A brand-new paper from Nature Ecology & Evolution used a clever technique to estimate the age of “LUCA”,. the “last universal common ancestor” of all living things. What that means is LUCA is the last creature whose descendants include every species alive: the ancestor of all of us. And it dates LUCA to about 4.2 billion years ago! That is far older than people thought. Previous estimates were in the 3.5-3.8 billion-year range, after the famous “Late Heavy Bombardment” (LHB), during which the Earth was continually battered with asteroids and comets. It was assumed that nothing alive on Earth could have survived those impacts. But if the authors are right, LUCA’s ancestors did survive this, for 4.2 billion years is probably a big underestimate of of when life on earth began.
The earliest generally accepted fossil evidence for life is about 3.7 billion years, which is based on isotopes that, scientists think, could have been produced only by living creatures. But the earliest genuine fossilized organisms occur a bit later than that: fossilized blue-green algae (“stromatolites”), whose fossils go back 3.5 billion years ago.
The new paper by Moody et al., which has an accompanying research brief (click screenshots below to access, or find the pdf here) pushes the age of LUCA back to 4.2 billion years ago, which actually precedes the LHB. And the new LUCA date comes soon after the Earth actually formed (about 4.54 billion years ago) and after the Moon was created, probably by a huge, Mars-size planet striking Earth and throwing off debris that consolidated to create our Moon. (That occurred soon after the Earth formed.) Surely no life could have survived that collision, so if the authors are right, it took only about 0.3 billion years, or 300 million years after the Earth was formed, before life existed.
But LUCA wasn’t the first life on Earth: it is simply the bacteria-like species of organism that gave rise to all living creatures. Surely life originated before that, and the new paper suggests that the 4.2 billion year old (byo) LUCA was only one of a number of life forms existing back then, with the rest going extinct without leaving descendants. The authors think this because LUCA probably needed complex carbon compounds to live, and is also likely to have provided niches for other creatures. That means that life itself began well before LUCA, especially because, based on its genome, the authors conclude that LUCA was quite complex— about as complex as modern bacteria. Surely it would take millions of years of evolution to get to the point where a LUCA-like creature could have existed. See below for the diagram of what LUCA was like.
The main lesson from the paper is that life began very, very soon after the Earth had cooled off and the dust had settled from the LHB and carving out of the Moon. If that’s the case, then perhaps life on other planets could evolve more easily than we thought.
But on to the paper. If you want the whole megillah, click on the first link, while the second gives a two-page précis. It’s a very complicated and long paper, so give me kudos for reading it twice to distill it here. But I can’t claim to have understood everything, as the analyses of the data, or even the methodology, is quite arcane and sophisticated.
A two -age summary from the same journal:
Why do we think that all life descended from a single species rather than having multiple origins? Because all living creatures have some similarities that probably reflect the workings of chance: whatever mutations happened to give rise to our ancestor. The paper explains:
The common ancestry of all extant cellular life is evidenced by the universal genetic code, machinery for protein synthesis, shared chirality of the almost-universal set of 20 amino acids [JAC: all amino acids used in modern creatures are the L rather than the D form] and use of ATP as a common energy currency. The last universal common ancestor (LUCA) is the node on the tree of life from which the fundamental prokaryotic domains (Archaea and Bacteria) diverge. As such, our understanding of LUCA impacts our understanding of the early evolution of life on Earth.
The way scientists usually estimate LUCA is using molecular dating based on DNA divergence among living organisms. Because there is a “molecular clock”, with the DNA changing roughly in a linear fashion with time, you can back-calculate from living creatures to estimate when their DNA sequences would have converged on a single sequence, which would be the DNA sequence of LUCA. But there are formidable problems with this, making DNA-based estimates contentious. But the authors found a way around this.
What they did is to estimate divergence times of all living creatures (for practicality, they used bacteria [prokaryotes] and Archaea, bacteria-like organisms that form their own kingdom) using DUPLICATED GENES. These are genes that, tracing the sequences of living organisms back, had already been duplicated in LUCA. As you may know, genes often get duplicated during cell division or (in sexual organisms) meiosis, so a single gene can now occur in two copies. Those two copies will initially be identical, but then, being genetically independent, will begin to diverge via mutation and then selection or drift. (Examples of duplicated genes are are different forms of globins in humans, two of which, alpha and beta, produce products that combine to make adult hemoglobin. But many, many genes have duplicated over the history of life.)
A gene that is duplicated (based on sequence similarity) in LUCA must have been present in the ancestor of LUCA, and have duplicated before LUCA existed. Thus an estimate of the age of a duplicated gene in LUCA gives us a lower-bound on the age of LUCA itself. And since some genes are already duplicated in LUCA, we can use them, combined with a molecular clock (and other statistics) to estimate how long it took for each copy to give rise to the diversity of DNA-sequences in descendant copies in modern microbes. The advantage comes because we have two estimated DNA sequences in LUCA that began identically but then diverged over evolutionary time. This gives us two chances to estimate the age of the creature. Using other methods, we can estimate how many genes there were in LUCA, the size of its genome, and what kind of genes it had. The latter can then give us an idea of what kind of creature it was and how it lived.
