Trans Isomerization of Secondary Amide

Feb 5, 2010 - E-mail: [email protected]. ... For a more comprehensive list of citations to this article, users are encouraged to perform a sea...
0 downloads 0 Views 176KB Size
Download PDF
J. Phys. Chem. B 2010, 114, 3387–3392

3387

Kinetics and Equilibria of Cis/Trans Isomerization of Secondary Amide Peptide Bonds in Linear and Cyclic Peptides Khanh Nguyen, Margret Iskandar, and Dallas L. Rabenstein* Department of Chemistry, UniVersity of California, RiVerside, California 92521 ReceiVed: January 2, 2010

The secondary amide peptide bonds that comprise up to one-third of the bonds of peptide or protein backbones can exist as cis and trans isomers, with the trans isomer being highly favored. However, there is little quantitative data on the kinetics and equilibria of cis-trans isomerization of secondary amide peptide bonds due to the difficulty of detecting the very small population of cis isomers. Knowledge of factors that influence the kinetics and equilibria of cis-trans isomerization of secondary amide peptide bonds will contribute to a more complete understanding of the structural and dynamic behavior of the backbones of peptides and unfolded proteins and of complex protein folding kinetics. We have characterized the kinetics and equilibria of cis-trans isomerization of the Xaa-Yaa secondary amide peptide bonds of the linear dithiol and cyclic disulfide forms of the peptides Ac-Cys-Xaa-Yaa-Cys-His-NH2, where Xaa-Yaa is Ala-Phe, Phe-Ala, Ala-Tyr, and Tyr-Ala, by 1H NMR. Resolved resonances were observed for the Ala-methyl protons of the trans and the much less abundant cis isomers due to differential shielding of the Ala-methyl protons of the trans and cis isomers by ring current effects from the Phe and Tyr side chains. The population of the cis isomers was determined from the relative intensities of the Ala-methyl resonances for the trans and cis isomers, and rate constants for cis-to-trans and trans-to-cis isomerization were determined by the magnetization transfer NMR method. The population of the cis isomers ranges from 0.07 to 0.12%, and the rate constants indicate that, when there is a trans-to-cis interchange, it is rapidly followed by a cis-to-trans interchange back to the more stable trans conformation. Although cyclization by disulfide bond formation imposes conformational constraints on the peptide backbones, cyclization is found to have relatively small affects on the dynamics of cis-trans isomerization. Introduction Due to its partial double bond character, the peptide bond is rigid and planar and exists in trans and cis conformations. Even though up to one-third of the bonds along a peptide or protein backbone are secondary amide peptide bonds, i.e., amide peptide

bonds to nitrogen of the nonproline amino acids, there is almost no quantitative data on the population of the cis isomer or the kinetics of cis-trans isomerization of secondary amide peptide bonds.1-3 It is well established that secondary amide peptide bonds favor the trans conformation due to less steric interaction between side chains of flanking amino acids,4-6 and that, in the native, functional state of folded proteins, secondary amide peptide bonds generally exist exclusively in the trans conformation.5-10 However, the secondary amide peptide bonds in peptides and unfolded proteins exist in a state of dynamic equilibrium between trans and cis conformations, with the trans conformation much more highly populated. Characterization of the kinetics and equilibria of cis-trans isomerization of secondary amide peptide bonds in peptides and unfolded proteins has been hampered by the lack of experimental * To whom correspondence [email protected].

should

be

addressed.

E-mail:

methods for detecting the very small populations of cis isomers. Consequently, only a few studies of cis-trans isomerization of secondary amide peptide bonds have been reported and these have been limited to small peptides, mainly dipeptides.1-3 Rate constants were determined for cis-trans isomerization of the secondary amide peptide bonds of Gly-Gly, Gly-Ala, and AlaGly by a UV/visible spectral method by exploiting differences in the absorption properties of the cis and trans conformations, and the dependence of the cis population on the protonation state of the dipeptides.1 Isomerization rate constants were determined by following spectral changes after pH jumps by stopped flow methods. Rate constants for cis-trans isomerization of Gly-Gly have also been measured by a UV resonance Raman method in which photon absorption was used to isomerize the trans isomer to the cis isomer.2 However, the rate constants obtained are much larger than those measured for GlyGly by the UV/visible method.1 Rate constants and thermodynamic parameters were determined by 1H NMR for the secondary amide peptide bonds linking alanine to the amino and carboxyl groups of phenylalanine and tyrosine in dipeptides and several small peptides.3 Conformer-specific chemical shifts were observed for Ala-methyl groups flanking the aromatic amino acids. Knowledge of factors that govern the kinetics and equilibria of cis-trans isomerization of secondary amide peptide bonds will contribute to a more complete understanding of the structural and dynamic behavior of the backbones of peptides and unfolded proteins and of complex protein folding kinetics. For example, even though the cis population of each secondary amide peptide bond in an unfolded protein is statistically very

