jim mcnamara said:
If our understanding of Archaea is correct - they evolved before modern cells, likely at high temperatures.
The three major cell types evolved way before their modern (extant) members of course. But it is true that of them eukaryotes split last. [See "Evolution of the ribosome at atomic resolution", Petrov et al, PNAS,
http://www.pnas.org/content/111/28/10251.full.pdf ; or the topologically equivalent usual 16S tree.]
Did Archaea evolve at high temperatures? Perhaps, but not necessarily.* It seems to be consensus that the universal common ancestry lineage bottlenecked through, or more likely rooted in, thermophilic cells.
Valentin's energy theory on Archaea, which is nicely consistent with Lane's energy theory on eukaryotes as high energy density specialists (due to coopting their
mitochondria, likely and ironically originally ATP stealing parasites), is that Archaea evolved to be low energy density specialists. Perhaps it was the reason for the split, a population isolated by utilizing a scarce niche. That would predict their specialized low leakage membranes, makes chemiosmosis more efficient, a trait that can be coopted by archaean thermophilic extremophiles. As I understand it, but I can be mistaken, the high temperature limits are set by metabolic stress rather than inability to evolve functional cells. At sufficiently high temperatures cells can't grow so can't procreate, and at higher temperatures they starve to death.
*Depends on the dating of the split. If one peruses Timetree, one finds that Archaea and Bacteria split ~ 4 billion years ago.
That dating coincide with the late bombardment, which was a stochastic high temperature environment. According to models of Abramov et al, mesophiles would have survived too. But they could easily been among the many extinct lineages as thermophile lineages would be much more populous and so survive much more frequently. On the other hand, seeing Valentin's model, it looks more like a coincidence than a correlation, as scarce niches would have been abounding for all sorts of reasons.
But it is a molecular clock date, and to this layman those clocks seem to prefer to push their earliest splits as far back as they can. And there is very little other evidence that would test such deep dating.
The dating look reasonable, earliest oceans and likely continental crust has been evidenced in a zircon at ~ 4.40 billion years ago, and using Abramov et al models life could emerge at that time as the late veneer bombardment rapidly quenched after Moon formation ~4.47 billion years ago (recently rather securely dated with multiple methods, and the bombardment quenching too). And the first splits (here between Bacteria and Archaea) that we have surviving extant evidence of despite lineage extinction could happen a few hundred million years later. But is it valid, with all the uncertainty involved?
jim mcnamara said:
what were the proteins like in cyanobacteria and relatives before the Great Oxygenation Event?
Much the same I think, but evolved for anaerobic uses. Luckily modern OO protection stress enzymes seem to root in NO protection, and NOx would be produced at significant levels by volcanism in a CO2 atmosphere with ammonia among volcanic gases. So life survived its own global poisoning event.
Wouldn't we expect lipids to be more changed by evolution? Fish meat has the unfortunate trait to taste 'fishy' after prolonged increased oxygen environment, which I understand as an oxidation effect of fish oils. Et cetera.