Here are the results, in short:
a.) LUCA lived about 4.2 billion years ago. Here’s the reconstructed phylogeny (note that there are two estimates of its age since they use two copies of each of the five genes they chose for age estimation). Click to enlarge. On the right are all the kingdoms of living organisms, traced back to LUCA. The use of two gene copies give similar estimates, about 4.2 billion years ago. I’ve circled the two LUCA estimates, which work out to a similar age (see age scale at top for divergence times):
b.) LUCA had a big genome and many genes. The authors estimate that LUCA’s genome had 2.75 million DNA base pairs, capable of making 2,657 proteins (an underestimate of gene number). That is a big and complex organism, comparable to existing bacteria. (Modern E. coli produce about 4288 proteins from 4.6 million base pairs.) This complexity shows that even LUCA was preceded by a long period of evolution.
c.) LUCA was probably an anaerobic and autotrophic creature, which means that it didn’t need oxygen to grow and flourish, and also that it produced its own “food”, getting energy from substances like hydrogen and carbon dioxide. The authors suggest two places where such a creature could have lived: in warm hydrothermal vents in the ocean, or on the ocean surface, where it would have ample access to the gases that constitute its food. There was no evidence that the organism was photosynthestic, as it lacked genes involved in modern photosynthesis.
Here’s a sketchy diagram of what kind of genes LUCA had (note the “immune” system, based on CRISPR-like genes that are used to destroy viruses. LUCA probably had a virus problem, too! Figure b) show us how LUCA fit into the tree of life:
d.) LUCA was part of a community of other organisms. It’s inconceivable that LUCA. which was a sophisticated organism, could live without a source of organic compounds (like amino acids) to use for constructing its body (remember, these organic compounds were not a “food,” but a construction material). Further, LUCA would itself provide organic compounds that would create niches for other species. (It’s likely that viruses, which aren’t good candidates for a LUCA-like creature, already existed.) The phylogeny in figure (b) just above shows how LUCA would fit into the tree of life, giving rise to all modern creatures via speciation events, but would itself also be part of an earlier family tree, all of whose members save LUCA went extinct without leaving descendants.
These are the four big conclusions of the paper, with the most interesting to me being how short the time was after Earth’s formation for complex life to have evolved. And the age of LUCA, remember, is an UNDERESTIMATE of how long it took complex life to evolve after the Earth’s conditions were suitable for such evolution.
I’ll end with the authors’ own conclusions, which are lucid enough for the layperson (bolding is mine)
Conclusions:
By treating gene presence probabilistically, our reconstruction maps many more genes (2,657) to LUCA than previous analyses and results in an estimate of LUCA’s genome size (2.75 Mb) that is within the range of modern prokaryotes. The result is a picture of a cellular organism that was prokaryote grade rather than progenotic [JAC: not having the characteristic of a prokaryote, which LUCA did] and that probably existed as a component of an ecosystem, using the WLP [JAC: the Wood-Ljungdahl pathway for producing energy, based on hydrogen and carbon dioxide] for acetogenic [JAC: producing acetate as a product of anaerobic metabolism] growth and carbon fixation. We cannot use phylogenetics to reconstruct other members of this early ecosystem but we can infer their physiologies based on the metabolic inputs and outputs of LUCA. How evolution proceeded from the origin of life to early communities at the time of LUCA remains an open question, but the inferred age of LUCA (~4.2 Ga) compared with the origin of the Earth and Moon suggests that the process required a surprisingly short interval of geologic time.
I think that this paper has important consequences for extra-terrestrial life. If life came about in such a hostile environment as the early earth then it is my view that life was inevitable and is a direct consequence of the chemical-geological-physical conditions that were found then and that, given similar conditions, life is, if not certain, highly probable to be found elsewhere in the universe. I suspect that the universe is teeming with life of one sort or another.
+1
We need look no further than (but we should) the thermal vents deep in the ocean trenches where no light penetrates.
Google:
“The heated waters spewing out of hydrothermal vents are rich in chemicals that chemosynthetic organisms, can use as a source of energy”.
“But around hydrothermal vents, life is abundant because food is abundant. Hot, mineral-rich fluids supply nutrient chemicals.”
… it’s a party down there at these vents, weird and a wonderful display of life.
Great post!
Appreciate this – a gargantuan scope for this fundamental question.
Great to see this fascinating post in the midst of gloomy turmoil. Is it OK to comment an an apparent typo? (million for billion): “Previous estimates were in the 3.5-3.8 million-year range” —
Yes, I fixed that, thank you!
Well if viruses are already present, too, than they may be the descendants of an even earlier common single ancestor, or the descendants of a 2nd origin of life. It’s a bet that is likely unwinnable, but I’d wager on the 2nd origin.
It’s contestable whether viruses are even “alive”, so they’re not considered when thinking about LUCA.
They replicate and evolve by natural selection and are rather sophisticated little parasite machines, I think they meet just about every requirement for living except having their own metabolism. And, as natural selection would predict – if you can use someone else’s metabolism for your ends, why not jettison your own? Really, what’s the difference in functionality between a virus and a sperm cell? They are much more than quirky chemistry.
