Path: utzoo!attcan!uunet!husc6!bloom-beacon!tut.cis.ohio-state.edu!rutgers!sunybcs!boulder!pell
From: pell@boulder.Colorado.EDU (Anthony Pelletier)
Newsgroups: sci.bio
Subject: Re: Nature Articles. Anyone read them?
Keywords: amino acid tRNA charging code
Message-ID: <7169@sigi.Colorado.EDU>
Date: 18 Jul 88 18:23:31 GMT
References: <1628@runx.ips.oz> <6209@bloom-beacon.MIT.EDU>
Sender: news@sigi.Colorado.EDU
Reply-To: pell@boulder.Colorado.EDU (Anthony Pelletier)
Organization: University of Colorado, Boulder
Lines: 48


(sorry, I tried to mail this, but it bounced--just hit "n" unless
you are the original poster)

I think you got a couple of things mixed up here.  The "two-base code"
is a bit of a misnomer.  They only did the work for one tRNA; unlike
translation, information for which must be constrained in a physical
space (the A nd P sites of the ribosome), there is no reason that all
tRNA sythetases need use the same number of contacts or even contacts
on analogous parts of the molecule.  One might imagine, for example,
that the variable loop might be important in some cases, not in others.
John Ableson has found that one cannot construct efficient amber suppressors
for all tRNAs.  That is, altering the anticodon to CUA does not mean
that it will work.  I have not heard these data for a while, but some of the
inefficient tRNAs were so because they were not charged properly.
Thus, even though the rest of the molecule is unchanged, the cognate
synthetase does not recognize and/or transfer the AA.
Also, recall that there is a tremendous amount of structure-information
in tRNA.  The possibility for allosteric effects on a binding pocket
by changes elsewhere in the molecule are strong.

Put this in an evolutionary context.  It is imagined by many that tRNAs were
self charging, at first--it is clear that RNA can be catalytic and it is clear
now that specific binding sites for amino acids can exist in RNA (See the 
article by Mike Yarus in the 6/24 Science).  If that is true then proteins
were added later.  Why should proteins developing contacts with different
tRNAs all interact with the same spot?  There are plenty of good reasons
why the number of spots will be limitted.  But some variation should be
expected.

Regarding functions of DNA:  In NO WAY could it be said that a function
of DNA is to "replicate itself" as you say.  DNA does not do that.  True,
self complimentarity does provide for the mechanism of replication.
DNA's main role is the storage of information.  What I think schimmel
meant was that his tRNA does store information.  In the RNA world (before DNA
and protein) imagined by most scientist, catalysis and information storage
were performed by  small RNAs.  The ancient "code" was the way pre-tRNAs
interacted with amino acids.  Later, after protein synthesis was developed a
bit, this was helped along by proteins (until you can code for proteins, you
can't really use them).  Anyway, personally, I think Schimmel is stretching
the importance of his work a bit.  One professor in our building who
shall remain nameless put the article on his notice board with the caption
"what an Asshole."  I have to be more kind since Schimmel gave me
a clone I needed for a rather important experiment.

There were other points I was going to make, but...back to experiments.

-tony