Abiogenesis isn’t simple


I’ve been reading Hazen’s Gen-e-sis: The Scientific Quest for Life’s Origins (amzn/b&n/abe/pwll). I haven’t finished it yet, but it led me to appreciate this post and this diagram more.

i-b383f1d62c2e6c1b172c916fa277fd79-abio_theory.gif

Isn’t that the truth? What creationists see as a weakness, that we have so many ideas (and so many gaps), we see as a strength.

Comments

  1. says

    Back when I studied chemistry, I used to have nightmares about carbon and carbon chains. Literally. Creationists just don’t understand how the unique structure of carbon allows it to readily combine with other elements.

  2. SkookumPlanet says

    Creationists are trying to manufacture explanatory weakness out of the strength of scientific approaches. What actually goes on inside their heads is anybody’s guess, but I’m inclined to favor choices low on the morality scale.

  3. Caledonian says

    To paraphrase Homer Simpson:

    “Pfft, facts. You can use facts to prove anything that’s even remotely true.”

    As long as certain important events are shrouded by ignorance and uncertainty, theists can pretend that the explanations they’ve been peddling for millennia are just as valid as rational speculation. Any attempt to shed light on those mysteries will be perceived as an attack and treated as such.

  4. says

    Rick: No, no, no. That would be a silly theory. Unfortunately, it’s just the sort of strawman that gets bandied about among the atheist cohort. The real theory of creation is:

           Dirt ——–> Man
                     Î›
                    /
                   /
                  /
                 /
                /
          Divine Breath

    See?

  5. says

    I thought the creationist idea of abiogenesis was “Dirt —> Man”.

    Or as Kent Hovind puts:
    “Evolutionists… believe a rock can turn into a rocket scientist.”

    That would be really funny if he wasn’t hopping all over the country teaching that crap to children.

  6. Theodore Price says

    I’m gonna have to run out and get that book. As a neuroscientist, I know dangerously little about this subject, but it sure is fun to ponder. As I understand it, current opinion is that there was a preponderance of RNA and perhaps simple proteins during the polymer and transitioning into the replicating polymer phase. Is there any evidence that there ever was, or that it is even possible, to have a reverse translatase (reverse transcriptase but protein to RNA) complex. Albeit the energy requirements for such a thing would make it improbable… Just a thought, and not something I can get a reasonable opinion on from a group of behavioral neuroscientists. Thought someone around here might offer a comment (before the thread gets hijacked by creationist babble).

  7. wamba says

    Is there any evidence that there ever was, or that it is even possible, to have a reverse translatase (reverse transcriptase but protein to RNA) complex.

    There is no evidence that there ever was such a thing. Possible? I suppose so.

  8. Kagehi says

    You can find replication polymers all over the place, even in space dust. What makes simple replication and RNA different is only complexity and some basic function. As for energy requirements… Its hardly any more energy than you find from sunlight or better yet, volcanic vents. And energy doesn’t have to come from external sources, the right chemicals can generate transient electrival charges, thermal variation, etc. I am not sure what part is “improbable” in the energy requirements, but then I am neither a neuroscientist nor a biologist myself, so maybe I am just missing something?

  9. Theodore Price says

    Kagehi, You’re missing something. I’m talking about a very specific process (translation of RNA into protein) working in reverse. It is well known that DNA into RNA can be reversed into RNA into DNA with a trascriptase called reverse transcriptase. In terms of energy, I have no question that sufficient energy could be found somewhere, I am just wondering out-loud if the RNA-protein complex that would be needed to reverse translation could harness enough energy to actually achieve the reaction.

    Wamba, I figured there was no evidence (kinda a silly question) but how about reasonable hypotheses on whether something of the sort could work or was ever involved in abiogenesis. I’ve been trying to find information on this for a long time and have never come up against anything other than exam question type google hits. Just figured this was big chance to ask a crowd that might know.

  10. wamba says

    I thought the creationist idea of abiogenesis was “Dirt —> Man”.

    Then there’s the Goo to You version.

  11. wamba says

    Wamba, I figured there was no evidence (kinda a silly question) but how about reasonable hypotheses on whether something of the sort could work or was ever involved in abiogenesis.

    I have never heard a hypothesis that such a thing was ever involved in abiogenesis. If I heard any hypothesis of the sort I would classify it as extremely unlikely (and unreasonable).