10.1021/jp1000286  2010 American Chemical Society Published on Web 02/05/2010

3388

J. Phys. Chem. B, Vol. 114, No. 9, 2010

Nguyen et al.

TABLE 1: Peptides Studied in This Research and Chemical Shift Data for the Ala-Methyl Resonances of cis and trans Isomers of the Indicated Secondary Amide Peptide Bonds Ala-CH3 chemical shifta

a

peptide

amino acid sequence

peptide bond

trans

cis

1a 1b 2a 2b 3a 3b 4a 4b

Ac-C-A-Y-C-H-NH2 Ac-C-A-Y-C-H-NH2 Ac-C-Y-A-C-H-NH2 Ac-C-Y-A-C-H-NH2 Ac-C-A-F-C-H-NH2 Ac-C-A-F-C-H-NH2 Ac-C-F-A-C-H-NH2 Ac-C-F-A-C-H-NH2

A-Y A-Y Y-A Y-A A-F A-F F-A F-A

1.295 1.280 1.321 1.296 1.274 1.211 1.328 1.305

0.841 0.922 0.984 1.083 0.791 0.886 0.986 1.085

ppm vs TMSP.

small, the necessary cis-to-trans isomerization of these peptide bonds during the folding of a protein with all trans secondary amide peptide bonds in the native state can give rise to complex folding kinetics, with the rate limiting step being cis-to-trans isomerization for a very small fraction of the protein.11 For example, 5% of the molecules in a proline-free variant of the 74-amino acid protein Tendamistat fold slowly, with a rate constant of 2.5 s-1, due to slow cis-to-trans isomerization of cis secondary amide peptide bonds.11 Likewise, trans-to-cis isomerization of a peptide bond during the folding of a protein with a cis secondary amide peptide bond in the native state can be the rate limiting step.12,13 In this paper, we report the results of 1H NMR studies of the kinetics and equilibria of cis-trans isomerization of the XaaYaa secondary amide peptide bond in a series of peptides of the sequence Ac-Cys-Xaa-Yaa-Cys-His-NH2, where Xaa-Yaa1 is Ala-Tyr, Tyr-Ala, Ala-Phe, and Phe-Ala. Rate and equilibrium constants for cis-trans isomerization were determined for each peptide in both its linear dithiol and cyclic disulfide forms to determine the effect of conformational constraints imposed by cyclization of the peptide on cis-trans isomerization of the secondary amide peptide bonds. The rate and equilibrium constants are among the very few reported for cis-trans isomerization of secondary amide peptide bonds in linear peptides1-3 and are the first to be reported for secondary amide peptide bonds in cyclic peptides. The peptides were chosen for study, as they are models for the active site of oxido-reductase enzymes that have the Cys-Xaa-Yaa-Cys active site motif. It has been proposed, on the basis of NMR structures, that functional differences between the oxidized and reduced forms of E. coli thioredoxin are related to differences in conformational flexibility in and near the active site loop of the oxidized form.14 Experimental Section Materials. Trifluoroacetic acid (TFA) and 9-fluorenylmethoxycarbonyl (Fmoc)-protected amino acids were purchased from Chem-Impex International Inc. N,N′-dicyclohexylcarbodiimide (DCC), 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU), and N-methyl-2-pyrrolidone (NMP) were obtained from Applied Biosystems. SigmaAldrich supplied triisopropylsilane (TIPS), R-cyano-4-hydroxycinnamic acid (CHCA), piperidine (PIP), sodium 3-(trimethylsilyl)propionate-2,2,3,3-d4 (TMSP), and N,N′-diisopropylcarbodiimide (DIPCDI), and N,N′-dimethylformide (DMF), phenol, methanol, acetonitrile, acetic anhydride, and methyl t-butyl ether (MTBE) were obtained from Fisher Scientific. DTT (1,4-dithiol-DLthreitol) was obtained from Fluka. Cambridge Isotope Laboratories supplied NaOD (40%), DCl (35%), D2O, and deuterated

