On Sat, 20 Dec 2003 22:05:09 +0000 (UTC), [email hidden]
(TomHendricks474) said:<< >I'd like to ask a very specific question.
Quoted message said:IF a single strand of nucleotides could h-bond to the carboxyl group;
1. Wouldn't only G bond to the COO- version (with 2 of 3 h-bonds)
2. Wouldn't A or U bond to the COOH version with 2 h-bonds (and G or C but with only 2 of 3 bonds)
I'm not a chemist and I don't know. Even if true though, if nothing else is holding them together
but 2-3 h-bonds, it would be a very momentary connection. That is not enough h-bonding, especially
with the temperatures you envision.
TH Well your reasonable concerns may well be the Achilles heel of this idea. But it would solve
many problems if my idea was correct, and there is this:
Joyce and Orgel; '93 show some triple stem-loop stuctures that RNA molecules can assume.
They look like proto tRNA's with 3 loops and two loose stem areas - though the loose stem areas
seem highly protected by the loops, and in all cases they are next to each other and look to be
bonded except for an end base.
If we suggest this 'evolution' of tRNA shape
1. folded single strand with two loose ends ( or even a string with knots in it where between the
knots there are 'naked' bases.
2. Joyce and Orgel single strand with triple loop and two loose ends near each other
3. tRNA today.
Now for our purposes lets stick with #2 Then it seems that the hydrogen bonds I suggest between a
'naked base and the carboxyl group of an aa, may be highly protected - at least as protected as
those hydrogen bonds within some of these same proto RNA structures.
Quoted message said:
A rule of thumb is that h-bonds are 5% the strength of covalent bonds. Much of enzyme chemistry
just involves bringing two or more molecules close together that then naturally react (changing
the nature of the covalent bonds) but with no energy required for the reaction. Without the
enzyme, the reaction isn't seen because the molecules are never close together. Enzymes, of
course, hold the molecules by h-bonds. This was seen in Fox's experiments and is seen in RNA labs
today, where just having random polypeptide-like polymers and metal ions in the mix, increases
catalytic activity. Of course it is a long way from natural uninformed enzymatic substrates and
the informed protein enzymes of today. But I would think some sort of naturally occuring enzyme-
like substrates were involved from the very beginning. William L Hunt
It may well be that an enzyme you suggest was a key, yet perhaps we shouldn't place all our
existence on it - hold it instead as an ace in the hole - and come up with some overriding and
bigger overall scenario that somehow shows reason for the symbiosis between aa's and a folded
nucleotide strand.
TH
Comment?
I think maybe you have become too fixated on the tRNA-aa connection. I am going to completely
change the direction and describe in outline a scenario I have seen and like that explains what
tRNA and ribosomes might have been doing before this connection was made. You know I think this
connection came late in the RNA world timeline when there would have been plenty of ribozmes to
handle a covalent aa connection.
Scenario: Just as the machinery that makes protein in the modern cell (tRNA-ribosome-aaRS complex)
might be thought of as the key piece of machinery today, in the rna world it would have been the
machinery that replicated the informational rna. There are some hints that tRNA was involved in
this. One hint is the use of a tRNA as a primer by rna retro viriuses. A tRNA must be attached to
their rna before their rna replicase can make copies. Some, like HIV, come pre-primed with a tRNA-
like sequence already on the end of the rna. Maybe this is an RNA world fossil, maybe not. Some
think there is no evidence of what this RNA world replicating machine looked like, other think it
is looking us in the face...it is the tRNA-ribosome, only today modified for a different function
with AARS complexes added. Today DNA replicases place single nucleotides one at a time with
proofreading. But in the rna world maybe there was a "goldilocks" effect. Presumably there was a
pool of small various length nucleotides, but placing single or even double length nucleotides may
not have had enough h-bonding to hold them in place. Placing longer lengths has other problems. It
may difficult to find exact matches of longer length sequences (you have to look through 4 to power
of n for an n length sequence), and even worse, sequences that are not quite perfect matches might
be accepted because of the strength of the h-bonding. So matching and placing 3-length nucleotides
may have been optimal for the machine. Single strand rna would enter the machine and double strand
would exit. The tRNA would provide the 3 nucleotides on its anti-codon. If a match was found, the
anti-codon 3 base section would be snipped by a ribozyme (no problem conceptually as many of these
sort of ribozymes exist today). The tRNA with its anti-codon removed would then be recycled.
Possibly the stem had a mirror anti-codon sequence that would allow the tRNA to have its own
template for replacing this 3 nucleotide sequence. The details might be messy, how many species of
tRNA for instance? But the scenario does provide for the evolution of a ribosome with A and P sites
and a 3-base rachet movement mechanism before any use of aa. The tRNA has and needs no connection
to aa yet. I have seen proposals of how an aa connected to the tRNA stem might be advantageous
later in this scenario, sort of how it might come to be involved to bridge the gap to the modern
function of tRNA-ribosome-AARS. One other thing I think you may not see the way I do is 3rd
position wobble. I think you may still see this as somehow intrinsic and unavoidable in the third
position. I have previously explained the view that Watson-Crick pairings are being enforced by the
ribosome and it could easily have evolved to enforce a Watson-Crick pairing for the third position
also (if there was any selective pressure to do so). In the above scenario the ribosome must be
enforcing WC pairings on all three postions or the copying would just be too error filled. So in
this scenario, (1)early in the RNA world there would be very high error filled copying without a
ribosome. (2)Some more evolution, use your model to fill in the details of how we get from 1 to
3.(3)Late in the RNA world the tRNA-ribosome would have evolved to do accurate copying (by RNA world
standards) of informational rna, with the tRNA binding with Watson-Crick pairings in all three
positions.
(4) A modified machine takes up a new function with tRNA-aa connection building polypeptides but
keeping its basic mechanisms. Wobble in third position evolves under selection pressure to
reduce number of tRNA species. (5)Protein replicases completely replace the original machine
leaving only the machine that does the later function of building polypeptides. William L Hunt
