Re: RNA World

From: "Perplexed in Peoria" <jimmene...@xxxxxxxxxxxxx>
I said it correctly. Proteins are phenotype. Phenotype is heritable.
Indirectly heritable (thru the genes) like all phenotype.

No, phenotype isn't inherited. Phenotype is constructed anew from
combinations of all the various inter-relating aspects of the
genotype, both the way different genes affect steps along a single
biochemical pathway, and the way multiple biochemical pathways
contribute to the overall effect which is just one phenotype
character. Depending how phenotype is defined, environmental
factors, and somatic mutations, may also may affect phenotype, and
those parts of phenotype are in no way whatsoever inherited.

Most people take 'RNA world' to mean the era before the
RNA->polypeptide step first appeared.

Oh, I didn't realize that. Somebody a year or so ago posted some
long multi-part treatise wherein he worked backwards in time, using
clues from present life back into the RNA world and even earlier. I
don't recall who that was. Do you remember? What was his definition
of the "RNA world"?

Given the confusion in term "RNA world", I guess we need to
disambiguate by making more than one different each-more-specific
term. Would accept these terms?
- RNA-ribozyme-only world
- RNA-ribozyme+polypeptide world

I certainly agree that the DNA->RNA step was added at some
point in the past, as of course was the RNA->polypeptide
step. The question is which came first. You have given no
reason why you think that polypeptides appeared before DNA.

Well I was just presenting a very first guess how the sequence
might have happened. It's been a long time since I first posted my
sequence, and this is the first time anyone has questionned my
hypothetical sequence. So let me think really hard to see if I can
come up with a reason why my original just-so-story is more likely
than your proposed alterative ... OK, here it is: Ribyzymes offer
only a very limited amount of possible catalytic activity.
Polypeptides offer a much wider variety of activity and much finer
tuning of shape to match specific reactants and desired products.
Polypeptides also offer disambiguation. Whereas RNA can fold into
an enzyme or unfold to be a template for replication or
transcription, polypeptides pretty much do only one thing, so they
don't need to serve two purposes, and they don't have to be
cross-optimized for both purposes, especially in cases where one or
the other function is INAPPROPRIATE (harmful) and must be somehow
suppressed at the cost of making the desired function grossly
sub-optimal.

Consequently the RNA+ribozymes-only world had a very limited
ability to adapt enzymes to very specific and efficient catalytic
steps, was hampered so badly that evolution didn't proceed very
well, compared to how it can proceed today. It's unlikely that
specific mechanisms for processing DNA and RNA separately could
have evolved with such non-speciic ribozymes as the only available
catalysts. More likely the major evolutionary step of polypeptides
occured, which even in crude form (template-matching shape of
3*RNA/aminoAcid) was still much better than the old ribyzyme
method, causing a "explosion" (as in Cambrian, except at the
biochemical level way back then) of evolution, which rapidly
improved the specificity of 3*RNA->aminoAcid coding simultaneously
with starting to improve the general specificity of polypeptide
folding shape. Once the genetic code was finally "perfected", the
evolutionary "explosion" rapidly produced several major clades and
sub-clades, freezing the penultimate changes to the genetic code,
all of which enjoyed much better ability to adapt (by
3*RNA->aminoAcid sequences) than the old ribyzyme world, at which
point one of these successful clades chanced upon DNA as an
alternative to RNA, more stable even if a bit harder to work with,
a tradeoff, which became more and more balanced in favor of DNA as
genome as the transcription and synthesis mechanism evolved better
specificity.

Summary, RNA+ribozyme-only world just didn't have the ability to
evolve lots of wonderful new mechanisms such as DNA genome. Only
RNA+ribyzyme+polypeptide has such evolutionary capability within
the short time span allowed (a few hundred million years). There
might in fact have been one clade of RNA+ribozyme-only life which
first invented some DNA chemistry, but it was overtaken and driven
to extinction by its much faster evolving RNA+ribozyme+polypeptides
cousins. (Or horizontal gene flow might have occurred, whereby
there was no pp-before-dna or dna-before-pp, rather both pp and dna
in parallel, but pp dominating, and stealing dna from the other
clade before driving it to extinction.)

