Energetics of an n → π* interaction that impacts protein structure

Article abstract of DOI:10.1021/ol061569t

(Figure Presented) The trans/cis ratio of the amide bond in N-formylproline phenylesters correlates with electron withdrawal by a para substituent. The slope of the Hammett plot (ρ = 0.26) is indicative of a substantial effect. This effect arises from a favorable n → π* interaction between the amide oxygen and ester carbonyl. In a polypeptide chain, an analogous interaction can stabilize the conformation of trans peptide bonds, α-helices, and polyproline type-II helices.

Full text of DOI:10.1021/ol061569t

ORGANIC
LETTERS
2006
Vol. 8, No. 21
4695-4697
Energetics of an n f π* Interaction that
Impacts Protein Structure
Jonathan A. Hodges†,‡ and Ronald T. Raines*,†,§
Departments of Chemistry and Biochemistry, UniVersity of WisconsinsMadison,
Madison, Wisconsin 53706
raines@biochem.wisc.edu
Received June 26, 2006
ABSTRACT
The trans/cis ratio of the amide bond in N-formylproline phenylesters correlates with electron withdrawal by a para substituent. The slope of
the Hammett plot (G ) 0.26) is indicative of a substantial effect. This effect arises from a favorable n f π* interaction between the amide
oxygen and ester carbonyl. In a polypeptide chain, an analogous interaction can stabilize the conformation of trans peptide bonds,
r-helices,
and polyproline type-II helices.
Certain noncovalent interactions are known to direct a
polypeptide chain to assume a folded structure. Those inter-
actions include the hydrophobic effect, hydrogen bonding,
Coloumbic forces, van der Waals forces, and cation-π
interactions.1 Recently, we suggested that n f π* interactions
should be added to this list.2,3 In an n f π* interaction (which
is not to be confused with an n f π* electronic transition),
the oxygen of a peptide bond (Oi-1) donates electron density
from one of its lone pairs into the antibonding orbital of the
carbon in the subsequent peptide bond (Ci′dOi). This
Oi-1‚‚‚Ci′dOi interaction, which mimics the approach of a
nucleophile to the electrophilic carbon of an acyl group, is
strongest when Oi-1 is positioned proximally and along the
Bu¨rgi-Dunitz trajectory to Ci′dOi. In polypeptides, this
geometry occurs in the R-helix and polyproline type-II (PPII)
helix,2c which is the structure assumed by the strands of the
prevalent collagen triple helix.3,4
The n f π* interaction can occur only if the peptide bond
containing Oi-1 is trans (i.e., Z), as opposed to cis (E).
Accordingly, the trans/cis ratio of an Xaai-1-Proi peptide
bond can report on the strength of an n f π* interaction.
We had demonstrated previously that the n f π* interaction
contributes to the endogenous preference for trans peptide
bonds.2c Here, we present a simple model system that uses
the trans/cis ratio of a peptide bond to report on the energetics
of an n f π* interaction.
† Department of Chemistry.
‡ Present address: Affinergy, Inc., P.O. Box 14650, Research Triangle
Park, NC 27709.
§ Department of Biochemistry.
(1) For reviews, see: (a) Anfinsen, C. B. Science 1973, 181, 223-230.
(b) Dill, K. A. Biochemistry 1990, 29, 7133-7155. (b) Pace, C. N.; Shirley,
B. A.; McNutt, M.; Gajiwala, K. FASEB J. 1996, 10, 75-83. (c) Dougherty,
D. A. Science 1996, 271, 163-168. (d) Guo, W.; Shea, J.-E.; Berry, R. S.
Ann. N.Y. Acad. Sci. 2005, 1066, 34-53. (e) Parsegian, V. A. Van der Waals
Forces: A Handbook for Biologists, Chemists, Engineers, and Physicists;
Cambridge University Press: Cambridge, U.K., 2005.
(2) (a) Bretscher, L. E.; Jenkins, C. L.; Taylor, K. M.; DeRider, M. L.;
Raines, R. T. J. Am. Chem. Soc. 2001, 123, 777-778. (b) DeRider, M. L.;
Wilkens, S. J.; Waddell, M. J.; Bretscher, L. E.; Weinhold, F.; Raines, R.
