Unexpected substituent effects in offset π-π stacked interactions in water

Article abstract of DOI:10.1021/ja016508z

We have measured the rotational barriers of meta- and para-substituted N-benzyl-2-(2-fluorophenyl)pyridinium bromides in aqueous solution by dynamic NMR as a model system for offset-stacking interactions in proteins. Because the benzyl ring can stack with

Full text of DOI:10.1021/ja016508z

Published on Web 02/09/2002
Unexpected Substituent Effects in Offset π-π Stacked Interactions in Water
Mark J. Rashkin and Marcey L. Waters*
Department of Chemistry, UniVersity of North Carolina, Chapel Hill, North Carolina 27599-3290
Received June 27, 2001; Revised Manuscript Received January 7, 2002
Table 1. Rotational Barriers (kcal/mol) of Substituted Benzyl
Aromatic interactions are common motifs in biomolecular
structure and molecular recognition.1-3 However, despite their
frequent occurrence, there is no unifying picture of the factors that
contribute to the interaction, which include electrostatic (quadra-
pole-quadrapole and quadrapole-dipole, and dipole-dipole),4
hydrophobic,2a,5 and van der Waals interactions.2a,4d This is
complicated by the fact that aromatic rings interact in several
different conformations, each of which is favored by a different
combination of forces.1 These conformations include the edge-
face interaction for two unsubstituted benzene rings, the offset
stacked conformation for substituted aromatic rings, and the face-
face stacked geometry for the interaction of benzene with perfluo-
robenzene. The offset stacked geometry is the most common
geometry for aromatic interactions, but the least well studied. To
clarify the factors that influence offset stacking, we have undertaken
a quantitative study of substituent effects on this interaction in
aqueous solution. We have found that the magnitude of the offset
stacked interaction is dependent on the orientation of the rings, and
it appears that a direct interaction between the ring hydrogens and
the substituents themselves may significantly influence the mag-
nitude of the interaction.
Pyridinium Bromides in Watera
q
q
q
compd
X
∆G
∆∆G (X-H)
∆∆G (m-p)
1
H
4-CH3
3-CH3
3,5-(CH3)2
4-CF3
3-CF3
3,5-(CF3)2
4-NO2
3-NO2
3,5-(NO2)2
16.81
16.80
16.86
16.88
16.87
17.17
17.47
16.91
17.13
17.32
-
-
2a
2b
2c
3a
3b
3c
4a
4b
4c
-0.01
0.05
0.07
0.06
0.36
0.66
0.1
-
0.06
-
-
0.30
-
-
0.32
0.51
0.22
-
a Energies were measured at 332 K as described in the text. Propagation
of errors gives an uncertainty of less than (0.04 kcal/mol.7
Inspection of the rotational barriers in Table 1 indicates that there
is a small substituent effect on offset stacking in which electron-
withdrawing groups increase the magnitude of the rotational barrier,
and that it is larger for meta-substituents than para-substituents.8
Comparison of the electrostatic potential surfaces of compounds 1
(X ) H) and 4a (X ) 4-NO2) suggests that the higher barrier found
with electron-withdrawing groups in the para position is due to the
reduced electron density on the face of the C ring relative to
compound 1 (Figure 1).9 Reduction in the electron density on the
face of one ring has been shown to decrease the repulsive
component resulting from interaction of the π-clouds in the face-
face stacked conformation.4b The effect is significantly smaller in
the offset-stacked conformation, presumably due to the reduced
interaction of the π-clouds in this geometry.
As a model system for offset stacking interactions, we investi-
gated meta- and para-substituted N-benzyl-2-(2-fluorophenyl)-
pyridinium bromides of the type 1, which stack in the offset
conformation in the solid state.6 The single methylene linker
between the B and C rings prohibits edge-face interactions. The
pyridinium ring provides water solubility and the ortho-fluoro group
was incorporated to desymmetrize ring A, making Ha and Hb
diastereotopic. The relative strengths of the stacking interactions
with various substituents, X, were determined by measuring rate
constants for restricted rotation around the biaryl bond in which
the stacking interaction between rings A and C must be disrupted
in the transition state. The rotational barrier was determined by
dynamic NMR through simulation of the line-broadened spectra
of Ha and Hb in D2O.7 Because the substituent, X, does not interact
with the A or B ring in the transition state either through space or
via conjugation, changes in the rotational barrier can be attributed
solely to changes in the ground-state interaction.
Figure 1. Ab initio electrostatic potential calculations of 1, 4a, and 4b
indicating the interactions of the faces of the A (top) and C (bottom) rings.9
The larger rotational barrier observed for the meta-substituents
relative to the para-substituents cannot be explained by changes in
the electron density on the face of the rings alone. For example,
compound 4b (X ) 3-NO2) has a similar degree of electron density
on the face of the C ring as compound 4a (Figure 1), but the
rotational barrier for 4b is 0.22 kcal/mol higher in energy than that
of 4a. Modeling studies indicate that an oxygen on the m-nitro group
is in close proximity to Hd of the A ring in compound 4b, suggesting
that there may be an attractive electrostatic interaction between the
edge hydrogen (δ+) and oxygen (δ-) of the nitro group that is
not possible in 1 or 4a (Figure 2), and that this may contribute to
the higher rotational barrier for 4b relative to 4a.
Support for the presence of an interaction between a ring
1
hydrogen and the nitro substituent in 4b comes from H NMR
9
1860 VOL. 124, NO. 9, 2002 J. AM. CHEM. SOC.
10.1021/ja016508z CCC: $22.00 © 2002 American Chemical Society

