MAGNESIUM AND CATALYTIC RNA 285
phate linkage to exist, although the one hydrogen bond seen in the crystal
structure pairing U 16.1 and A 15.1 was broken. Minimization of the starting struc-
ture provided a solution that required stem I and II helices to become approxi-
mately parallel with the stem I helix pushed up into a higher position compared
to its crystal structure position (see Figure 4 of reference 49 ). These authors
then attempted to fi t the best plausible structure to other experimental data.
For example, crosslinking experiments between residues in stems I and II had
concluded previously that the distances between 2 ′ - OH groups of residues 2.6
(stem I) and 11.5 (stem II) was set at 16 Å or less. In the crystal structure, this
distance was approximately 11 Å. However, if this distance and associated
stereochemical constraints were imposed on the energy - minimized structure
obtained in this molecular modeling study, the stem I helix would be required
to unwind fully — an energetically unlikely event. The reference 49 authors
believe a large - scale rearrangement in the hammerhead ribozyme into a tran-
sition state as described by Herschlag ’ s group is unnecessary, citing the ability
of a hammerhead cleavage reaction product to be formed within a crystalline
lattice as they have shown in their previous work (reference 44 , PDB: 379D
and reference 39 , PDB: 301D). They further cite X - ray crystallographic data
on a ribozyme – product complex obtained in 2000. This work is discussed in
the following paragraphs.
Scott and co - workers next used crystal lattice trapping and X - ray holo-
graphic reconstruction to capture and visualize a hammerhead enzyme - product
complex.^52 The data have been deposited in the Protein Data Bank as PDB:
488D. First, it should be stated that, upon cleavage, hammerhead ribozyme –
complex crystals become disordered and the quality of X - ray diffraction dete-
riorates to the extent that structures of the ribozyme – product complex cannot
be solved. These authors circumvented this problem by mixing a small percent-
age of modifi ed (or inhibited) RNA substrate into the crystallization solution,
resulting in crystalline lattices of suffi cient rigidity to trap the ribozyme –
product complex in a matrix of unreacted modifi ed ribozyme – substrate
complex. The collected data were then manipulated using X - ray holographic
methodology, resulting in clearer and more reliable identifi cation of structural
elements and conformational changes. Data sets were collected on crystals
shown (by HPLC UV absorbance integration) to have undergone 40% and
60% cleavage to product. The data were refi ned against a previously deter-
mined initial state structure (PDB: 299D) and further manipulated using the
X - ray holographic procedure (which was itself tested using the initial state
structure). The conformational changes were observed in both the 40% and
60% cleavage - to - product crystals.
The fi nal refi ned structure showed a 2 ′ ,3 ′ - cyclic phosphate product at the
C 17 ribose position and a C 17 nucleotide that had moved dramatically and
become almost perpendicular to the Watson – Crick faces of conserved nucleo-
tides G 5 and A 6 in the catalytic pocket. Interactions of functional groups of G 5
and A 6 with the product and C 17 included (1) a hydrogen bond formed between
the exocyclic amine (N 6 ) of A 6 and the cyclic phosphate nonbridging oxygen