(98%) dithiothreitol, and Rink amide 4-methylbenzhydrylamine (MBHA) resin (0.56 mmol/g) was purchased from NovaBiochem. Peptide Synthesis and Purification. Peptides 1a-4a in Table 1 were synthesized on an Applied Biosystems ABI-433A solid phase peptide synthesizer using standard Fmoc peptide synthesis methodology. Fmoc-MBHA with a loading capacity of 0.56 mmol/g and a 10 times excess of each Fmoc-protected amino acid were used for the peptide synthesis. Support-bound peptides were cleaved from the resin and side chains deprotected with a cleavage cocktail comprised of 88% TFA, 4.2% H2O, 5.8% phenol, and 2% triisopropylsilane (TIPS) by volume.15,16 Cyclic disulfide-bridged peptides (peptides 1b-4b in Table 1) were prepared by oxidation of peptides 1a-4a with trans[Pt(en)2Cl2]2+, where en is ethylenediamine, an oxidizing agent for rapid and quantitative formation of intramolecular disulfide bonds with high selectivity.17 The crude peptides were purified on a Varian HPLC using a Vydac C18 semiprep reversed phase column (10 × 250 mm). Elution from the column was monitored at a wavelength of 215 nm. Peptides were eluted from the column with a gradient of mobile phase A (0.1% TFA in H2O) and mobile phase B (0.1% TFA in acetonitrile). Optimum retention time and resolution were achieved with a linear gradient of 2-30% mobile phase B in 30 min. Identities of crude peptides after cleavage from the resin and of pure peptides isolated by reversed-phase HPLC were confirmed by matrix-assisted, laser desorption/ionization mass spectrometry. NMR Samples. 320 µL solutions of ∼20 mM peptide were prepared in 90% H2O/10% D2O at pH ∼3.0. TMSP was added for a chemical shift reference. To remove any disulfide peptide formed by air oxidation in solutions of the dithiol peptides, a 3-fold excess of deuterated DTT was added, the pH was adjusted to ∼7.5 with 0.1 M DCl and 0.1 M NaOD, the peptide solutions were allowed to react with DTT for 1 h, and the pH was then adjusted to ∼3.0. Sample solutions were degassed with N2 for 30 m. NMR experiments were run immediately after degassing the sample to minimize air oxidation. NMR Spectroscopy. One- and two-dimensional 1H NMR spectra were measured at 500 MHz with a Varian Unity-Inova spectrometer. Chemical shifts are reported relative to the methyl resonance of TMSP at 0.000 ppm. The residual HOD resonance was suppressed with a presaturation pulse. Two-dimensional total correlation spectroscopy (TOCSY) and rotating frame Overhauser effect spectroscopy (ROESY) spectra and bandselective, homonuclear-decoupled (BASHD)-TOCSY and BASHD-ROESY spectra were measured with standard pulse sequences.18-21 TOCSY and ROESY spectra were measured with the following parameters: spectral width of 5500 Hz in both dimensions, 8196 data points in the directly detected (F2)

Isomerization of Secondary Amide Peptide Bonds

J. Phys. Chem. B, Vol. 114, No. 9, 2010 3389

Figure 1. The Ala-methyl region of the 1D 1H NMR spectrum of a 22 mM solution of the disulfide form of Ac-Cys-Tyr-Ala-Cys-His-NH2 (peptide 2b) in 90% H2O/10% D2O at pH 2.91 and 25 °C. The spectrum was measured by the single pulse method, with suppression of the H2O resonance by presaturation. Spectrum B is plotted with a vertical scale 320× that of spectrum A.

