Text 670, 70 rader
Skriven 2004-11-06 06:11:00 av Catherine Woodgold (1:278/230)
Ärende: Beginning of Transcriptio
=================================


Today I succeeded in imagining intermediate steps in the
evolution of transcription of proteins.  Of course, I don't
know whether that's how it happened.  Or whether others have
described it similarly in the past.  I'd be interested in
suggestions of books that talk about this sort of thing.

I imagine that the first life involved replication of
RNA but not manipulation of proteins.  I imagine it could
have progressed through the following steps:

-- The RNA is shaped to attach to short, commonly-available
proteins which increase its stability, or its ability to
cause the disintegration of other RNA strands, or its ability
to replicate.

-- The RNA is shaped to attach to two short, commonly-available
proteins, which it attaches together to make a protein that's
useful to it.

-- The RNA not only attaches to proteins, but lets them go again
after joining them together, leaving them to perform useful
functions.

-- The RNA is shaped to attach to individual amino acids which
it joins together.

-- A piece of RNA might have the same base repeated several times
at its ends, and might create an enzyme that tears bases off the
ends of RNA strands but only if they have some other base near the
ends.

-- For one of the amino acids which is rarer and/or more difficult
to form an RNA receptor for, the RNA uses other amino acids to form
a receptor.

-- Having developed rather complex mechanisms for specifying the
various amino acids, the RNA forms elegant, simple control
procedures to specify when to use those mechanisms, thus
only having to have one copy of the mechanism specifying amino acid X
even if X is used multiple times.

-- Up to here, a given bit of RNA could specify a given protein,
directly or indirectly.  But then it begins to rely on
some proteins already being there  -- like compilers for the
C programming language that are themselves written in C.
The first one couldn't have been written in C, but it's OK
to lose the copies of that first one as long as there's always a compiled
compiler around.  Similarly, the original mechanism for specifying
amino acid X using only other amino acids could be lost, and 
a method that uses X in forming the receptor for X could remain.
From then on, the RNA by itself would be helpless;  it would
have to have a copy of the protein it makes, too.  At first, perhaps
the original method for only one of the amino acids is lost;  the
others are still specified completely, if indirectly, from the RNA.

-- I'm a little hazier here, but:  methods for specifying
the proteins more and more concisely would develop,
eventually settling on the optimal 3 bases per amino
acid code.
--
Cathy
---
ū RIMEGate(tm)/RGXPost V1.14 at BBSWORLD * Info@bbsworld.com

---
 * RIMEGate(tm)V10.2á˙* RelayNet(tm) NNTP Gateway * MoonDog BBS
 * RgateImp.MoonDog.BBS at 11/6/04 6:11:43 AM
 * Origin: MoonDog BBS, Brooklyn,NY, 718 692-2498, 1:278/230 (1:278/230)