12072 J. Am. Chem. Soc., Vol. 119, No. 50, 1997
if nLG,TS > nNu,TS then R ) Roxocarbenium
Berti and Schramm
endo and 2′-endo reactants (ribosyl rings, in general, have weak
conformation preference in solution and would be expected to
exist in a combination of states75). Only the 3′-endo reactant
gave calculated KIEs that matched the measured KIEs. NMR
spectra of NAD+ and nicotinamide mononucleotide in solution
indicate an unusual distribution of conformers skewed toward
the 3′-exo and 2′-exo conformations,76 with a predominance of
the 3′-exo conformer.77 These conformers have not been
observed for pyridine nucleotides in the solid state but may be
favored in solution because of electrostatic interactions between
the anionic phosphate and the cationic aglycone. Like the 3′-
endo conformer, the lone pair orbital of the ring oxygen is nearly
antiperiplanar to the C1′-N1 bond in the 3′-exo and 2′-exo
conformations. The bond lengths in the ribosyl ring of these
conformers, and therefore the resulting KIEs, are expected to
be closer to those produced using the 3′-endo conformer as the
reactant model. In locating appropriate transition state struc-
tures, a model based on the 3′-endo conformer was required to
match the measured KIEs, while the 2′-endo conformer as the
reactant model did not provide agreement for any combination
,
+
(Rreact - Roxocarbenium)[(nLG - nNu)/nreact](1 - OC)
where nr5P is the C1′-O1′ bond order of the ADP-ribose
molecule found in the X-ray crystallographic structure of liver
alcohol dehydrogenase (PDB66 structure: 5ADH). The angle
R
r5P is the “ N1-C1′-H1′ bond angle” 67 ) 38.1°, optimized
for the structure of the R-ribose-5-phosphate moiety of ADP-
ribose from the 5ADH structure at the RHF/3-21G level of
theory using the program Gaussian 94.68
CTBi. A computer program, CTBi (Cartesian coordinates
to Transition state structures to BEBOVIB input using internal
coordinates) was written to generate trial transition state models.
It reads in structures in Cartesian coordinates, parses the
structures to find leaving group and nucleophile atoms and to
identify ring closures, converts the reference structures to an
internal coordinate system, then generates a series of transition
state structures based on a list of nLG,TS and nNu,TS values, and
writes out the structures in polar coordinates for input into
BEBOVIB.
of nLG,TS and nNu,TS
.
First Reference Structure: The Reactant State. The
reactant state model of NAD+ was from the X-ray crystal-
lographic structure of the lithium salt of NAD+,69 truncated to
nicotinamide mononucleotide with a monoanionic phosphate.
Hydrogens were added and optimized at the RHF/6-31G** level.
Unlike other (e.g., adenylyl) nucleotides,70,71 the conformation
of the ribose ring attached to the nicotinamide has a significant
effect on bond orders. Since KIEs are a function of the
differences in vibrational modes between the reactant and
transition state structures, the reactant model is as important in
determining the predicted KIEs as the transition state. In a
comparison of five X-ray crystallographic structures of NAD+
in the 3′-endo conformation72 versus 4 structures in the 2′-endo
conformation,73 the 3′-endo structures had lower bond orders
by 0.17 for C4′-O4′ and C1′-C2′, 0.15 for C1′-N1, and higher
by 0.03 for C1′-O4′. These differences may arise partly from
the anomeric effect in the 3′-endo conformation, where interac-
tion of the lone-pair electrons of the ring oxygen antiperiplanar
to the C1′-N1 bond would lead to a lengthening of the C1′-
N1 and C4′-O4′ bonds and shortening of the C1′-O4′ bond.
In the 2′-endo conformation, the angle between the C1′-N1
bond and the lone pair electrons is wider, leading to a weaker
anomeric effect. Similar effects have been observed between
axial and equatorial acetals and O-glucosides.74
The 3′-exo and 2′-exo conformations would be expected to
be more reactive than the 2′-endo conformer. Jones and Kirby78
showed a correlation between C-O bond length and reactivity
and estimated that the free energy of activation for unimolecular
heterolysis decreased with increasing reactant bond length by
250 kcal‚mol-1‚Å-1. Assuming that the same effects obtain
for C-N bonds and that the energies of activation are of the
same magnitude, then given an average difference of ∆rC1′-N1
) 0.05 Å, the free energy of activation for 3′-endo conformers
would be 12.5 kcal‚mol-1 lower than for 2′-endo conformations.