C O M M U N I C A T I O N S
In summary, our results indicate that the orientation of two
stacked rings can have a significant effect on the magnitude of the
interactions.11 Although variations in interaction strength are small,
these variations can have a considerable effect in the context of a
protein, in which many of such weak interactions are involved.
The variation in the magnitude of the stacking interaction with the
meta-substituents in this system appears to be due in part to direct
interaction of the edge hydrogens of one ring with electronegative
substituents on the other ring. This may have implications for
stacking of other substituted rings such as the DNA bases. The
generality of these findings is currently under investigation.
Figure 2. Space-filling models of 1, 4a, and 4b indicating the interaction
of Hd on ring A (front) with the substituent on ring C (back).
chemical shift data. If interaction of Hd with the nitro group in 4b
does indeed increase the stability of the offset stacked geometry,
conformation I should be favored over conformation II (Scheme
1). We compared the chemical shifts of Hg and Hk on the C ring
of 4b to a control compound, 5, in which the A ring is missing.
The relative upfield shifts of Hg and Hk should reflect the relative
populations of I and II. Both Hg and Hk in compound 4b are upfield
shifted relative to the ortho hydrogens in compound 5, but Hg is
upfield shifted by about 0.1 ppm more than is Hk, indicating that
I is in fact the lower energy conformation.10
Acknowledgment. We gratefully acknowledge the University
of North Carolina College of Arts and Sciences for startup funds
and the National Science Foundation for a Career Award (Grant
CHE-0094068).
Supporting Information Available: Synthesis, characterization,
and kinetic measurements of reported compounds (PDF). This material
is available free of charge via the Internet at http://pubs.acs.org.
Scheme 1. Possible Conformations for Meta-Substituted
Compounds
References
(1) For a recent review on aromatic interactions, see: Hunter, C. A.; Lawson,
K. R.; Perkins, J.; Urch, C. J. J. Chem. Soc., Perkin Trans. 2 2001, 651-
669 and references therein.
(2) For recent examples of aromatic interactions in biomolecules, see: (a)
Guckian, K. M.; Schweitzer, B. A.; Ren, R. X. F.; Sheils, C. J.;
Tahmassebi, D. C.; Kool, E. T. J. Am. Chem. Soc. 2000, 122, 2213-
2222. (b) Blakaj, D. M.; McConnell, K. J.; Beveridge, D. L.; Baranger,
A. M. J. Am. Chem. Soc. 2001, 123, 2548-2551. (c) Kirksey, T. J.; Pogue-
Caley, R. R.; Frelinger, J. A.; Collins, E. J. J. Biol. Chem. 1999, 274,
37259-37264.
(3) For recent examples of aromatic interactions in molecular recognition,
see: (a) Georgopoulou, A. S.; Mingos, D. M. P.; White, A J. P.; Williams,
D. J.; Horrocks, B. R.; Houlton, A. J. Chem. Soc., Dalton Trans. 2000,
2969-2974. (b) Inouye, M.; Itoh, M. S.; Nakazumi, H. J. Org. Chem.
1999, 64, 9393-9398. (c) Ponzini, F.; Zagha, R.; Hardcastle, K.; Siegel,
J. S. Angew. Chem., Int. Ed. 2000, 39, 2323-2325.
(4) (a) Hunter, C. A.; Sanders, J. K. M. J. Am. Chem. Soc. 1990, 112, 5525-
5534. (b) Cozzi, F.; Siegel, J. S. Pure Appl. Chem. 1995, 67, 683-689.
(c) Ferguson, S. B.; Diederich, F. Angew. Chem., Int. Ed. Engl. 1986, 25,
1127-1129. (d) Breault, G. A.; Hunter, C. A.; Mayers, P. C. J. Am. Chem.
Soc. 1998, 120, 3402-3410. (e) Heaton, N. J.; Bello, P.; Herrandon, B.;
del Campo, A.; Jimenez-Berbero, J. J. Am. Chem. Soc. 1998, 120, 12371-
12384. (f) Smithrud, D. B.; Sanford, E. M.; Chao, I.; Ferguson, S. B.;
Carcanague, D. R.; Evanseck, J. D.; Houk, K. N.; Diederich, F. Pure Appl.
Chem. 1990, 62, 2227-2236.
The difference in barriers of 3a (X ) 4-CF3) and 3b (X ) 3-CF3)
shows the same trend as for the m- and p-nitro-substituted