    Energetics is not the problem; you would only have to do it once for each polypeptide, since you could then reproduce the result using standard polynucleotide replication technology.

    To make it happen, you would need to build a separate capability to recognize each amino acid, either while it is in the polypeptide chain – or I suppose while you are taking it apart piece by piece. This would be more difficult than the forward direction of tRNAs and tRNA charging enzymes, since those can be done off-site and fit interchangably into the same ribosome. Polypeptide do not have the sort of base-pairing properties that polynucleotides do. The actual ligation of the RNA pieces together is enzymatically simple, it’s all the specific recognition; and of course you would have to take care to match up the same genetic code in reverse as you did in the forward system. I cannot at present think of a way the code-matching could be done naturally.

    Is that enough BS for one helping?

  12. wamba says

    Such a system could be designed, at least theoretically, but it would take a level of technology beyond what we have today.

  13. Theodore Price says

    I’ve actually managed to finally find some papers that support the hypothesis (albeit they are from the 70s and 80s). I did find one paper that shows evidence for the engineering of an rt tRNA (J theor Biol (2001) 209 181-187). I’m not qualified to comment on the validity of the claim, but it sounds reasonable. Found all this business in pubmed by searching (in quotations) reverse translation. Reading the theory papers from the 70s was great — its wild to see in writing just how much we’ve learned in a short period of time.

    As for getting the message right in reverse, maybe that could actually be an advantage. Assuming a diversity of codons for the same amino acid (at a prebiotic stage) you could generate an astonishing diversity of RNA codes simply.

    Thanks for the helping of BS… I’m listening if you’re willing to offer more.

  14. tacitus says

    Then there’s the Goo to You version.

    I’ve heard “Goo to Zoo to You” before and, believe it or not, “who’s ever heard of a walking, talking rock?”

  15. Siamang says

    “I believe ‘in the beginning, God.’ and you believe ‘in the beginning, dirt!'”

    That’s hovind I think. There’s something magical about his delivery when he says this. He’s like an amazing salesman pushing aluminum siding.

  16. wamba says

    And then there’s this… http://www.miraribiosciences.com/peplica.phtml . Its from a biotech so of course there is no info but it sounds interesting (and if it really works I would use it in the lab — what a tool!!!)

    Interesting, but absolutely no info on how they accomplish the “reverse translation”.

    It can be done with modern machinery: run the protein through peptide sequencing equipment to derive the sequence, use a lookup table to find the corresponding RNA or DNA sequence, and synthesize it on a DNA synthesizer. This would take a lab bench full of equipment; I assume the original question was about accomplishing it at the molecular level.

    Assuming a diversity of codons for the same amino acid (at a prebiotic stage) you could generate an astonishing diversity of RNA codes simply.

    There are a number of known genetic phenomenon for generating diversity of RNA sequences that are much simpler than what we are talking about here.

  17. wamba says

    I’ve actually managed to finally find some papers that support the hypothesis (albeit they are from the 70s and 80s). I did find one paper that shows evidence for the engineering of an rt tRNA (J theor Biol (2001) 209 181-187).

    I would discount anything from the 70s and 80s, as the “RNA World theory” really took off with the discovery of catalytic RNA in the 1980s and obsolesced a lot of previous stuff. I dug up the Nashimoto paper and will give it a look. I also ran it through Science Citation Index to see what subsequent papers have cited it; 3 hits:
    Moreira A,
    Genetic algorithms for the imitation of genomic styles in protein backtranslation
    THEORETICAL COMPUTER SCIENCE 322 (2): 297-312 AUG 30 2004
    Trevors JT
    Early assembly of cellular life
    PROGRESS IN BIOPHYSICS & MOLECULAR BIOLOGY 81 (3): 201-217 APR 2003
    Trevors JT
    Genetic material in the early evolution of bacteria
    MICROBIOLOGICAL RESEARCH 158 (1): 1-6 2003
    I’m not sure how much time I want to spend on this.

    After a night’s sleep, I still see the same difficulties I did last night: the difficulty of recognizing the residues in the polypeptide chain, lack of an analogy to the nucleotide base-pairing that helps make forward translation work, and the indepence of the forward and backward genetic codes.

    Cheers,

  18. wamba says

    OK, I read the Nashimoto paper. I see numerous weaknesses in his proposed scheme for reverse translation, and no indication that the author is aware of those weaknesses.

