All of this is just speculation. When everything is said and done the
most implausible explanation could be the one that was used because we
are only dealing with one chance occurrence. I don't follow origin of
life research so this is only my opinion. I have not seen this model
published, so if I am stealing someone else's idea it is just out of
ignorance.
A lot of things have to happen before DNA can replicate. One hint is
that DNA replication requires an RNA primer to be laid down before
initiation of DNA replication.
The current thinking is that RNA came before DNA. You still need a
lot of things to make RNA even if RNA could be a genetic material.
Something made the nucleotides to make the first polymers of RNA.
These somethings were likely not RNA polymers. I don't think that
anyone knows what the first self replicating molecules were made of.
They had to make more copies of themselves, and do it with some
accuracy, but inaccurately enough to create a diversity of self
replicating entities.
We know that such complex molecules would have had a complex three
dimensional surface. This surface could have served multiple
catalytic functions, so my notion is that groups of self replicating
units would evolve so that each had secondary functions such as making
purines or pyrimidines from whatever was available as well as making
more copies of themselves. Once something like a lipid bilayer could
form (you probably need self replicators with the secondary function
of making lipids) you can get bags of self replicators that would
likely just bleb off when the bags (proto cells) got too large.
Somewhere in this type of mess of bags of independent self replicators
with multiple secondary functions you would evolve the first RNA
producing bags of self replicators. An RNA polymer has the property
that it can be used to make a copy of RNA or DNA that can be used to
make an exact copy of what was copied. DNA may be a better template
to make RNA from, so there could have been a fairly rapid evolution to
making DNA templates for RNA production.
tRNAs come in because they can be used to make peptides. So if the
first self replicators used peptides to make themselves or do
something useful, crafty self replicators might evolve a means to use
tRNAs to make peptides more efficiently. The original tRNAs would
likely replicate as RNAs, but probably lacked the anticodon and were
just used to make peptides using some other template (probably the
surface of some other self replicator). If these tRNAs were made off
a set template they could evolve in their sequence and specific tRNA
sequences could have been acylated with specific amino acids expanding
the kinds of peptides that might be made with some degree of
specificity.
Ribosomal RNA likely evolved as a more efficient means of using tRNAs
to make peptides. My guess is that once DNA had appeared on the scene
to be used as a template to make RNA polymers like tRNAs that the DNA
could mutate and increase in size, and some of the RNA made might not
serve any function, but could be used as random template. The genetic
code could start evolving as tRNA might have one part of themselves
coopted as an anticodon that could be used to make peptides based on
RNA nucleotide sequence template. It looks like the first code was
two bases instead of three, but there was either a spacer nucleotide
or there was an expansion phase from two to three nucleotide codons.
Making peptides off these crude templates would be going on in the
background of the bag of multiple self replicators with secondary
functions, and the peptide making machinery might be selected for
because they made raw material for the self replicators or did useful
things in the proto cells.
The important part of this scenario is that it is the bunch of self
replcators with secondary functions that are heart of the proto cell.
Once DNA arrives and RNA templates can be made that do things like
tRNAs or ribosomal RNA. They could also be used as mRNA templates to
make larger more complex polypeptide proteins. Once longer
polypeptides could be replicated with accuracy the polypeptides could
evolve functions that replace the secondary functions of the original
self replicators. This creates a path from the original self
replicators to genetic information in the form of DNA.
1. Self replicators evolve that also have secondary functions due to
their complex surfaces.
2. Groups of such self replicators evolve the ability to make
glucose, amino acids, RNA nucleotides and RNA polymers etc.
3. Short RNA polymers could be replicated off RNA or possibly DNA
templates and set sequence tRNAs could evolve. It is the start of
nucleotide genetic material. The original RNA and DNA polymerases
could be self replicating units with that secondary function.
4. The first tRNAs would not have anticodons, but be used by the
protocells self replicators to make peptides. They would start out as
acylated RNA molecules that were used to transfer amino acids to
peptide chains.
5. With stable but not perfect replication tRNA sequences could be
selected for that could be charged with specific amino acids and could
be used to more accurately reproduce needed peptides. My guess is
that it could be part of the recognition sequence to specifically
acylate a tRNA that might be coopted as the initial anticodon.
6. Ribosomal RNA evolves that can be replicated off a stable RNA or
DNA template that can use the tRNAs to make peptides. These ribosomal
RNAs would likely begin to replace some of the original self
replicators that had been doing the same thing.
7. RNA would start to be used as mRNAs to more accurately replicate
polypeptides that would be selected to have functions already being
done by self replicating units with those secondary functions. The
existing specifically charge tRNAs could evolve anticodons used to
read the mRNA template. The RNA/DNA genetic templates would likely
become a more accurate means of storing needed information.
8. The genetic code would evolve in this fashion replacing the
functions of the individual self replicators as proteins evolve that
duplicate the existing functions. With the self replicators as backup
the genetic code could evolve in steps until it was good enough to not
just produce useful peptides for the cell, but replace the individual
self replicators with more efficiently stored instructions to do what
needs to be done using protein polypeptides instead of the original
self replicators.
Things would be much more complex than this summary, but it is a way
that the proto cell could slowly evolve the functions it needs to be a
viable reproducing cell, and evolve the means to switch over from the
original self replicators to DNA an RNA based genetics. It doesn't
have to be linear. Different bags of self replicators could be fusing
and blebbing off as the genetic code was evolving bringing in their
own templates and tRNA anticodons. The code could evolve as different
protocells contributed what had evolved in their little part of the
biosphere and contribute what RNAs and peptides that they could
accurately code for.
Ron Okimoto