We identified protein domains in LUCA (Last Universal Common Ancestor). Their distinctive amino acid usage reveals the order amino acids were added to the genetic code, based mostly on size. Older proteins hint at earlier alternative codes. 1/15 @seekingluca.bsky.social www.pnas.org/doi/10.1073/... - ThreadSky

joannamasel.bsky.social • 87 days ago

We identified protein domains in LUCA (Last Universal Common Ancestor). Their distinctive amino acid usage reveals the order amino acids were added to the genetic code, based mostly on size. Older proteins hint at earlier alternative codes. 1/15 @seekingluca.bsky.social https://www.pnas.org/doi/10.1073/pnas.2410311121

Comments

petrovadmitri.bsky.social•87 days ago

Truly fascinating stuff Joanna

chasewnelson.bsky.social•83 days ago

Just amazing! Bravo!

montsepa.bsky.social•87 days ago

📌

ericfauman.bsky.social•87 days ago

Very interesting. Can you say anything from this about the evolution of the genetic code, eg the hypothesis that an early codon map was based on 2bp instead of 3
I see most of the codons where the 3rd position is superfluous belong to early amino acids (VGATP) only two for late (LR).

joannamasel.bsky.social•87 days ago

We provide key constraining info for such speculations, but don't speculate in the paper. Re one hypothesis that originally the second position carried most/all information, I did notice that there are ancient amino acids represented in each of the four cases: Gly, His, Ile/Met/Val, and Thr/Ala.

ricardoazevedo.bsky.social•87 days ago

Very cool!

pharmblog.bsky.social•82 days ago

📌

tshandy.bsky.social•87 days ago

Neat-o!

genomesevolve.bsky.social•87 days ago

Congratulations Joanna. Downloaded, will take a look.

cslamo.bsky.social•83 days ago

📌

jung-gt.bsky.social•87 days ago

📌

jukebarosh.bsky.social•82 days ago

I love paper. The logic is beautiful. M, C, W & H make so much sense in context of 1-C, I was surprised to learn they were not already considered to be among the earliest aa's

jukebarosh.bsky.social•82 days ago

If glutamine was one of the last aa, do you have a guess what the amide donor was (eg for purine biosynthesis)? I know NH3 can sub, but has to be at very high concentrations

seekingluca.bsky.social•82 days ago

glutamine (and its derivatives) could still be present in the cell and used as an amide donor. Our results show that its incorporation into the code was late but that does not necessarily mean it was absent.

jukebarosh.bsky.social•82 days ago

Interesting to consider - why would the amino acids that also pull most from chorismate (and therefore 1C pool) be the 1st to be incorporated into proteins given SAM demand? And met the literal 1st - at N-term? Seems like early life benefited from controlling the flow

joannamasel.bsky.social•81 days ago

You mean F, Y, and W as those that pull from chorismate? They are all late to our code, and W is depleted pre-LUCA, but they are less depleted in the most ancient domains diversified pre-LUCA than single-copy LUCA. The unambiguously early ones are GVIMHTA.

joannamasel.bsky.social•81 days ago

A caveat is that an early amino acid that subsequently gained codons would show up as late. The assumption is that new aa were added by taking codons away from old aa, not that old aa took codons away from other old aa.

joannamasel.bsky.social•87 days ago

Protein domains are the modules of evolution. A domain can be much older than the protein it is currently in. We dated each “clan” of evolutionarily related Pfam domains, including relatives that are highly diverged at the sequence level. 2/15

joannamasel.bsky.social•87 days ago

Not all protein clans go back to the primordial soup. We compared the oldest clans that had diversified prior to LUCA vs. clans descended from just one LUCA copy vs. younger but still-ancient clans not in LUCA but appearing before the last archaeal or bacterial common ancestor. 3/15

joannamasel.bsky.social•87 days ago

When an amino acid is enriched in LUCA relative to ancient but post-LUCA, we deem it early. Trifonov previously classified cysteine and methionine as late because of abiotic unavailability – but of course they weren’t produced in the Miller-Urey experiment, because it added no sulfur! 4/15

joannamasel.bsky.social•87 days ago

Amino acid size strongly predicts the order of recruitment. (Trifonov's statistical consensus among 40-60 metrics does worse.) Partial outliers are that methionine, histidine, and isoleucine are earlier than expected, and glutamine is later. 5/15

joannamasel.bsky.social•87 days ago

Histidine was presumed late because it isn’t abiotically available. But the code evolved in a cell capable of ribosomes, not directly from the prebiotic soup. Histidine resembles a purine, and could have been biotically available. https://journals.plos.org/plosone/article/figure?id=10.1371/journal.pone.0196349.g001 6/15

joannamasel.bsky.social•87 days ago

Methionine would have had high biotic availability given the importance of S-adenosylmethionine (SAM) to early life, e.g. for methylation @paolalaurino.bsky.social https://doi.org/10.1371/journal.pbio.1002396 7/15

Comments

Posting Rules

Reply