Hi Jacob-
What an intriguing proposition. I can think of multiple reasons why
such a system would not exist, but there is a mechanistic one which is
most fundamental, having to do with the nature of the genetic code.
Say that there is a cellular machine which would unfold a protein and
expose it to some sort of reading system (already a hard problem).
There is now the issue of transforming the amino acid information into
nucleic acid information. For simplicity let's assume that this
system only uses one codon per amino acid, bypassing the degeneracy
problem.
How would the cell then read off the amino acid sequence? It seems
the simplest solution would be analogous to translation, i.e. having
adaptor molecules analogues to tRNAs which would guide an enzyme that
synthesized the nucleic acid. Otherwise, one would have to invoke the
idea of a single enzyme recognizing every amino acid, which seems
impossible to me.
As we know, the problem of protein-protein recognition is relatively
complex. At a minimum, one would need 20 adaptor proteins to
recognize the 20 canonical amino acids: however, it seems unlikely
that recognition of a single amino acid would be robust enough to
select for the correct adaptor molecule.
So, let's say instead these adaptors recognize 2 amino acids at a time
(still probably not robust enough). Then, one would need 2^20
adaptors, already a far greater number of gene products than that
present in any genome than I know of...
It might be tempting to draw an analogy between this system and the
immune system, where an incredible diversity is generated from a small
number of genes. However, diverse immune proteins all take the same
input sequence (say antigen recognition) and lead to a single
response, whereas this system has a 1 to 1 correspondence between
inputs (protein sequence) and outputs (nucleic acid sequences), and
thus there is no way that a randomization system could generate the
required diversity.
Cheers,
-Greg Alushin
Nogales lab
UC Berkeley
On Sep 6, 2010, at 7:12 PM, Michael Thompson wrote:
Jacob,
The idea is enticing, but don't forget that there are multiple
degenerate codons for a given amino acid. Once the protein is
synthesized, the specific codon information is lost.
I think that's a fundamental problem.
Keep the ideas coming,
Mike Thompson
----- Original Message -----
From: "Jacob Keller" <j-kell...@fsm.northwestern.edu>
To: CCP4BB@JISCMAIL.AC.UK
Sent: Monday, September 6, 2010 6:36:14 PM GMT -08:00 US/Canada
Pacific
Subject: [ccp4bb] Reverse Translatase
Dear Crystallographers,
does anyone know of any conceptual reason why a reverse translatase
enzyme
(protein-->nucleic acid) could not exist? I can think of so many
things for
which such an enzyme would be helpful, both to cells and to
scientists...!
Unless there is something I am missing, it would seem to me
conceptually
almost impossible that it *not* exist.
Best Regards,
Jacob Keller
*******************************************
Jacob Pearson Keller
Northwestern University
Medical Scientist Training Program
Dallos Laboratory
F. Searle 1-240
2240 Campus Drive
Evanston IL 60208
lab: 847.491.2438
cel: 773.608.9185
email: j-kell...@northwestern.edu
*******************************************
--
Michael C. Thompson
Graduate Student
Biochemistry & Molecular Biology Division
Department of Chemistry & Biochemistry
University of California, Los Angeles
mi...@chem.ucla.edu