- RNA world, period. (Biochemical synthesis of RNA bases, RNA
replicase, and the use of loose RNA strands as enzymes.)
- Above, plus biochemical synthesis of loose amino acids.
- Above, plus mapping from RNA triples to single amino acids,
yielding modern enzymes.
As in the Sidney Harris cartoon, I think you need to provide more
detail on that third step. ;-)

As I said, I posted only a quick outline/summary of my sequence.
The full details were posted years ago, part by me, and part by
other people in this newsgroup or the other I don't remember.
Summary of the transitional steps:
- First the synthesis of lone amino acids, some already occurring
from abiotic processes but long ago gobbled up and now a scarce
resource, but now that they are synthesized they are available in
whatever quantity a cell might need them, plus brand new amino
acids that never were abiotically produced are now available for
the first time ever.
- Crude polymerization processes, such as production rules
(for example, whenever you see MetAla then insert a Phe, and
whenever you see AlaPhe then insert Met, and whenever you see
PheMet then insert Ala, which together means any time MetAla occurs
by chance it gets auto-extended to a repeating 3-pattern). Such
production rules could have been the result of ribozymes that match
the shape of the first two amino acids then match an activated
third amino acid to force attachment then "lose interest" in
attachment as soon as the amino bond has happened.
- Longer and longer production rules, whereby a short "seed"
polypeptide attaches to the start of a long RNA strand, and then
all the rest of that RNA strand has triples with affinity to
various amino acids to draw them into the polypeptide down the full
length of the RNA strand. At this point, in a crude way, just
about any polypeptide is possible, not just random sequences and
repeating patterns. But still the simple affinity of RNA triples
to amino acids (and usually only the first two RNA bases of each
triple have any real effect, the third being more of a "spacer")
results in considerable ambiguity as to *which* amino acid will
take each location in a long polypeptide. So of all the
polypeptides resulting from a single RNA strand, there's a lot of
single-base polymorphism.
- Helper enzymes to make the 3*RNA->1*aa coding system more specific,
which gradually evolve to the general mechanism of tRNA.
- Gradually the helper enzymes including tRNA develop to a
fullfledged ribosome with very specific mapping from 3*RNA to 1*aa.

Note that even today, ribosomes are 65% RNA and only 35%
polypeptides. Apparently the original 3*RNA->1*aa system was nearly
100% RNA, and over time (while the 3*RNA->1*aa system was evolving
to be more and more specific) new parts were a mix of more RNA and
the brand-new polypeptides, and more recently as coding for
proteins became more specific and better evolvable the mix of *new*
parts became more and more polypeptide, and occasionally a more
specific polypeptide would actually replace an older RNA enzyme
(ribozyme) in the ribosome, so there's a RNA->AA takeover in
progress now, which has run only 35% to completion so-far. Whether
it will stall about where it is now, or eventually result in nearly
100% polypeptide enzymes, is anybody's guess.

Note this solves a chicken-and-egg problem: The mechanism to map
RNA to amino-acid sequences first started to develop in an
environment where there were *no* pre-existing amino-acid sequences
(except random sequences or repeating patterns by accident, nothing
that would be useful as a specific enzyme), so the first versions
of 3*RNA->1*aa mapping had to rely on RNA enzymes entirely, which
indeed seems to have been the case, given the preponderence of RNA
over polypeptides in ribosomes even now. A positive feedback loop
occurred later, when the already-existing crude mapping system
allowed the first pretty-specific amino-acid enzymes to evolve,
which could be used to improve the mapping system, which could then
produce more specific amino-acid enzymes, which could be used to
make the mapping system even more specific, etc. etc. etc.

So, if I understand you correctly, at this stage we have RNA in a
genetic role and DNA in a purely structural role (a role which no
longer exists). Odd, and somewhat interesting.

Yes, that was my first "just so story" idea. However I see you've
suggested a different idea (why didn't you think of it when I first
posted my idea years ago?) ...

But my intuition tells me that RNA and DNA are so similar that
early ribozymes and enzymes would have trouble telling them apart.
In fact, even today some enzymes don't discriminate. (For
example, the ones which transfer phosphates between the nucleotide
diphosphates and triphosphates thus moving phosphate potential from
ATP to all of the other nucleotide diphosphates.)
It seems more likely to me that at some stage there was no discrimination
and the polymerase enzymes or ribozymes would add whichever (ribo-
or deoxyribo-) nucleotide happened to be present.