If you jettison your metabolic machinery, that implies you had it before. By that argument, viruses appeared after LUCA and couldn’t be it. I’d bet that viruses, mycoplasmas, and chlamydiae evolved as cell fragments cast out at various long ago times from previously evolved organisms — the envelope of an enveloped virus like herpes is actually composed of the host cell membrane. They are probably offshoots, foreign object debris, not ancestors.
A virus is not remotely like a spermatozoon. It’s like a crystal. It just sits there, inanimate. To me, a virus is good evidence that life evolved from large molecules because under the right conditions, other life forms can allow viral molecules to “come alive”. But I don’t think life evolved literally from viruses.
1. Most viral genes have no known homologs in bacteria, archaea or eukaryotes. Cell fragments would.
2. Sperm are genome injection machines like viruses. They clearly evolved from eukaryotes, but the process of cell simplification yet retaining functionality is a window on what might be possible in early virus evolution. Further, sperm only ‘work’ because of horizontal gene transfer of viral genes that facilitate cell membrane fusion.
3. I suspect viral ancestors did have metabolism once, but likely different in origin and more inefficient to that in LUCA. Hence, neither LUCA nor the descendant viruses had any use for it. Long gone into the evolutionary reject pile.
Wait…..I thought viruses themselves could not replicated. Instead they needed the cells of living organism to replicate. Is this true?
Viral genes code for virus-specific function and structure that the host cell doesn’t do, so lack of homology between them would not be surprising.
Any unnecessary host cell DNA or RNA that chanced to get incorporated into an assembling virus particle would be pared down and lost as superfluous to the tight packaging constraints of the viral capsid leaving only those virus-specific genes alien to host-cell life.
Agreed this doesn’t explain where all those viral genes came from, unless they just got lucky from doing it many trillions of times under selective pressure. But this pressure would optimize for viral imperatives, not cellular, and this seems to me to be a reason to lose homology, not retain it.
I would speculate that identifying viral proteins that mimic cellular functions at the virus-cell interface, such as initiating the interaction between viral RNA and the host ribosome (and out-competing the highly adapted normal interaction with host mRNA) would be the place to look for homology. In enteroviruses, the simplest human viruses, this interaction seems to be mediated by the 3-D structure of the RNA strand itself, even before any protein synthesis occurs. In later rounds of replication before the cell lyses, these other proteins synthesized in the first round may play a catalytic role. They don’t need to get packaged into the assembling viral particles. These “early proteins” are produced in tiny numbers compared to the late structural proteins produced in large quantities and are hard to harvest enough to study.
Even viruses that interact specifically with human receptors, like the famous spike protein of SARS-Cov2 and the ACE-2 receptor (or influenza’s hemagglutinin) could be examples of convergent evolution without requiring any preserved archaic genetic homology. (The key that opens many locks vs. the lock that any key will open.)
The virologist Anthony Racaniello takes the view that a virion is not alive, but once it invades a cell it springs to life.
And then it “dies” again when it leaves the cell as an intact and now-infectious virion but is capable, as was its parent, of springing to life in its turn. The passage of something from alive to not alive is always referred to as (irreversible) dying, yet a virus which has so died can come back to life.
When it’s infectious it’s not alive. When it’s alive it’s not only not infectious but it’s also invisible until the last few minutes when the maturing particles start to assemble. But it is still meaningful to talk about “killed” virus which is neither alive nor infectious and so doesn’t come to life. Without killed viruses there would be no Salk polio vaccine.
I find it hard to accept viruses as alive in the sense we developed before we knew that viruses existed (or bacteria or prions for that matter.) I suppose it doesn’t matter. All these things just are. If our concepts that date from antiquity can’t encompass them, that’s the limitation of our concepts, not the viruses’ fault.
The definition of living thing is that it is a self-replicating organism. Viruses cannot self-replicate: in order to reproduce they must get inside a living cell and co-opt the cell’s internal mechanisms (such as the ribosome in a eukaryotic cell) to make more copies of itself. It follows that, as Jerry suggested, they are not considered to be living entities. It is important to understand that LUCA was a highly complex entity, orders of magnitude more complex than a virus, and would already have been the product of a long evolutionary process that preceded its existence. Which is perhaps why this paper’s conclusions are so surprising. Before LUCA there would have been a time where life was dominated by protocells, but in the conclusion of the paper it states that LUCA was: “a cellular organism that was prokaryote grade rather than progenotic”: not a simpler protocell but a prokaryotic one. And even protocells would have been complex and would have had to have precursors. It is thought that the first living things evolved in an “RNA world” which would have been characterised by countless molecules like RNA endlessly combining and recombining, possibly in a highly energetic environment fizzing with energy from a “rain” of cosmic rays. (There would have been little or no oxygen in the atmosphere then – so there would have been no ozone layer to filter out high energy particles coming from space). This would have been the time that life began. Richard Dawkins defined the first living thing as the first self-replicating molecule. This is because once a molecule could reproduce independently, with variation, then this would have allowed natural selection to begin. Although I rather suspect that some form of selection would have been active before that, after all, non-living entities like viruses are still subject to natural selection. This is one of many points I want to explore with Richard the next time I, hopefully, get to speak to him.