dimension, 64 transients, and 128 t1 increments. Mixing times of 120 and 200 ms were used for TOCSY and ROESY experiments, respectively. Shifted sine bell and Gausian apodization were applied in the F1 and F2 dimensions, respectively. BASHD-TOCSY and BASHD-ROESY spectra were measured with the same parameters, with the exception that the spectral width in the F1 dimension was less, as needed to cover a specific band of resonances.20,21 Rate constants for cis-trans isomerization were determined by the inversion-magnetization transfer method using the Alamethyl resonances. The trans resonance of a given cis/trans pair of resonances was selectively inverted with the pulse sequence: 23 90°x-τ-90°x-t-90°(x,(y -acquisition, where τ is a fixed delay time which equals 1/(2|νt - νc|), νt - νc is the chemical shift difference in Hz of the resonances for the trans and cis isomers, and t is a variable delay during which magnetization transfer takes place by interchange between the cis and trans isomers.22 The more intense trans resonance was selectively inverted, and the transfer of inversion to the less intense cis resonance was monitored as a function of t. Typically, inversion-magnetization transfer spectra were measured at 14-16 t values ranging from 0.0001 s to at least 5 times the longest T1 of the cis and trans resonances; T1 values were determined in a separate experiment by the inversion-recovery method. For each t value, 256 transients were collected. Results Assignment of 1H NMR Spectra. The 1H NMR spectrum of each peptide in Table 1 is comprised of resonances for the peptide with all peptide bonds in the trans conformation, and a much less intense resonance for the Ala-methyl protons of the isomer in which the Ala-Phe, Phe-Ala, Ala-Tyr, or Tyr-Ala peptide bond is in the cis conformation. To illustrate, the Alamethyl region of the spectrum of peptide 2b is shown in Figure 1. Only the Ala-12CH3 resonance of the isomer having the trans conformation across the Tyr-Ala peptide bond is observed in spectrum A. The Ala-12CH3 resonance of the cis isomer and the 13C-satellites of the Ala-12CH3 resonance of the trans isomer are also observed in spectrum B, which is plotted with a 320 times vertical scale expansion.

Figure 2. The NH(F1)-CRH(F2) region of the BASHD-ROESY spectrum of 22 mM peptide 2b in 90% H2O/10% D2O at pH 2.91 and 25 °C, measured with F1-band selection of the NH region. The NHi-CRHi+1 cross peaks that establish the all-trans conformation for the most abundant isomer are identified. The cross peak in the inset is from the standard ROESY spectrum; it is comprised of two overlapping cross peaks (TyrNH-TyrCRH and TyrCRH-AlaNH) that are resolved in the BASHD-ROESY spectrum as a result of collapse of multiplets to singlets in the F1 dimension.

The resonances of the all-trans peptide were assigned using two-dimensional BASHD-TOCSY and BASHD-ROESY spectra. First, the backbone amide NH resonances were assigned to specific amino acids in the peptide by measuring BASHDTOCSY spectra with F1-band selection of the amide NH region. The trans conformation across each secondary amide peptide bond was established by dipolar cross peaks in BASHD-ROESY spectra measured with F1-band selection of the amide NH region. To illustrate, the NH (F1)-CRH (F2) region of the BASHD-ROESY spectrum of peptide 2b is shown in Figure 2, with the CRHi-NHi+1 cross peaks that establish the trans conformation across the secondary amide peptide bonds identified. The inset shows the overlapped TyrNH-TyrCRH and

3390

J. Phys. Chem. B, Vol. 114, No. 9, 2010

Nguyen et al.

TABLE 2: Population of the cis Conformation, Equilibrium Constants, and Rate Constants for cis/trans Isomerization of the Indicated Secondary Amide Peptide Bonds, and Free Energy Differences of the cis and trans Isomers of the Peptides in Table 1a peptide

peptide bond

1a 1b 2a 2b 3a 3b 4a 4b

A-Y A-Y Y-A Y-A A-F A-F F-A F-A

a

% cis

Keq

∆G (kcal/mol)

kct (s-1)

0.081 ( 0.0003 0.074 ( 0.0003 0.070 ( 0.0004 0.097 ( 0.0008 0.098 ( 0.0007 0.087 ( 0.002 0.074 ( 0.0003 0.12 ( 0.001

1232 ( 4 1327 ( 5 1422 ( 9 1027 ( 9 1019 ( 7 1082 ( 16 1353 ( 6 843 ( 10

4.22 4.26 4.30 4.11 4.43 4.14 4.27 3.99

2.0 ( 0.09 0.60 ( 0.03 1.4 ( 0.05 1.3 ( 0.08 0.86 ( 0.06 0.37 ( 0.05 1.2 ( 0.08 1.1 ( 0.1

ktc (s-1) (1.6 ( 0.07) × 10-3 (0.45 ( 0.02) × 10-3 (0.99( 0.04) × 10-3 (1.2 ( 0.08) × 10-3 (0.85 ( 0.06) × 10-3 (0.34 ( 0.05) × 10-3 (0.91 ( 0.06) × 10-3 (1.3 ( 0.2) × 10-3

25 °C.