Hmm, I like that idea. One problem for the genome is that RNA and
DNA have very different coiling tendencies. So I can't see there
was ever a time when genome replication and consequent cell fission
would indiscriminately work with strands that were sometimes mostly
RNA and sometimes mostly DNA. But a random mix of the two is
not *totally* inconceivable. Maybe a statistical mix of the two in
a single [RD]NA loop could actually work most of the time, with
occasional cells having too much of one or the other so they fail
to replicate and die out, but pre-prokaryote cells multiply so
quickly that who cares if a few cells just *DIE*. What do you
think? But I can easily see *most* of the enzymes working
indiscrimantly, giving a really simple means for transition from
RNA-mostly use to DNA-genome-use. Actually now that you've gotten
me thinking outside my earlier box, I see several dimensions of
variation in the just-so stories. Here's one possible sequence:
- Maybe originally there was only RNA, no DNA at all, and all the
cell mechanisms depended on absense of DNA to make things run
smoothly, so any mutation that synthesized DNA (which then got
ambiguously used in place of RNA) got eliminated pretty quickly.
- In some local environment, there was a nutrition crisis, such as
a severe shortage of oxygen, or some metal needed for a key
RNA-synthesis co-enzyme but not needed for the corresponding DNA-synthesis
co-enzyme was in short supply, so that DNA was cheaper to produce
than RNA, which countered the way DNA messed up things, allowing a
mutation to make DNA to survive for the first time.
- As the nutrient-starved clade lived longer in this local
environment, it evolved detoxifying mechanisms, specifically ways
to block DNA out of some key places where only RNA was safe to use,
and/or modifications of RNA-using systems such as the genome to be
more tolerant of occasional DNA monomers mixed with the majority of
RNA monomers.
- As that clade continued to evolve, the genome itself started to
drift to larger fraction of DNA, and the necessary adaptions in
[DR]NA looping mechanisms to deal with that minority but not
infinitesimal portion of DNA monomers. The mechanisms for mapping
to mRNA, however, tightened up the restriction against using any
DNA whatsoever, because the 3*RNA->1*aa mapping was too critically
dependent on RNA (and *not* DNA) being in the mRNA. (Note this
could be true *only* after the genetic code was precise, which was
true *only* after the genetic code happened at all, which is
another reason for believing that polypeptides and the genetic code
were invented long before the RNA->DNA-genome takeover/transition.)
- Eventually the fact that DNA was more stable than RNA, hence it
made a better material for making the genome, won out over the
complexity needed to handle DNA in genome while absolutely
protecting against DNA in mRNA. At this point, the organisms
using significant amounts of DNA in the genome had an advantage
even outside that local nutrient-starved place, so the new
[DR]NA genome m[Ronly]NA system underwent radiative explosion
whereby it replaced the older [Ronly]NA genome system.
- Eventually, co-evolution between the mechanism to assign probability
of the two bases in the genome strands, and the mechanism to deal
with differing coiling characteristics of different relative
proportions of DNA and RNA in the genome, drove the genome to 100% DNA.
(Stability of DNA was the directional driving factor, while dealing
with coiling characteristics was simply following it.)

Robert, has anyone ever told you that you write too long?

Yeah, and in fact sometimes I notice I'm writing a very long
article that covers a lot of points, and I then split the article
into more than one separately-posted part.

But nothing's stopping you from bookmarking my too-long article,
replying to the first part until you get tired, and picking up
where you left off another day. Nothing is forcing you to reply to
my entire article in one single reply of yours.

MetaLilitu: In another newsgroup somebody known as
Lilitu@xxxxxxxxxxx challenged me to balance good and bad remarks I
make. Tonight I've been making effort to answer her challenge.

ObLilitu: My reply above already contains a mix of positive and
negative remarks, including heated dispute over whether it can
properly be said that phenotypes actually replicate, and thanks for
a new idea that allowed to think outside an old box and come up
with an alternate just-so story for the RNA->DNA genome
takeover/transition.

.

Relevant Pages

  • Re: RNA World
    ... Proteins are phenotype. ... clues from present life back into the RNA world and even earlier. ... reason why you think that polypeptides appeared before DNA. ... after proteins rather than before. ...
    (talk.origins)
  • Biochemistry of Genetic Mechanisms
    ... RNA, the same 20 amino acids & ATP) and many identical or nearly ... guanine & cytosine) in DNA. ... protein, acts as a template for the synthesis of RNA in transcription. ...
    (sci.bio.evolution)
  • Re: More shotgunning of ocean microbial ecology.
    ... But OOL and evolution are two very different topics, ... I see you present some additinal reasons why "RNA world" likely came ... before our current DNA world, ... did you specify a sequence of steps involved ...
    (sci.bio.evolution)
  • Re: views of academic scientists
    ... prayer causes some miraculous, 100% immediate healing of all ails. ... Messenger RNA is the RNA that carries information from DNA ... because you said i am against teaching COMPARATIVE Religion. ...
    (talk.origins)
  • Re: views of academic scientists
    ... distorts the data that is trying to be read back into the DNA. ... Messenger RNA is the RNA that carries information from DNA ... acid sequence in the protein that is produced.Many RNAs do ... creationism is religion. ...
    (talk.origins)
Loading