I suppose if we carry that thought to its logical conclusion we would have to say obligate parasites are not alive! Reductio ad absurdam indeed.
To Christopher,
I don’t think @Peter Fisher was offering a definition of life so much as setting out tasks that life must do to be called life. Obligate intracellular parasites such as chlamydiae use their host’s respiration systems to provide energy for growth and replication but they carry out their own DNA synthesis and their own protein synthesis on their own prokaryotic ribosomes as do other bacteria. They can also be fitted into the phylogenetic tree without difficulty. Their relationship with other prokaryotes and to the LUCA can be productively discussed.
The also remain as discrete, identifiable membrane-bound organisms during their intracellular (and extracellular) life. This is another feature of living things: they are bounded, with an inside and an outside and they actively defend the difference between them using metabolic energy or robust structural barriers (e.g., in endospores) made with metabolic energy earlier in their life cycle. The astounding thing about viruses is that they disintegrate and literally disappear into the crowd of macromolecules after entering the cell and must do so before they can “spring to life”. Reassembling themselves into inert crystals is their last living act before they (temporarily) die.
I would say it’s the nature of the dependence that viruses have on cells that makes them not living things, not just its existence. When we talk about this dependence we need to describe it fully, as @Peter Fisher did.
Interesting post! Thanks!
Hearty thanks, PCCE, for this clear and detailed essay on these findings!
Thanks for this explanation. I had seen the original article but this explanation is easier to understand (for me, anyway).
They used MCMCTree (=Bayesian program).
Did they run their molecular dating analysis “on empty” (i.e., without considering the sequence data) to see how the priors they used for calibration dates interact to produce an estimate for LUCA?
I looked through their Methods, but I didn’t see an example of them running it “on empty”. Their result might just be from the way they set the program without their data actually contributing anything to the analysis? They used hard priors for LUCA of 3347-4520 Ma.
I’d ask the authors, not me; as I said, this isn’t exactly my bailiwick.
Fascinating, thank you for writing about this study, Jerry. These results continue a long trend of discoveries and research results that indicate that life occurs easily and readily. At this point it seems implausible to me that the universe isn’t teeming with life.
Nice point about L-amino acids. As you say, only the L-enantiomers are incorporated into polypeptide chains of all known life forms. Yet D-amino acids do exist in nature, some produced (but not incorporated) by bacteria and others by abiotic racemization. D-amino acids are used as carbon-source fuel by bacteria and some higher life forms. So primordial biochemistry was presumably “aware of” both enantiomers but solved the trick of peptidyltransferase (probably initially with catalytic RNA) with a L-enantiomer substrate once, first, and then never looked back. (Recall there is no specificity in the peptidyltransferase spot welder. The specificity resides in the RNA codons and the enzymes that attach each amino acid to the correct transfer RNA, which is also a neat trick. If that matchup is corrupted, peptidyltransferase will happily attach the wrong amino acid.). The chiral carbon atom in an amino acid is the one* that participates in the peptide bond, which indicates the hand but not the forearm has to fit the glove.
It’s reassuring that this took much more than merely a billion years to get right.
———————-
* Wikipedia reminds me that two amino acids have two chiral carbons. In these, the L- refers to the chirality of the carbon that will form the peptide bond.
I saw an abbreviated version of this the other day, but did not dig into it like you did. Thank you! It’s a super-interesting paper.
Even as an undergraduate student I knew that stromatolites went back to about 3.2 billion years ago, but I never expected the date to be pushed back this far! And yes, LUCA was already a sophisticated metabolic machine—a life form—even at 4.2 Ga. Because of how quickly life formed on Earth after its formation, I’ve long thought that life must be abundant in the universe, and I’m even more confident of that now. Once the conditions on Earth became stable enough to support it, life appeared. Not only that. Conditions on Earth at that time were probably not all that stable, implying that once formed, life must be pretty resilient indeed.
Thanks, Jerry!!
I thought that oceans, or maybe even liquid water, weren’t formed until less than 4 billion years ago. So does this new paper conflict with the hydrothermal vent theory? I’d previously been pretty convinced of the latter.
There is some interesting new research about this:
https://www.sciencenews.org/article/sulfur-key-first-water-earth
Dr. Coyne, Will you be producing a refreshed version of your book, Why Evolution Is True?
Interestingly enough, it was by reading your book that as a one-off, I came to understand what cancer is. Mainly, cancer is evolution going awry inside an organism.
Not likely given that the evidence is pretty much the same, but stronger! Thanks for asking. If I were to change it at all, I would add more molecular evidence, of the kind that Ken Miller and others use. But I think the book has enough to convince people of its title!
“Awry” – that implies that you think there is some “purpose” to cancer. What is that purpose, and how does it contrast with a non-cancerous cell’s “purpose(s)”?
It seems that there is still room for the idea that the precursors of life on Earth somehow made it from elsewhere in the Solar System — Mars, for instance.
According to some planetary scientists, Mars is thought to have had a clement environment earlier than Earth did.