T.; Markley, J. L. J. Am. Chem. Soc. 2002, 124, 2497-2505. (c) Hinderaker,
M. P.; Raines, R. T. Protein Sci. 2003, 12, 1188-1194.
(3) For reviews, see: (a) Jenkins, C. L.; Raines, R. T. Nat. Prod.
Rep. 2002, 19, 49-59. (b) Raines, R. T. Protein Sci. 2006, 15, 1219-
1225.
We reasoned that the strength of an n f π* interaction
could be altered by using para-substituted phenyl esters of
(4) (a) Ramachandran, G. N.; Kartha, G. Nature 1954, 174, 269-270.
(b) Ramachandran, G. N.; Kartha, G. Nature 1955, 176, 593-595. (c) Rich,
A.; Crick, F. H. C. Nature 1955, 176, 915-916. (d) Rich, A.; Crick, F. H.
C. J. Mol. Biol. 1961, 3, 483-506. (e) Bella, J.; Eaton, M.; Brodsky, B.;
Berman, H. M. Science 1994, 266, 75-81.
10.1021/ol061569t CCC: $33.50
© 2006 American Chemical Society
Published on Web 09/22/2006

N-formyl-L-proline to modulate the electrophilicity of the
carbonyl (Ci′dOi). Our expectation was that the trans/cis ratio
would correlate with that electrophilicity. We chose N-
formylproline because its trans/cis ratio is close to unity and
can thus serve as the basis for a highly sensitive assay.2c
The esters were synthesized by condensation of N-formyl-
proline with the corresponding phenol using PyBOP activa-
tion and were purified by flash chromatography.
1
We used H NMR spectroscopy to measure the trans/cis
ratio of each ester (90-170 mM) in CDCl3. We chose
chloroform as the solvent to minimize effects from dif-
ferential solvation of the trans and cis isomers. Because the
isomerization of the peptide bond is slow on the NMR time
scale, two sets of signals were observed in the spectrum of
each ester. Though most of the resonances exist as multiplets
and overlap with those from their isomer, the resonance of
the formyl proton appears as two singletssone for each
isomer. The larger, upfield resonance arises from the trans
isomer. Integration of the two signals provides the trans/cis
ratio at the amide bond.
We find that more electron-withdrawing para substituents
correlate strongly with larger trans/cis ratios (Table 1).
Table 1. Trans/Cis Ratios (CDCl3) and σp Values of X in
Fm-Pro-OC6H4-p-X at 25 °C
a
X
σp
Ktrans/cis
∆G (kcal/mol)b
NO2
CN
H
OMe
NMe2
0.78
0.66
0.00
-0.27
-0.83
4.13 ( 0.05
3.95 ( 0.03
2.12 ( 0.01
2.01 ( 0.04
1.67 ( 0.02
-0.84
-0.81
-0.45
-0.41
-0.30
Figure 1. Relationship between the trans/cis ratio and the electron-
withdrawing ability of X in Fm-Pro-OC6H4-p-X. (A) Values of
KX are from experimental data (Table 1) and give F ) 0.26.
(B) Values of KX are from density functional theory calculations
on the trans isomer with Cγ-exo pyrrolidine ring pucker (Table 2)
and give F ) 0.60.
a From ref 6. b ∆G ) -RTlnKtrans/cis
.
Moreover, a linear relationship exists between the log of the
trans/cis ratio and the Hammett parameter (σp) of the
substituent (Figure 1A), which is indicative of an electronic
effect.5 The values of Ktrans/cis and σp are related by F )
∂log(KX/KH)/∂σp ) 0.26. This F-value signifies that the
electronic effect of X on Ktrans/cis is substantial, especially
given that the interconversion of the cis and trans isomers
of Fm-Pro-OC6H4-p-X does not involve a change in
covalent bonding.