However, when the different conformers are in rapid equilibrium
in solution, the difference in reactivity will have no effect on
the KIEs.
Second Reference State Structure: The Oxocarbenium
Ion + Nicotinamide. The starting point to establish the
structure of the oxocarbenium ion model was the X-ray
crystallographic structure of ribonolactone.79 The carbonyl
oxygen was changed to hydrogen, a phosphate group derived
from the NAD+ structure was added, and all atoms were
minimized at the RHF/6-31G** level, with minimal bond bend
and torsional constraints added to prevent the anionic phosphate
from forming a covalent bond with C1′. The nicotinamide
molecule was the X-ray crystallographic structure,65 with
hydrogens added and optimized at the same level of theory.
Hyperconjugation at H2′. The Hartree-Fock method used
for structure optimization is inherently unable to reflect the
electron correlation effects necessary to model the hypercon-
jugation that is observed in the â-secondary (2′-3H) KIE.
Hyperconjugation arises from the interaction of the occupied
π-symmetry orbital of the â-carbon, C2′, with the developing
vacant p-orbital of the anomeric carbon as the leaving group
departs80 (Figure 2). This leads to a lengthening of the C2′-
In attempting to locate the transition state, no combination
of nLG,TS and nNu,TS could be found that matched the measured
KIEs when using a 2′-endo reactant or a combination of 3′-
(66) Bernstein, F. C.; Koetzle, T. F.; Williams, G. J. B.; Meyer, E. F.,
Jr.; Brice, M. D.; Rodgers, J. R.; Kennard, O.; Shimanouchi, T.; Tasumi,
M. J. Mol. Biol. 1977, 112, 535-542.
(67) The angle between the C1′-H1′ bond and the normal to the plane
defined by the atoms O4′, C1′, and C2′.
(68) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.; Petersson, G.
A.; Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski,
V. G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski, J.; Stefanov, B. B.;
Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen, W.;
Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; Head-
Gordon, M.; Gonzalez, C.; Pople, J. A. Gaussian 94, ReVision C.2; Gaussian,
Inc.: Pittsburgh, PA, 1995.
(74) Briggs, A. J.; Glenn, R.; Jones, P. G.; Kirby, A. J.; Ramaswamy,
P. J. Am. Chem. Soc. 1984, 106, 6200-6206.
(75) Altona, C.; Sundaralingam, M. J. Am. Chem. Soc. 1973, 95, 2333-
2344.
(76) Oppenheimer, N. J. In The pyridine nucleotide coenzymes; Everse,
J., Anderson, B., You, K.-S., Eds.; Academic Press: New York, 1982; pp
51-89.
(69) Reddy, B. S.; Saenger, W.; Muhlegger, K.; Weimann, G. J. Am.
Chem. Soc. 1981, 103, 907-914.
(70) Moodie, S. L.; Thornton, J. M. Nucleic Acids Res. 1993, 21, 1369-
1380.
(77) Oppenheimer, N. J. In Pyridine nucleotide coenzymes; Dolphin, D.,
Poulson, R., Avramovic, O., Eds.; Wiley-Interscience: New York, 1987;
pp 185-230.
(71) Gelbin, A.; Schneider, B.; Clowney, L.; Hsieh, S. H.; Olson, W.
K.; Berman, H. M. J. Am. Chem. Soc. 1996, 118, 519-529.
(72) Reference 69 and PDB66 structures: 1GEU and 2NAD (two
independent NAD+ molecules per structure).
(78) Jones, P. G.; Kirby, A. J. J. Am. Chem. Soc. 1984, 106, 6207-
6212.
(79) Kinoshita, Y.; Ruble, J. R.; Jeffrey, G. A. Carbohydr. Res. 1981,
92, 1-7.
(80) Hehre, W. J. Acc. Chem. Res. 1975, 8, 369-376.
(73) PDB structures: 1HLD (two NAD+ molecules), 2OHX, and 1LDM.