Phosphoserine as a coded amino acid

From: Perplexed in Peoria (jimmenegay_at_sbcglobal.net)
Date: 03/25/05 

Date: Fri, 25 Mar 2005 11:33:18 -0500 (EST)

This Week in SCIENCE, Volume 307, Issue 5717, at:
http://www.sciencemag.org/content/vol307/issue5717/twis.shtml

---------------------------------------------------------------------------
Biochemical Prehistory
---------------------------------------------------------------------------

The transition from an early RNA-based biochemistry to one that was
(and is) based on proteins required a set of components that could
convert the nucleic acid code for amino acids into the actual amino
acid. The set of aminoacyl-transfer RNA (aa-tRNA) synthetases does
just that, attaching the amino acid to its cognate tRNA, which is then
used by the ribosome to translate the genetic code into proteins. There
is, however, evidence that some of the 20 canonical amino acids are
relative latecomers, and (p.1969) show that cysteine may be one of these
add-ons. Archaea that lack the aa-tRNA synthetase for cysteine rely on
an alternative pathway (likely a relic) in which phosphoserine is attached
to tRNA and then enzymatically converted in an anaerobic, pyridoxal
phosphate-dependent reaction to cysteinyl-tRNA.

--------------------------------------------------------------------------

My comment. This is, of course, further confirmation of the hypothesis
that at least some evolution of the genetic code took place as a result
of the development of some biosynthetic pathways for amino acids that
work on amino acids precursors already attached to the tRNAs.

This paper is not much of a surprise - it was already known that seleno-
cysteine is made from phosphoserine on tRNA.

I have believed for some time that phosphoserine was once a standard
amino acid which had the UG_ block to itself, and perhaps also the
UC_ block as well. Note that phosphoserine is an essential component
of many proteins, but that the current way of inserting it into a
protein is to phosphorylate serine residues AFTER translation. But
doesn't it seem odd that the phosphorylation is almost always required
to ACTIVATE the enzyme - almost never to turn it off. This is a
clue that the enzymatic activity of the phosphorylated enzyme is
older that the process of post-transcriptional phosphorylation.

The same argument applies to phosphothreonine. Every biochem textbook
will show phosphothreonine as the biosynthetic precursor of threonine.
But few will mention phosphoserine as the precursor of serine, cysteine,
and tryptophan. It is though, if you take 3PG as the starting material.
And my version of J. T. Wong's theory does insist on this pathway as
the original one.

Incidentally, one of the reasons why I think that phosphoserine was still
present in the genetic code when translation was developed is that it
is difficult for me come up with a plausible tRNA-based synthesis for
aspartate and glutamate in code block GA_. So we need phosphoserine
as the only anionic amino acid until Asp and Glu can be added to the
code. It is ironic that in my theory, Asp and Glu are the "parents"
of whole "families" of amino acids, but they are not themselves members
of those "families".