But how such precursors can survive interplanetary space and the effects of asteroid & cometary collisions remains a puzzle for me.
This version of the panspermia idea, of course, doesn’t actually address the biochemical origin of life on Earth. It just puts the beginning somewhere else.
Some people, yes. Planetary scientists … would have a big problem with that.
Except for the ignition flash – a period of relatively high stellar luminosity for a few million years early in the life of a star (proportionate to Sol’s ~10 Ga lifetime), the star’s luminosity is fairly steady, only increasing by (order of) a percent (luminosity) per percent of the stellar mass burned.
For a Solar mass star, that translates to a few % luminosity increase per gigayear – the so-called “faint young Sun paradox” (Wiki has a page on it, under that title).
Mars being around about 1.5 [check – 1.523] times the distance of the Earth from the Sun, it is going to receive about (1 over (1.5)squared)= 1/2.25 [check – 1/2.319529] of the illumination from the Sun that the Earth does.
The Stefan-Boltzmann law applies, that energy radiated varies as the fourth power of the absolute temperature difference, So a 1/2.25 irradiation difference becomes a (1/2.25)^1/4 difference in (absolute temperature), which is a factor of 0.810307853919 (using my checked factor for Mars’ distance) for the absolute temperature on Mars.
Geological evidence (not common knowledge, but I’ve mentioned it here before) of mineral grains of different densities transported together by a fluid of water’s density (liquid ammonia won’t do ; feel free to propose your alternative fluids) ties the temperature of Earth, in the Archean, into the melting range of water. Which puts the temperature range of Mars in the range 273*0.810307853919 (melting point of water [K] * factor calculated above) ~= 221 K.
Now, you can add a chunk of global warming to that from greenhouse gases (major CO2 on Mars, minor CO2 and major H2O on Earth), but you’re going to struggle to get Mars up to water-melting temperatures without pushing your model Earth into ocean-boiling temperatures.
There is abundant evidence of water-carved features on Mars. Most of which are also fairly short lived – flood channels ; lake maximum shorelines. So the general opinion is that Mars experienced brief periods of warming (probably following a significant impact into CO2-rich deposits ; possibly following volcanic action in/on those same CO2-rich deposits) with ice melting, followed by extended periods of the water freezing, then sublimating back into the polar regions as the planet settles back into it’s normal “iceball” state. Time period – a few tens of thousands of years, to a few hundreds of thousands of years. Compare the 100~120 thousand years the Earth’s atmosphere takes to return to “normal” after a few petatonnes of CO2 is injected to it (volcanic/ tectonic action 55 million years ago ; fossil fuel burning today.)
Which is not a good recipe for developing life dependent on the abundance of a particular solvent, which disappears into the solid for millions of years at a time.
That’s me giving the “Mars early clement conditions” the time of day, and due consideration. To me, the concept doesn’t survive due consideration. Over to you for counter-argument.
I think there are more planetary scientists who subscribe to an early clement Mars environment than you suggest (possibly because of wishful thinking on their part). But Mars certainly looks old and mostly barren, with — as you note — evidence of liquid water on the surface being episodic and brief in its history.
Nevertheless, I think it’s reasonable to think that long-lived pockets of “Goldilocks”-type warm & moist environments underground on Mars may have existed even during “iceball” periods … sufficient to nurture and sustain the early stages of life before Earth did. (Or after early life on Earth was destroyed by impacts.)
Without the slippery tectonic plates of Earth, it seems that the old volcanoes on Mars could have provided clement underground environments — possibly in caves — for long periods (and even to this day). I’m speculating as a lay person, of course, and I certainly don’t have the academic background you have.
The question that I find more problematic is, “If life started on Mars and came to Earth by way of impacts, how did it survive the enormous dynamic forces involved?”
I doubt even the hardy tardigrade would find such a transit survivable.
Knew about LUCA. Learned there are acronyms for other ancestors LBCA, LACA, LECA. Always learning and loving it. Thank you for your science posts!
A month or so ago, Auntie Beeb had a programme on the “Mysterious Origin of Insects” (it may have been a PBS co-production, so probably on US/ Canada TV imminently) which added “LUC[i]I[/i]A” to the alphabet soup.
If you have the technology (“get-iplayer” ; geoblocking or a geographical correcting VPN), “https://www.bbc.co.uk/programmes/m001zhxz” “Last on Thu 23 May 2024 00:30”, but no mention of a US co-production, so you’ll have to snarf it direct from Auntie.
I’ll have to look closely at TFP for the biochemistry, but this geologist at least doesn’t get terribly stressed over “putting” LUCA before the LHB, for several reasons:
a heavy impact may not have been lethal to life, overall
Most of the popular scenarios for the origin of life (assuming there was one ; not everyone agrees on this point) involve, at some point, proto-life eating the free lunch of thermal energy and biochemically interesting compounds provided by warm (or hot) hydrothermal systems interacting with seawater. In particular, the hydrothermal systems found today (emphasised) at depths of several kilometres below (today’s) sea level.