    • The author proposes that polypeptides once replicated themselves. There is no evidence for that, and immense practical difficulties to the possibility. Polypeptides lack any property analogous to the double helical base-pairing of polynucleotides, which is what allows polynucleotide replication to work.
    • The author proposes that reverse translation used to occur. There is no evidence at all.
    • The author suggests that the development of a useful suite of polypeptides through mutation of the RNA genetic material, translation and selection would have taken “tremendous time”, and throws in a line about monkeys producing Hamlet. He blithely ignores the fact that the same criticism applies to his proposal that useful proteins came about through random thermal polymerization of amino acids, and he’s lacking any replication method for polypeptides to ratchet him out of it.
    • Figure 3, in which he’s managed to get a couple pieces of RNA to swap a few bases doesn’t prove much of anything. It doesn’t address the most severe difficulties of his scheme.
    • The author blithely ignores the independence of the forward and backward genetic codes, an obstacle I have already mentioned.
    • The forward translation system smuggles in a lot of recognition specificity in that the amino acids are attached to the tRNAs by tRNA charging enzymes. This smuggling is possible because of the base-pairing properties of RNA. I can’t come up with any way to acheive such smuggling in the reverse translation system. The author suggests rtRNAs which both recognize the peptide residue and carry the condon information; i.e. they perform the job that in the forward system is done by the tRNA AND the tRNA charging enzyme. This has serious obstacles.
    • The author expects to have a working reverse translation system working by 2011. I admire but do not share his optimism.
  19. Theodore Price says

    Wamba, thanks for taking the time. I agree with alll your points (especially your third bullet), although I have to say I got the impression that the author acknowledged the weaknesses of the idea in the manuscript (at least in terms of the tone of the wording). Nevertheless, its an interesting idea and if anyone was ever able to engineer such an enzyme complex it would be an incredibly valuable tool. For instance, I often work with cultured neurons that form synaptic contacts in vitro. The synapses can be easily microdissected by a laser (or by sucking it into a patch pipette) but measuring the amount of a single protein at the synapse is nearly impossible. Using this technology such an experiment could easily be realized. That would make me a very happy man!

  20. David Harmon says

    Some more issues about “anti-transcriptases”:

    1) RNA and DNA are linear molecules, while proteins fold into all sorts of shapes. The endpoints may not be accessible, and folding can involve guided transformations, clipping off sections or assembling polypeptides from different sources. Information about those changes is not to be found in the sequence of amino acids. Likewise for metal ions, saccharide attachments, etc.

    2) IIRC, some amino acids have multiple codings. That means the reverse mapping is ambiguous.

  21. wamba says

    To add to David Harmon’s post, some of those “post-translational modifications”, actually change the identity of the residues; phosphorylation of serine & tyrosin, formation of disulfide bonds from cysteine, etc. Codon ambiguity wouldn’t actually hurt anything; translation of the resulting gene would give you the same protein regardless of which codon was used.

    but measuring the amount of a single protein at the synapse is nearly impossible.

    Are the amounts too small for TOF mass spec to work?

    On a tangent, there are a few polypeptides that are made by nonribosomal methods. I have a vague recollection of something from biochem class with both D and L residues; possibly a poison frog toxin?

    Oh well, just 5 more years until Nashimoto has proved us all wrong.

  22. Theodore Price says

    I don’t do TOP mass spec myself but I have tried to set up a collaboration to do it and no one seems to think it will work. Maybe I just haven’t asked the right or willing person yet, but based on what I do know about i think the chances are slim (although I’d be willing to give it a go for a few months).

    In terms of losing post-translational modifications, that would certainly be a downfall, but plenty of applications don’t really require that level of information.

    5 more years indeed…

  23. Theodore Price says

    I don’t do TOF mass spec myself but I have tried to set up a collaboration to do it and no one seems to think it will work. Maybe I just haven’t asked the right or willing person yet, but based on what I do know about i think the chances are slim (although I’d be willing to give it a go for a few months).

    In terms of losing post-translational modifications, that would certainly be a downfall, but plenty of applications don’t really require that level of information.

    5 more years indeed…

  24. Torbjorn Larsson says

    This is one (of many) reasons to hang around Pharyngula – one is often invited to see an area of knowledge sketched in a post, and inform ones naive views a little.

    This post perhaps settled a naive discussion about why life has a common ancestor in another thread. If one searches wikipedia for hypercycle one is redirected to the quasispecies model ( http://en.wikipedia.org/wiki/Hypercycle ).