favored8 due to π-polarization.9 The diminished delocaliza-
tion decreases the energy of the Ci′dOi π*-orbital, making
Ci′dOi more electrophilic, and leads to an increased contri-
bution from resonance form A, which results in greater
Density functional theory (DFT) calculations predicted a
similar correlation between Ktrans/cis and σp (Table 2; Figure
1B) and enabled a detailed analysis. Electron-withdrawing
substituents on a phenyl ring are known to increase the
electrophilicity7 (and reactivity5) of aryl esters. The electron-
ics of an ester can be described by three major resonance
forms: A-C (Scheme 1). As the phenyl substituent becomes
more electron withdrawing, forms B and C become dis-
Table 2. Results of Density Functional Theory Calculations
(B3LYP/6-31+G*) and Natural Bond Order Analysis on the
Trans and Cis Isomers of Fm-Pro-OC6H4-p-X with Cγ-exo
Pyrrolidine Ring Pucker
∆E
vC′ dO
Eπ*,C′ dO
Enfπ*
(Hartrees)c (kcal/mol)d
rO
i-1‚‚‚C′i
(Å)d
i
i
i
i
b,c
X
(kcal/mol)a (cm-1
)
NO2
CN
H
OMe
NMe2
2.22
2.08
1.27
1.22
0.94
1837
-0.02994
-0.02723
-0.01240
-0.00962
-0.00422
1.45
1.39
0.86
0.83
0.68
2.88
2.89
2.98
2.99
3.02
1835
1829
1826
1824
(5) (a) Hammett, L. P. Chem. ReV. 1935, 17, 125-136. (b) Hammett, L.
P. Physical Organic Chemistry; McGraw-Hill: New York, 1940; pp 184-
228. (c) For a historical review, see: Shorter, J. Chem. Listy 2000, 94,
210-214.
(6) Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165-195.
(7) Conteras, R.; Andres, J.; Domingo, L. R.; Castillo, R.; Perez, P.
Tetrahedron 2005, 61, 417-422.
a In Figure 1B, Ktrans/cis ) e-∆E/RT
isomer.
.
b Unscaled. c Cis isomer. d Trans
4696
Org. Lett., Vol. 8, No. 21, 2006

role of an electronic n f π* interaction in the conformational
preference of the amide bond. This effect cannot be steric.
Located in the para position, the aryl substituent is remote
from the amide bond. Moreover, if the effect was steric, then
the trans/cis ratios would fall in line with the steric bulk of
the substituent. Such a correlation was not observed, as the
Scheme 1
Ktrans/cis value of the smallest substituent, a proton, is between
those of the largest substituents, dimethylamino and nitro.13
We conclude that an n f π* interaction occurs between
the amide and ester groups in this model system. In addition,
we propose that this interaction has important ramifica-
tions for protein structure. An n f π* interaction could
stabilize not only a trans peptide bond but also an R-helix
and a PPII helix.2c The R-helix,14 like the â-sheet,15 is
stabilized by Ci′dOi‚‚‚H-N hydrogen bonds between main-
chain atoms.1b,16 The Ci′dOi bonds in R-helices are, however,
demonstrably longer than those in â-sheets.17 This greater
bond length is consistent with the manifestation of an n f
π* interaction in an R-helix. In contrast to the R-helix and
â-sheet, the PPII helix does not contain any hydrogen bonds
between its main-chain atoms. Yet, recent data indicate a
notable prevalence of the PPII helix in both folded proteins
and unfolded polypeptide chains.18 We propose that the n
f π* interaction stabilizes the PPII helix, contributing
significantly to its prevalence. Accordingly, the n f π*
interaction is a noncovalent interaction, like the hydrophobic
effect, hydrogen bonding, Coloumbic forces, van der Waals
forces, and cation-π interactions,1 that directs a polypeptide
chain to assume a folded structure.
double-bond character and higher Ci′dOi vibrational fre-
quency.8 These effects are observed theoretically in the cis
isomer, which represents the ester electronics in a state
unperturbed by the n f π* interaction. In the cis isomer,
the calculated vibrational frequency (VC′ dO ) of the carbonyl
i
i
increases with the electron-withdrawing ability of the para
substituent (Table 2). Moreover, natural bond order (NBO10)
analysis of the cis isomer demonstrates a decrease in the
ester π*-orbital energy (Eπ*,C′ dO ) with increasing electron
i
i
withdrawal (Table 2). These two effects are indicative of
the increased electrophilicity of Ci′dOi that results from
electron-withdrawing groups.