Now, consider the Chicxulub impactor. Order of 8km diameter ; global effects ; 100% of non-avian dinosaurs went extinct (‘Birds are not dinosaur descendants; birds are dinosaurs, for all useful meanings of “birds”, “are” and “dinosaurs”.’) ; 50% of non-dinosaur reptiles ; an unknown percentage of mammals (I’ve never seen someone even attempt to estimate this. Odd.) ; all the large marine reptilian non-dinosaur reptiles (mosasaurs, ichthyosaurs, plesiosaurs, pliosaurs) ; and significant numbers of genera (but no higher-level taxa, TTBOMK) of marine unicellular organisms. All in all, a very odd selection of effects in different taxa and environments. But nowhere have I heard of any significant effects on the oceans. Sure, the lights go out for a few months to a year ; lots of sand falls from the sky, averaging fractions of a mm globally ; the sky turns a dull-red heat for a few hours (from lots of secondary impactors. But did 100m of the top of the oceans get boiled off? I’ve never seen anyone propose that, nor have I heard of anyone proposing an effect like that.
During the Late Heavy Bombardment (see next section), there were dozens to hundreds of impactors in the scale capable of boiling off a kilometre of ocean. One (just one!) field area I’ve seen sediment logs for has multiple kilometres thickness of “decomposed spherule sands” (Chicxulub’s blanket of such reached to Cuba, but barely to Texas ; what’s that – a couple of % of the earth’s surface) sometimes in single beds hundreds of metres thick. Now that’s an impact! It might even have been a multiple ocean-kilometre-boiling impact. And having the deposits “stacked” with little intervening material suggests that either this location was unfortunately sited, and was hit by overlapping impact ejecta sheets ; or this site is an average location, and the whole planet was getting whacked with such impacts on a regular basis.
Thin streaks of sediments in the region I’ve just described (the Barberton Greenstone Belt of South Africa), in between the thick sheets of decomposed spherule sandstones, are chert beds which are a well-known locality of Archean microfossils. [Treat that Barberton Wiki link with more than normal caution. It describes chert as an “evaporite” ; which can be true, but very definitely is not the normal environment for forming cherts]. Life was present while this bombardment was going on, and didn’t seem particularly bothered by it. Sure – each impact would have sterilised a few thousand km around the astrobleme. And within mere decades to millennia, the sterilised zone would have been re-inoculated by life from the rest of the planet. From the extinction point of view, “meh” – says Life, and carries on reproducing.
This was quite derived life – most likely photosynthetic – and far removed from the hydrothermal vent systems so-often considered as a locus for the origin of life. To extinguish life down there (rather than having “just another mass extinction”), you’d have needed to boil off multiple ocean-kilometres of ocean. 11-plus km on modern Earth (though whether there was a full Wilson cycle of plate tectonics with subduction trenches 3+ Ga ago is debated).
When you think carefully about it, if life originated (or rapidly inoculated) the edges of hydrothermal systems (where it can be found today), then it gets very hard to extinguish. You’d really have to approach the “magma ocean” state postulated for the aftermath of the “Moon-forming” impact. That was a 1:10 impactor:primary mass ratio.
That’s part of the reason that I don’t think that Earth will be sterilised until the oceans boil – probably between 1 and 2 Ga form now.
was the “LHB” an actual thing?
The hypothesis of the LHB was proposed in the 1970s to explain the range of rock and mineral grain ages returned in samples from the Moon. Unfortunately, this group of samples has a strong internal bias – they come from a small p[art of the Moon’s surface, and all severely affected by the Mare Imbrium impact. There is strong reason to believe that the dating of the “LHB” is strongly influenced by the date of the Imbrium impact, and that the whole putative changes in impact frequency that makes the LHB a thing (compared to the implied pre-LHB decrease in frequency-of-impacts and the post-LHB decrease in frequency-of-impacts) may simply be an artefact of that initial biased sample selection.
Don’t take that as a criticism of the Apollo programme. After the geopolitical need to show off nuclear missile systems, the scientific window dressing needed to be done with a politically manageable level of risk, and that resulted in a perfectly reasonable decision process for the initial landing (and so, sampling) sites. The tragedy is that the programme stopped, and we don’t have any “ground truth” sampling from, say, the Lunar Highlands relics of the (putative) pre-Imbrium anorthositic cumulate crust.
And yes – one modern bolt-on to the LHB concept is that it is the result of the “Grand Tack” of Jupiter’s orbital migration. Or, alternatively, it’s the consequence of Jupiter and Saturn passing through a 2:1 mean motion resonance, which destabilised the small bodies of the Solar system, throwing them all over the place, and some into Earth- (and Moon-) encountering orbits. That’s a modern bolt-on to the LHB concept ; it’s not evidence for the LHB. It would also help, maybe, if the various different explanations didn’t have a half-Ga variation in predicted dates for the Grand Tack, or MMR.
While it is literally in the textbooks, the LHB is a much debated and disputed idea. The simple model of a monotonically decreasing bombardment rate through the Solar system remains very much on the table, and has Ockham silently arguing in it’s corner.