compounds. As with the 3-NO2 group, modeling studies indicate a
close contact between the CF3 group and Hd and He of the A ring
in 3b (Figure 3) that is not possible in 3a. Because fluorine is poorly
solvated by water, the larger interaction with the CF3 group may
also be due to the hydrophobic effect. However, no significant
difference between the rotational barriers is observed for the p-
and m-methyl compounds 2a (X ) 4-CH3) and 2b (X ) 3-CH3),
despite the fact that the 3-CH3 group is also in close proximity to
the edge of the A ring. This suggests that the hydrophobic effect
alone cannot explain the difference in rotational barriers of
compounds 3a and 3b.
(5) (a) Newcomb, L. F.; Gellman, S. H. J. Am. Chem. Soc. 1994, 116, 4993-
4994. (b) Gellman, S. H.; Haque, T. S.; Newcomb, L. F. Biophys. J. 1996,
71, 3523-3525. (c) Pang, Y.-P.; Miller, J. L.; Kollman, P. A. J. Am.
Chem. Soc. 1999, 121, 1717-1725.
(6) Martin, C. B.; Mulla, H. R.; Willis, P. G.; Cammers-Goodwin, A. J. Org.
Chem. 1999, 64, 7802-7806.
(7) Spectra were simulated using the program gNMR version 3.6.5 (Adept
Scientific). Each rate constant was measured at least twice, with an error
of less than 5%. The temperature was calibrated before and after each
spectrum based on the temperature dependence of ethylene glycol chemical
shifts, with an uncertainty of (0.2 K, resulting in an uncertainty of less
than (0.04 kcal/mol. The spectra were found to be independent of
concentration under the conditions studied, indicating that no aggregation
was occurring. The chemical shifts of Ha and Hb were found to be
independent of temperature.
(8) Similar energy differences have been observed for edge-face interactions
and the interactions of heteroaromatic rings. See: Kim, E.; Paliwal, S.;
Wilcox, C. S. J. Am. Chem. Soc. 1998, 120, 11192-11193. McKay, S.
L.; Haponstall, B.; Gellman, S. H. J. Am. Chem. Soc. 2001, 123, 1244-
45.
(9) Electrostatic potential surfaces were generated by mapping 6-311G*
electrostatic potentials onto surfaces of molecular electron density using
Spartan (Wavefunction, Inc, Irvine, CA). The potential energy range is
50-125 kcal/mol for all surfaces shown. Red indicates areas of greater
electron density and blue indicates areas of less electron density.
(10) The sum of the upfield shifts of Hg and Hk (1.25 ppm) agrees well with
the value of 1.24 ppm calculated from the crystal structure of N-benzyl-
2-phenylpyridinium bromide,6 suggesting that the rings spend a significant
proportion of their time in the offset stacked conformation in solution.
See: Pople, J. A. J. Chem. Phys. 1956, 24, 1111.
(11) A geometric dependence on aromatic interactions has recently been
observed in a flavoenzyme mimic: Goodman, A. J.; Breinlinger, E. C.;
McIntosh, C. M.; Grimaldi, L. N.; Rotello, V. M. Org. Lett. 2001, 3,
1531-1534.
Figure 3. Space-filling models of 2b and 3b indicating the interaction of
Hd and He on ring A (front) with the substituent on ring C (back).
Because the meta-substituted compounds are in rapid confor-
mational equilibrium, the magnitude of the rotational barrier reflects
the relative ground-state populations of I and II. If the interaction
of the meta-substituent with the ring is important, as proposed, then
the rotational barrier should be higher for a compound with two
meta-substituents. This is in fact what was found with compounds
3c and 4c. For example, the rotational barrier for compound 3b (X
) 3-CF3) is 0.36 kcal/mol greater than that for compound 1 (X )
H), and the barrier for 3c (X ) 3,5-(CF3)2) is 0.66 kcal/mol greater
than that for compound 1. In contrast, as for 2b (X ) 3-CH3),
compound 2c (X ) 3,5-(CH3)2) showed a negligible increase in
rotational barrier.
JA016508Z
9
J. AM. CHEM. SOC. VOL. 124, NO. 9, 2002 1861