MAGNESIUM AND CATALYTIC RNA 283
site is critical for hammerhead catalytic activity and that another divalent
metal site is probably required as well. In comparing this statement to the
“ rearrangement mechanism ” of Herschlag described in previous paragraphs,
DeRose ’ s research group postulates that the A9/G101.1 site (higher affi nity,
larger effect of added Cd 2+ ) is required for both ground and transition states
while the P1.1 - S Rp metal binding site (lower affi nity, smaller effect of added
Cd2+ ) is required for only the transiently populated transition state. The^31 P
NMR technique described in reference 20 is useful in that the substituted
RNA site is immediately observable with a signal that is sensitive to metal ion
binding if it is catalytically important. One limitation may be that phosphoro-
thioate substitutions may also cause conformational changes that interfere
with catalysis. This is not suspected in the reference 20 work because Cd 2+
addition supports full activity of the phosphorothioate - substituted ribozyme
and substrate complexes.
The third paper in the series compared hammerhead cleavage reactions
stimulated by monovalent and divalent cations.^47 In general, the hammerhead
cleavage reaction is found to be most effi cient in millimolar concentrations of
divalent metal cations. This is not a general truth for all ribozymes because
some, such as the hairpin ribozyme, can operate in the absence of divalent
metal ions. Also, Scott and co - workers described catalytic activity for ham-
merheads in the absence of divalent metal ions provided very high concentra-
tions of monovalent ions are present.^48 The phosphorothioate – Cd 2+ “ rescue ”
experiments described above suggested that certain phosphoryl groups par-
ticipate in metal ion interactions important for catalysis. Other experiments
have indicated participation by metal ions in catalysis carried out by ribo-
zymes. However, metal ions are also known to be important for the folding
and structural properties of DNA and RNA, and it is diffi cult to separate these
structural effects from those only involved in catalysis.
To attempt to differentiate between structural and catalytic effects of metal
ions, Herschlag and co - workers^47 use kinetic experiments, with reactions per-
formed under single - turnover conditions, to test the importance of monovalent
metal ions for catalysis. They used the HH16 and HH8 ribozyme – substrate
complexes in this study. First, they found that for HH8, fi rst - order dependence
on mM Mg 2+ concentrations was found up to a saturation rate of 1.3/min at
approximately 10 mM Mg 2+. In contrast, the catalytic cleavage rate in LiCl was
very slow below 0.5 M and showed second - order dependence up to a rate of
0.24/min in 4 M LiCl. When abasic (C3X, U7X, G10X, and so on, for HH16) and
other mutations (C3U, C17A, and so on, for HH8) were introduced into HH16
and HH8, kinetic experiments like those described previously in the Herschlag
group ’ s work (references 41 and 42 ) showed that reaction rates in 10 mM Mg 2+
and 4 M Li + were mostly analogous. One point of difference was found for the
abasic replacements at positions G10.1 and C11.1 — that is, G10.1X and C11.1X.
While the rate in 10 mM Mg 2+ decreased relative to wild - type HH16 behavior,
the rate in 4 M Li + increased from the wild - type rate. Since G10.1 is an
important divalent metal ion ligand, the researchers concluded that some