Roland Hirsch, e-mail: Roland.Hirsch@science.doe.gov
Comments (unpublished)

This news article in Nature came up elsewhere, and here is part of my
response. Note in particular that the news article makes substantial
extrapolations from what was said in the actual paper in the Journal of the
American Chemical Society (JACS).
=================================
Here are the reaction conditions as given in a footnote to the paper:
"General reductive amination conditions and product analysis. Templates, 2
mM; substrate, 4mM for tetramer, 8 mM for dimer, 16 mM for monomer; 20 equiv
of NaBH3CN was added at room temperature" (the mM and equiv refer to
concentrations of the reagents). The reagent, NaBH3CN, is required to change
the chemical form of the DNA strand to a reactive one. It is most unlikely
that this, or any other imine-forming reagent, was present in useful amounts
in prebiotic conditions. Further, the reaction requires the absence of
common metal ions that must have been present under prebiotic conditions and
would have disrupted this type of reaction. Examples are copper, iron, zinc,
manganese.
I do not understand how the author of the news article drew the
conclusions about abiogenesis. The JACS article does not refer to this at
all. Indeed it clearly is a designed, not a chance and natural law type of
experiment: "The reaction was designed to proceed via intermediate
template-bound imine", speaking about the critical step in making the
synthesis work.
Then the full last paragraph of the JACS article says: "The ability
to read a DNA template sequence and chain length specifically represents a
critical extension of biology's template-directed syntheses, represented by
its Central Dogma. The step-growth kinetics outlined here are clearly
different than the chain-growth processes selected in biology, and it should
be possible to exploit strengths and limitations of other synthetic
reactions for reading biopolymers. Knowing that amine-linked DNAs are
catalytically active phosphodiester ligases suggests that these materials
may be further extended to access synthetic replication and selection
strategies. Therefore, the accurate synthetic translation of information
encoded in biological macromolecules, exploiting more standard
polymerization reactions, should enable the preparation of a diversity of
monodisperse, sequence-specific functional materials for synthetic biology."
Note how many times the authors state that what they have done is
different from what biology does in nature: "critical extension of biology's
template-directed syntheses"; "clearly different than the chain-growth
processes selected in biology". Indeed, if DNA replicated this way under
prebiotic conditions, why wasn't this more efficient way of replication
(requiring no proteins with the enormous energy spent to make them) retained
in cellular biochemistry?

POWRÓT