnesium, tabula rasa, and showed that the system, centering on the ribosome, would have thrived basically as it is today 4 billion years ago at the earliest foundations of life on Earth.

In the system inside cells that translates genetic code into life, he replaced about 1,000 essential linchpins with primitive substitutes to see if the translational system would survive and function. It seemed impossible, yet it worked swimmingly, and Bray had compelling evidence that the great builder of proteins was active in the harsh conditions in which it evolved 4 billion years ago.

The experiment’s success reaffirmed the translational system’s place at the earliest foundations of life on Earth.

Every living thing exists because the translational system receives messages from DNA delivered to it by RNA and translates the messages into proteins. The system centers on a cellular machine called the ribosome, which is made of multiple large molecules of RNA and protein and is ubiquitous in life as we know it.


This is a very interesting study. There is a (remote, admittedly) chance that it will be relevant to @pnelson’s essay he was working for later this evening, or maybe parts 2 and 3 of his discussion of the ideas in the Signal + Noise thread. (Just a playful prod, @pnelson. I am sure you have plenty on your plate.)

1 Like

I find the subtitle comical:

So audacious was Marcus Bray’s experiment that even he feared it would fail.

Just about all of our experiments we “fear they will fail.” What in the world does this tell us about the audacity of the experiment? Hehe.

1 Like

@John_Harshman, take a look at this:

Eat your magnesium

Those linchpins Bray yanked out and replaced were metal ions (atoms with charges, in this case positive).

In today’s ribosome, and in the whole translational system, they are magnesium ions, and Bray’s experiment replaced them all with iron ions and manganese ions, which were overabundant on primordial Earth. Williams and Jennifer Glass, the principal investigators in the new study, also had their doubts this was doable.

“I thought, ‘It’s not going to work, but we might as well try the moonshot’,” said Williams who has led similar work before but on simpler molecules. “The fact that swapping out all the magnesium in the translational system actually worked was mind-boggling.”

That’s because in living systems today, magnesium helps shape ribosomes by holding them together. Magnesium is also needed for some 20 additional enzymes of the translational system. It’s one reason why dietary magnesium (Mg) is so important.

“The number of different things magnesium does in the ribosome and in the translational system is just enormous,” said Williams. “There are so many types of catalytic activities in translation, and magnesium is involved in almost all of them.”

As I recall, one of the examples from @pnelson’s test (Nelson's Test of Common Descent Comprehension) was about the type of ion in a protein. I remember responding to him that it is fairly easy to swap different metal ions in and out of a protein, so I expected there be a lot of homoplasy. Do you remember that example?

Of course, it is far more surprising that this worked for the ribosome, swapping all of them out at once. I don’t know if I would have predicted it. And on reflection is it really that surprising?

1 Like

You are thinking of Yan Zhang and Vadim N. Gladyshev. 2010. General Trends in Trace Element Utilization Revealed by Comparative Genomic Analyses of Co, Cu, Mo, Ni, and Se. J. Biol. Chem. 285: 3393–3405. But it isn’t about use within a gene; it’s much fuzzier than that and much more dubious as a question of homology. It’s about the percentage of species within particular groups that use particular ions in any capacity at all. Nelson stripped out all the taxa that lack one or more of the ions, showed you the few taxa that use them all, and hopes you will decide that those ions are all universal. Very cheesy maneuver.

1 Like

Isn’t this a major result in the study of the origin of life on Earth?

I’m not sure if that is the case. I’m not sure what this buys origin of life research.

This allows for the evolution of the ribosome using cations apart from magnesium, many of which may have been much more abundant in the primordial environment.

When my mind wanders, I sometimes ponder how difficult it would be to identify riboorganisms - cells that use only RNA and not DNA. This report adds an item to this musing (not so idle), the search for an organism that has ribosomes that use other than magnesium. How might one do this? Just wondering …


Please explain more.

It is hard to argue against an argument that isn’t presented. I’d have to imagine how it helps the origin of life case so to argue against it. Maybe I’m just ignorant, but I don’t see anything particularly important. All this is shows is that extant ribosomes are able to use different metal ions. I’m unclear why that is so important.