I’d be worried about pushing LUCA so far back into the history of Earth – but only because going so far back is excluding a lot of wriggle room. But for reasons to argue against it … I don’t really have anything. It is compatible with the evidence from the Jack Hills (and Acasta) zircons which showed that the core zircons formed in contact with water that had been through a meteoric (weather : evaporation, transport, condensation, rainfall) cycle before 4.2 Ga ago. That’s … interesting.
But I’ve got something on tonight. Gotta go.
Hey, this post exceeds the 600-word limit by more than a factor of two. Have you read the posting Roolz? Please do so, and limit posts to about 600 words, thanks!
To quote Voltaire, I think, “I apologise for writing such a long letter. I did not have time to write a shorter one.”
You are also dominating the thread; please read the roolz about frequency of posting.
A propos exactly this, next time you’ve got the “web mechanic” in to kick the tyres and change the metaphorical oil, could you ask him (her/ it/ whatever pronouns small furry creatures from Alpha Centauri use) how much a “word count” widget in the post editor would cost?
I knew I was covering a lot of ground, but didn’t have time for copying the text, opening a word processor, pasting it, counting the words, editing it. Putting a “WC” function on the edit box would give the “post length Rool” some teeth.
Jerry, you mention the widespread view, now challenged, that life could only have started after the late heavy bombardment. Yet many years ago Stephen Mojszis and his student showed that once established, life would not have been so easily extinguished: Microbial habitability of the Hadean Earth during the late heavy bombardment, https://www.nature.com/articles/nature08015
Yes, I missed that; I was quoting from the paper and didn’t know of other suggestions. Thanks for adding this.
Is that the paper that considers the amount of ocean you’d need to boil off to eradicate near-surface hydrothermal systems favoured as a site for Origin(s) Of Life?
I don’t have the wherewithal for £200/year of Nature, and the abstract is coy about which terrestrial environments they’re considering. But a 2009 publication sounds about right.
Yes, that’s the one. However, a 2019 paper from the same group now says there probably wasn’t a LATE heavy bombardment anyway; I think it’s open access: https://iopscience.iop.org/article/10.3847/1538-4357/ab2c03 Onset of Giant Planet Migration before 4480 Million Years Ago, Stephen J. Mojzsis et al., The American Astronomical Society. All rights reserved.
The Astrophysical Journal, Volume 881, Number 1
Thank you Jerry.
What an ingenious piece of research! Many thanks for doing the heavy lifting in summarizing it for us.
Jerry, most of this is fine and interesting, but can I make a terminological quibble? LUCA is last, in the sense of most recent. You describe it as
“What that means is LUCA is the first creature whose descendants include every species alive: the ancestor of all of us.”
No, it is not the first, it is the last. The most recent (and that’s what most of your interesting article is indeed talking about). If there was a lineage leading to all present-day life, LUCA is not at the start of it. It is at the oldest fork that has descendants of both branches today.
I see this all the time these days, that LUCA = The origin of life. No!!
Joe,
Thanks. I hope I emphasized that LUCA was not the beginning of life; you did read it, right? It was just a thoughtless word and I’ve fixed it.
Jerry
——-
You know where this is going, don’t you!
Creationists are going to argue this proves that there was not enough Darwinian “Deep Time” for life to have began in only 300 million years.
I, too ,am scratching my head about such a short time period. Maybe chemical fore-runners of life did indeed originate in space, and the the Late Heavy Bombardment brought them to earth. Continuous collisions could have mixed up all the prebiotic chemicals too, and created extensive temperature gradients across the planet providing optimal conditions for bio-chemical reactions.
Just a thought.
+
I’ve seen less stressing over the origin of the molecules involved in more recent decades. Various classes of organic molecules can be generated by a variety of inorganic interactions between geological liquids, minerals and carbon-containing gases (in solution, typically) – that doesn’t seem to be a real problem.
But the “concentration” problem remains : regardless of the origin of the molecules under consideration in a terrestrial, marine, geological or an astronomic context, when your proto-life is bobbing around in the pre-life oceans, it is in an extremely dilute biochemical “soup”. Which would explain why autotrophic organisms typically latch onto the commonest carbon-containing molecules in their environment, and make everything else they need from that. (It’s somewhat different for terrestrial heterotrophs like us – but could the first life forms have been anything other than autotrophs?)
The shortening period of time for origin of life makes me uncomfortable too. But the “concentration problem” makes the putative extraterrestrial origin of (pre-)biological molecules pretty moot to me. Sure, they may have been in the rain of debris coming down to Earth. But they’d still have been very dilute in the oceans, making the metabolic construction of biomolecules a useful trick for proto-life to incorporate.
At least the news that Earth had a rainfall (meteoritic) cycle in the early Hadean has been around for a decade and a half now. So people have had some time to digest that.
Didn’t somebody (some team?) use a similar duplicated-gene phylogeny to root the big tree of life about 30 years ago?
Are you thinking of Woese and his “three domains” model. Which was late-70s, so closer to 50 years ago (where’s my Zimmer frame?)
He used one of the RNA->protein converting ribozymes, IIRC.
No. As I recall, Woese couldn’t really root the tree because there is no outgroup to the three domains. Ten or so years later someone had the idea of using duplicated genes to construct two three-domain subtrees, with each serving as the root for the other.