TyrCRH-AlaNH cross peaks in the regular ROESY spectrum, to demonstrate the significant increase in resolution achieved by collapse of multiplets to singlets in the F1 dimension in the BASHD-ROESY spectrum.21 The Ala-methyl resonance at 1.083 ppm in Figure 1 was assigned to the isomer having the cis conformation across the Tyr-Ala peptide bond by one-dimensional magnetization transfer experiments. The Ala-methyl resonance at 1.296 ppm for the trans isomer was selectively inverted, and the intensity of the cis resonance measured as a function of the exchange time in the inversion-magnetization transfer pulse sequence. As described below, the intensity of the cis resonance decreased as the exchange time was increased, indicating the Ala-methyl resonance assigned to the trans isomer is linked by chemical exchange to the resonance at 1.083 ppm. The chemical shifts of the Ala-methyl protons of the trans and cis isomers are listed in Table 1. The Ala-methyl resonances assigned to the cis isomers are shifted upfield of the trans resonances, with the magnitude of the shift being larger for the Ala-Phe and AlaTyr sequence isomers.3 Equilibrium Constants for Cis/Trans Isomerization. Equilibrium constants for cis/trans isomerization (Keq ) [trans]/[cis]) of the Tyr-Ala, Phe-Ala, Ala-Tyr, and Ala-Phe peptide bonds of the peptides listed in Table 1 were determined using the relative areas of the Ala-methyl cis and trans resonances. Because of the large differences in intensity of the cis and trans resonances (Figure 1), relative resonance intensities were determined using the upfield 13C-satellite resonance of the Alamethyl trans resonance, which comprises 0.554% of the trans resonance intensity. Due to the very low intensities of the 13Csatellite and cis resonances, relative areas were determined from the relative masses of the 13C-satellite and cis resonances. The percent cis isomer, equilibrium constants for cis/trans isomerization, and the difference in free energy of the cis and trans isomers for each peptide are reported in Table 2. Kinetics of Cis/Trans Isomerization. Rate constants were determined by selective inversion of the Ala-methyl resonance for the trans isomer of each cis/trans pair. The intensity of the cis resonance was then measured as a function of the length of the exchange delay in the inversion-magnetization transfer pulse sequence. To illustrate, inversion-magnetization transfer spectra for peptide 2b are shown in Figure 3. The decrease in intensity of the cis resonance as the delay time is increased is due to transfer of inverted trans resonance intensity to the cis resonance by trans-to-cis interchange. At longer delay times, the cis resonance intensity recovers to its equilibrium intensity by T1 relaxation. Rate constants for cis-to-trans isomerization (kct) for peptides 1-4 in Table 1 were determined from the dependence of the intensity of the cis resonance on delay time by a nonlinear least-

Figure 3. Resonances for the Ala-methyl protons of the isomers of peptide 2b having the cis and trans conformations across the Tyr-Ala secondary amide peptide bond as a function of the length of the exchange time t in the inversion-magnetization transfer experiment. The sample was 22 mM peptide 2b in 90% H2O/10% D2O at pH 2.91 and 25 °C. The trans resonance was selectively inverted. The resonances for the cis isomer are plotted with a vertical scale expansion 1000× that of the resonances for the trans isomer. The exchange times were 0.0001, 0.0100, 0.1000, 0.2500, 0.4000, 0.6000, 0.8000, 1.0000, 1.5000, 2.0000, 3.0000, 4.0000, 5.0000, and 6.0000 s.

squares analysis.23 To illustrate, the nonlinear least-squares fit of the inversion-magnetization transfer spectra plotted in Figure 3 is presented in Figure 4; a value of 1.3 ( 0.08 s-1 was obtained for rate constant kct. The rate constant for trans-to-cis isomerization (ktc) at 25 °C was then calculated from Keq and kct (ktc ) kct/Keq). Rate constants for cis-to-trans and trans-to-cis isomerization for all of the peptides listed in Table 1 are summarized in Table 2. Rate constants for peptides 1a-4a and 2b were measured directly at 25 °C. Rate constants for peptides 1b, 3b, and 4b were measured at elevated temperatures; rate constants at 25 °C were then calculated using activation parameters obtained with the Eyring equation. Discussion Even though up to one-third of the bonds that form peptide and protein backbones are secondary amide peptide bonds, there is a paucity of data on the kinetics and thermodynamics of their cis-trans isomerization.1-3 The spectra in Figure 1 and the results in Table 2 demonstrate why this is the case. The resonance for the Ala-CH3 protons of the cis isomer is much

