DNA Structure: Puzzle Number 12 (Discussion)

clive delmonte clived at ndirect.co.uk
Mon Dec 21 06:14:18 EST 1998

I have been cheerfully chipping away at the internal contradictions and
inconsistencies of DNA crystallography, undermining DNA fibre
crystallography (for example, Puzzles 2, 4, 5, 6 & 9) and dwelling on the
circularity of reasoning in so many X-ray diffraction studies of
oligonucleotide crystals (for example, Puzzle 12).   Future Puzzles would
address issues in heavy atom, MAD and synchrotron X-ray diffraction of
oligonucleotides, but Dr Harry Powell's posting warrants an early response,
however abbreviated.

Although difficult to include all pertinent data in one Puzzle, I should
have made it clearer in Number 12 that my reference to best resolutions down
to about 0.15nm related to the sort of studies I was discussing there.

There could be many points to be made in response to Dr Powell and I ask the
indulgence of the newsgroup in selecting at least a few of them for
inclusion here.

First, we still have no answer to the Puzzles 1, 2, 3, 4, 5, 6, 7, 8, 10 and
11 already set out in this series, all of which relate to duplex DNA.  Is
there anyone who would care to offer an explanation based upon a
plectonaemic double helix ?

Next, the structure of quadruplexes has been raised.   One, at least, has
been studied by both X-ray diffraction and NMR, and major backbone
differences have been reported (J. Feigon et al., in Structural Biology,
ISBN 0-940030-42-X (1994) Pages 127 - 134, and references therein).   What
is the way forward here ?

Third, what real, actual, direct experimental evidence is there that
crystallising short lengths of oligonucleotides in vitro actually produces
the same structures as exist in high polymer DNA synthesised and assembled
by proteins in vivo ?    I know of none.

Fourth, moving to ever higher resolutions in synchrotron oligonucleotide
diffraction only makes the third point even more pressing.

Oligonucleotide crystallographers have proved themselves very unwilling to
consider, and very resistant to, the notion that their structural work can
be usefully informed by other studies of high polymer DNA changes in fibres
(for example, Puzzles 2, 4, 5, 6 and 9), or with methylating agents (for
example, Puzzle 7), or on nucleosomal histone cores (for example, Puzzle 8,
10, 15 (to come)), or from general chemistry (for example, Puzzle 3).

Just to summarise; for polynucleotides, a very large number of
crystallographic fibre studies conflict with each other when viewed overall;
a large number of oligonucleotide crystallographic studies are circular and
reproduce the structures which were fed in; and for the relatively few
oligonucleotide crystallographic structural studies at very high resolution,
which are claimed to be ab initio, there is no direct experimental evidence
at all that the structures exist in vivo.

Biophysics needs and deserves better than this.

The true side-by-side duplex, high polymer DNA structure found
experimentally by Lee et al. (Puzzle 1, ref. 1), and identified
theoretically in even earlier work reported elsewhere, offers a coherent,
comprehensive explanation of all the Puzzles.

Is there any explanation available based on a plectonaemic model ?

Clive Delmonte

>--Previous message from
>Dr Harry Powell, MRC Laboratory of Molecular Biology, MRC Centre, Hills
>Road, Cambridge, CB2 2QH

More information about the Xtal-log mailing list