Someone has to make the argument for why it is important for me to assess it.

Probably no more difficult than identifying RNA viruses or doing RNAseq. I suppose a riboorganism would first appear to use like an RNA virus, till we were stunned by its ability to live independently. That might get someone a Nobel Prize, right?


Hi @swamidass, @Art and @Patrick,

I can’t lay claim to any expertise in this area, but it strikes me that the latest finding does seem to tell somewhat in favor of abiogenesis, even if the pathway continues to elude us.

Here’s what I mean. Suppose that the Creator designed the first living cell ex nihilo, including its basic biochemistry, which would be transmitted to all of its descendants. Presumably, the Creator would build designs that would last, at the biochemical level. Changing the structure of the key molecules of life later on sounds like changing horses in midstream.

So if an experiment suggests that a molecule that’s currently used by all living things once used iron and manganese instead of magnesium, the conclusion I would draw is that little, if anything, about the design of life appears to be invariant over time. But if the molecules of life are constantly changing, then one has to ask: what, exactly, did the Creator design?


Ah I see what you are saying. Don’t we already know that most of these things are not invariant?

For example the “universal” genetic code is more of a “rule with exceptions” than a “law.” We know that it can and does evolve into none standard codes. Of course, that isn’t what large complex organisms will do, but microbial systems and and do change.

Ribosomes, polymerases, and other core proteins also change over time too. Some of these changes are functionally important, others are not.

1 Like

Very cheesy, maybe, but also highly illuminating. (Can cheese illuminate?)

Here’s why. Two versions (x and y) of the phylogenetic exercises were prepared for the students, and distributed randomly. In the y version, which I didn’t send to John Harshman, only the taxa that used all the trace elements were masked – exactly the opposite of the x version, which John saw. Students who received the y version had the rest of the character distribution unmasked, with widely varying trace element usage among the lineages.

My expectation going into the exercise was that students who received the y version would not predict universal trace element usage, given the information they had. And that’s exactly what happened.

I realize this looks rigged and cheesy, but a serious and statistically rigorous version of the same kind of test could be conducted with evolutionary biologists. Systematic biologists, however, would probably refuse to participate, because they know about confounding patterns such as non-orthologous gene displacement, trait reversal, rampant homoplasy, long-branch attraction, parallelism, and the like.

1 Like

We already did this here @pnelson, avoiding the selection bias in your experiment: A Test of Common Descent vs. Common Function. When do you want to see how phylogenomics systematically tests this?

Better yet, it avoids htis problem:

Instead, we would use the brute facts to guide a test of the hypothesis.

1 Like

Exactly. It looks rigged and cheesy because it is rigged and cheesy. You’re choosing data to force the result you want. You could do a rigorous version by omitting characters from taxa at random. But what would that show? That there is such a thing as homoplasy? Such a thing as reversal? You don’t need an elaborate test to show that. What do you think it would show?

1 Like

OK, I’m going to go out on a limb here (and potentially expose my ignorance).

I don’t see how this is a remarkable result. As a non-biological chemist, I’m not surprised that swapping out Mg for Mn or Fe would still leave you with a functional molecule. Admittedly the sizes are a bit different but they have the same charge (+2) as the Mg, which I would think would be the major contributor towards functionality. I’d be more surprised if they did it with Al or K. I think I would also be surprised maybe if swapping out improved the functionality.

I even found the following on wikipedia, so it must be true: "Mn^{2+} often competes with Mg^{2+} in biological systems."

I do see how the result could be useful for abiogensis studies though since you could show that ribosome could be functional in a different environment, more similar to what early earth would have looked like.


Interesting comment. I was surprised by this report, but I can see how, from a purely chemical perspective, it isn’t all that remarkable. I guess part of the surprise comes from the facts that Mg functions to neutralize negative charges in the rRNA backbone, but also plays other roles in the RNA structure and RNA-protein interaction network. These latter, one might imagine, would be more sensitive to the sizes of the ions involved.