The strength of the n f π* interaction in the trans isomer
was examined directly with NBO analysis. Second-order
perturbation theory indicates that as the trans/cis ratio
increases so does the energy (Enfπ*) of the interaction
between the lone pair on the amide carbonyl oxygen and
the π*-orbital of the ester (Table 2). The increasing strength
of the n f π* interaction is reflected in the decreasing
distance (rO ‚‚‚C′ ) between the amide oxygen and the ester
i-1
i
carbon (Table 2). The value of F ) 0.60 for the theoretical
data (Figure 1B) is greater than that for the experimental
data (Figure 1A). The pyrrolidine ring was held in the
Cγ-exo conformation during the theoretical analyses, as this
ring pucker allows for much greater n f π* interaction than
the Cγ-endo conformation.2b,11 The pyrrolidine ring has,
however, a slight preference for the Cγ-endo pucker.12 Hence,
the effect of the para substituent on the trans/cis ratio is
amplified in the calculations.
Acknowledgment. We are grateful to F. Weinhold
(University of WisconsinsMadison) for contributive dis-
cussions. J.A.H. was supported by postdoctoral fellowship
AR48057 (NIH). This work was supported by grant AR44276
(NIH).
Supporting Information Available: Procedures for the
preparation of N-formylproline phenylesters and related
analytical data. This material is available free of charge via
the Internet at http://pubs.acs.org.
The link between the electron-withdrawing ability of the
remote para substituent and the trans/cis ratio supports the
(8) Neuvonen, H.; Neuvonen, K.; Koch, A.; Kleinpeter, E.; Pasanen, P.
J. Org. Chem. 2002, 67, 6995-7003.
OL061569T
(9) Bromilow, J.; Brownlee, R. T. C.; Craik, D. J.; Fiske, P. R.; Rowe,
J. E.; Sadek, M. J. Chem. Soc., Perkin Trans. 2 1981, 753-759.
(10) (a) Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.; Carpenter, J.
E.; Bohmann, J. A.; Morales, C. M.; Weinhold F. NBO 5.0; Theoretical
Chemistry Institute, University of WisconsinsMadison: Madison, WI,
2001; http://www.chem.wisc.edu/∼nbo5. (b) Weinhold, F.; Landis, C. R.
Valency and Bonding: A Natural Bond Orbital Donor-Acceptor Perspec-
tiVe; Cambridge University Press: Cambridge, U.K., 2005.
(11) The pyrrolidine ring of proline actually prefers two distinct twist
(rather than envelope) conformations. As Cγ experiences the largest out-
of-plane displacement in these twisted rings, we refer to pyrrolidine ring
conformations (herein and elsewhere) simply as “Cγ-exo” and “Cγ-endo”.
For additional information, see: Giacovazzo, C.; Monaco, H. L.; Artioli,
G.; Viterbo, D.; Ferraris, G.; Gilli, G.; Zanotti, G.; Catti, M. Fundamentals
of Crystallography, 2nd ed.; Oxford University Press: Oxford, U.K., 2002.
(12) The pyrrolidine ring of Ac-Pro-OMe has 66% Cγ-endo pucker
and 34% Cγ-exo pucker in dioxane at 25 °C (ref 2b).
(13) In contrast to para substituents, ortho substituents can have adverse
steric consequences. For example, DFT analysis of the highly electron-
withdrawing pentafluorophenyl ester showed a lower trans/cis ratio than in
the unsubstituted phenyl ester, presumably due to a steric interaction between
the ortho-fluoro groups and the amide oxygen (data not shown).
(14) Pauling, L.; Corey, R. B.; Branson, H. R. Proc. Natl. Acad. Sci.
U.S.A. 1951, 37, 205-211.
(15) Pauling, L.; Corey, R. B. Proc. Natl. Acad. Sci. U.S.A. 1951, 37,
251-256.
(16) For a historical review, see: Eisenberg, D. Proc. Natl. Acad. Sci.
U.S.A. 2003, 100, 11207-11210.
(17) Lario, P. I.; Vrielink, A. J. Am. Chem. Soc. 2003, 125, 12787-
12794.
(18) For reviews, see: (a) Shi, Z.; Olson, C. A.; Bell, A. J., Jr.;
Kallenbach, N. R. AdV. Protein Chem. 2002, 62, 163-240. (b) Creamer,
T. P.; Campbell, M. N. AdV. Protein Chem. 2002, 62, 263-282. (c) Rath,
A.; Devidson, A. R.; Deber, C. M. Biopolymers 2005, 80, 179-185.
Org. Lett., Vol. 8, No. 21, 2006
4697