I found the paper. It’s Evolution of the vacuolar H+-ATPase: implications for the origin of eukaryotes.
JP Gogarten, H Kibak, P Dittrich, L Taiz, EJ Bowman… – Proceedings of the National Academy of Sciences, 1989.
Great paper; great summary! But I prefer LUCA-AFAWK (as far as we know) or ALUCA (apparent) since the discovery of an organism descended from the grey area in the reproduced tree diagram is far from improbable. Recent surprises include Asgardarchaeota and giant viruses, so who’s to say what we still may find? And then there’s the mystery of viral origins, much discussed above. To me, it just seems a bit hubristic to confidently assert the term “LUCA”.
It seems to me that the molecular evidence points to the fact that all current lifeforms must have had a common ancestor. If there were an organism, as you suggest, in the grey area, then that creature would have been LUCA. Although and I suspect the authors of the paper would agree, that horizontal gene transfer might well have had significant influence on the evolution of early life, so the picture might be quite complex.
To be clear: LUCA was a good descriptor, until we had the kind of focus which this paper provides. If some radically strange sister clade emerges then the conclusions in this paper won’t necessarily apply to the newly shifted LUCA, so there is considerable scope for confusion. It’s like we decided to apply H. neanderthalensis to H. heidelbergensis because we’ve just discovered Denisovans: imagine how confusing that would be!
But the parallel taxon you are suggesting would have to have had the same genes as the LUCA that is being proposed in the paper because all modern taxa share them. I do agree that it is difficult to be certain about precise details this far back in time, but the DNA that is shared between all living organisms suggest that LUCA would also have had the same DNA, that’s how we know it’s LUCA.
Thanks for all your hard work analysing and summarising the paper, Jerry. I’m sure everyone learned new things. I certainly did.
Just a thought, and I want to quickly clarify the fact that I’m not a palaeontologist, but as I understand it, according to the molecular clock, modern phyla diverged from their common ancestors deep in the Proterozoic, much earlier than the fossil evidence suggests.
Modern phyla had certainly diverged by the mid-Cambrian, as evidenced by the Burgess Shale assemblage, and others, but as far as I know, no representatives of modern phyla have been found in the Ediacaran in the very late Proterozoic. In fact, I’ve heard it suggested that it is hard to say whether the Ediacaran “fauna” even belong to the animal kingdom. This far back in time, of necessity, the fossil record is very sparse, and it just may be that representatives of modern phyla just haven’t been found yet, but if the molecular clocks ran a lot faster earlier in Earth’s history, perhaps because of quicker generation times, than they do now, then this would explain the anomaly and would mean that LUCA may have lived much later than this paper suggests.
Does anyone know whether this really is a live unresolved anomaly, or have I simply misunderstood?
I’m not clear on what you’re understanding, or misunderstanding. But …
Well, there is (somewhat disputed, but accumulating) fossil evidence of complex multicellular, sediment deforming life well back into the Proterozoic – upwards of a gigayear ago. (The oldest I have in RAM is the “chain of beads” structure, probably an ichnofossil, whose name will come back to me in a minute or 5, and dated to about 1.5 Ga.) The Francevillian biota of Gabon (explored because of it’s association with the uranium mines of the natural nuclear reactor at Oklo) is dated to about 2.2 Ga, and is interpreted as multicellular, if not clearly related to modern phyla.
The Burgess is fairly high in the Cambrian. Well into the upper. The Chenjiang biota fomr China shares many elements with the Burgess, but is 30-odd million years (~5%) older, and is mid-Cambrian. The Cambrian is one of the longer periods of Earth history.
This point is argued on a regular basis. People are asserting “this Ediacaran is a un-derived member of that modern phylum” on a regular basis, followed by months of discussion, comments, and replies. I don’t think anyone has found a “smoking gun” (was the gun at Butler smoking? Is this a good metaphor?), but people are trying it on a regular basis.
I think the problem is more that they haven’t been recognised yet, not that they haven’t been found. See above description of frequent arguments. That the arguments are polite and use long words disguises the blood dripping form the ceiling. Slightly.
When I was a student, in the 80s, I was very dubious of “molecular clocks”. But over time their methodology has improved, and they’re approaching the limits imposed by the palaeontology they depend on. The first-known fossil of a particular group is almost certainly not the origin of that group, and the accuracy of molecular clocks isn’t ever going to exceed that limit. But they’re within that level of uncertainty now. I still don’t trust them as much as a few decimal places on an isotope ratio – but that’s because I’ve worked with MS and understand their nuts, bolts and metal-to-metal seals better than biological muck-aboutery. But I’m sure squishyologists don’t trust hammer-wielding geologist’s methods either. (I used that as an email address once – I think for taunting Creationists.)
Thanks, this filled in a lot of the gaps for me. Corrections and clarifications much appreciated.
I am behind on my reading, but God I didn’t skip this.
Interesting paper, great summary, thanks!
The most interesting book I have read on LUCA and the early evolution of eukaryotes is Nick Lane’s The Vital Question. Like everything else he has written, I highly recommend it.