    The theory (since modern RNA quasispecies are observed) chooses one quasispecies. It also seems to balance the mutation rates through pushing them close to the error threshold, is that correct?

    It also both confirmed and refuted my naive view that abiogenesis seems simpler to look at as a process instead of looking at life specifics. The model of quasispecies specifics is process based. :-) It’s always fun then the right choice is a third option.

    I have put the real deal Genesis in my reading list too.

  25. says

    Creationists are trying to manufacture explanatory weakness out of the strength of scientific approaches.

    yeah, and all they need are their principles and Wiiliam Jennings Bryan-caliber oratory to do it. and Falwell knows it.

  26. Xiao Bao Clark says

    I realise this thread is stupidly old, but (coming from an incredibly naive 1st year biomed student) couldn’t the problem of the protein folding be over come by simple denaturation by high or low pH, temperature etc? If that’s the case, then it’s not hard to imagine a ‘reverse-ribosome’ binding onto one end and working it’s way across, then using the same method reverse transcriptase uses to bring it back to DNA.
    In response to the cutting out of parts, I believe this only occurs when the molecule is to be activated? So it would then produce inactive DNA. This is obviously going way out on a limb, but did somebody say ‘introns’? :) I realise they get cut out before translation occurs, but what’s to say that isn’t a process that has undergone development to the point where they do cut out the unnecessary parts? Saying this, if the protein got to the backtranslation complex fresh out of production (ie before any part removals) then what you’d get is a functional DNA gene. It would definately help to explain why the majority of DNA doesn’t appear to code for anything.
    As for the codon triplet specificity: I could be very very wrong here, but do certain codons appear for an amino acid more than other ones? If this is the case, it could well be that those common ones were the ‘ancestor’ codons, the original ones the system made knowing(I use this term loosely) full well that DNA replication can fuck up alot, and reducing the effects of base pair substitution(mutation) significantly.

    I’m really hoping someone smart will see this and thoroughly discredit everything I’ve said here. Please inform me :) Thanks.

  27. Xiao Bao Clark says

    I realise this thread is stupidly old, but (coming from an incredibly naive 1st year biomed student) couldn’t the problem of the protein folding be over come by simple denaturation by high or low pH, temperature etc? If that’s the case, then it’s not hard to imagine a ‘reverse-ribosome’ binding onto one end and working it’s way across, then using the same method reverse transcriptase uses to bring it back to DNA.
    In response to the cutting out of parts, I believe this only occurs when the molecule is to be activated? So it would then produce inactive DNA. This is obviously going way out on a limb, but did somebody say ‘introns’? :) I realise they get cut out before translation occurs, but what’s to say that that isn’t a process that has undergone development to the point where they do cut out the unnecessary parts? Saying this, if the protein got to the backtranslation complex fresh out of production (ie before any part removals) then what you’d get is a functional DNA gene. It would definately help to explain why the majority of DNA(human) doesn’t appear to code for anything.
    As for the codon triplet specificity: I could be very very wrong here, but do certain codons appear for an amino acid more than other ones? If this is the case, it could well be that those common ones were the ‘ancestor’ codons, the original ones the system made knowing(I use this term loosely) full well that DNA replication can fuck up alot, and reducing the effects of base pair substitution(mutation) significantly.

    I’m really hoping someone smart will see this and thoroughly discredit everything I’ve said here. Please inform me :) Thanks.

  28. David O'Hagan says

    I have talked with Mark Martin and Nashimoto about their experimental evidence for reverse translation… and they have none. I tried my own approach to the pRT chemistry using RNA aptamers to recognize the N-termini of peptides… and to sort them based on the N-teminal residue. After sorting I was going to tag them with an artificial codon, then use edmund degradation to remove the n-termini and repeat the process. It was so far out there that I could only find temporary funding for the project. On my Myspace profile I have published the grant application that describes the mechanism and preliminary findings. I believe that with further work it might be possible… but the preliminary results suggest that the specific sorting of peptides based on the recognition of the last three N-terminal residues with RNA aptamer pools is not specific enough for a complex mixture. I tried to affinity purify those peptides with an RNA aptamer population that was selected against peptides of the sequence L-X-X, and this population of RNA aptamers selected the L-X-X-L-X peptides from a random population of peptides (X-X-X-X-X). If anyone else is working on this… I would like to here from them. Dave