Isomerization of Secondary Amide Peptide Bonds

Figure 4. Integrated intensity of the resonance for the Ala-methyl protons of the cis isomer of peptide 2b as a function of the exchange time t in the inversion-magnetization transfer pulse sequence. The sample was 22 mM peptide 2b in 90% H2O/10% D2O at pH 2.91 and 25 °C. The smooth curve through the points is the theoretical curve obtained by nonlinear least-squares analysis of the data.

less intense than the Ala-12CH3 resonance for the trans isomer (Ala-12CH3 trans:cis ) 1016:1), and is even significantly less intense than the 13C-satellite resonances for the trans isomer, which each comprise 0.554% of the total intensity of the AlaCH3 resonance for the trans isomer. Furthermore, the very weak Ala-CH3 resonance for the cis isomer of peptide 2b is observable only because of differential shielding of the Ala-methyl cis and trans resonances by a ring current effect from the aromatic ring of the adjacent Tyr residue.3 The results reported in Table 2 indicate that the cis isomer of the Tyr-Ala, Phe-Ala, Ala-Tyr, and Ala-Phe peptide bonds of the linear and cyclic peptides in Table 1 is present at very low abundance, ranging from 0.070 to 0.12%, and that cyclization has relatively little effect on the population of the cis isomer. The results indicate that constraints imposed on the peptide backbone by cyclization also have relatively little effect on the dynamics of cis-trans isomerization: rates for both cis-to-trans and trans-to-cis interchange of the secondary amide peptide bonds of the Tyr-Ala and Phe-Ala sequence isomers are essentially the same for the linear and cyclic forms of peptides 2 and 4, while those for the Ala-Tyr and Ala-Phe peptide bonds of the cyclic forms of peptides 1 and 3 are decreased by only a factor of ∼2-3. For both the linear and cyclic forms of the peptides, the rates of cis-to-trans isomerization of the Phe-Ala, Ala-Phe, Tyr-Ala, and Tyr-Ala peptide bonds are approximately a 1000-fold faster than the rates of trans-to-cis isomerization. Thus, while cis-trans isomerization of secondary amide peptide bonds causes conformational heterogeneity along the backbone of peptides and unfolded proteins, the vast majority of each secondary amide peptide bond is in the trans conformation, and when there is a trans-to-cis interchange, it is quickly followed by a cis-to-trans interchange to the much more stable trans conformation. For example, the trans isomer of the Ala-Tyr peptide bond of peptide 1a has a t1/2 value of 433 s as compared to 0.347 s for the cis isomer. The trans isomer of the Ala-Tyr peptide bond of 1b is even kinetically more stable, with a t1/2 value of 1540 s, while t1/2 of the cis isomer is 1.16 s. Nevertheless, the relatively slow rate (on the protein folding time scale) of cis-to-trans interchange can cause complex protein folding kinetics. As mentioned above, 5% of the molecules in

J. Phys. Chem. B, Vol. 114, No. 9, 2010 3391 a proline-free variant of the 74 amino acid protein Tendamistat fold slowly, with a rate constant of 2.5 s-1, due to slow cis-totrans isomerization of the secondary amide peptide bonds in the cis conformation.11 If the secondary amide peptide bonds in the cis conformation are uniformily distributed along the backbone of the unfolded protein, a slow folding rate for 5% of the unfolded protein corresponds to 0.07% of each secondary amide bond in the cis conformation, in qualitative agreement with the populations found for the cis isomers of the secondary amide peptide bonds in the present study. It is of interest to compare the kinetics and equilibria of cis/ trans isomerization of the secondary amide peptide bonds to those of the much more abundant and well-studied Xaa-Pro tertiary amide peptide bonds. The population of the cis conformation of Xaa-Pro peptide bonds is typically in the 5-20% range for linear proline-containing peptides, while the population of the cis conformation in cyclic, disulfide-containing peptides covers a much wider range, depending on the amino acid sequence and size of the disulfide-containing ring, e.g., from 1.4% for the Cys-Pro peptide bond of the disulfide form of Ac-Cys-Pro-Thr-Cys-NH2 to 77% for the Cys-Pro peptide bond of the disulfide form of Ac-Cys-Pro-Phe-Ala-Ala-Ala-CysNH2.24 The much higher population of the cis isomer of XaaPro tertiary amide peptide bonds, as compared to the Phe-Ala, Tyr-Ala, Ala-Phe, and Ala-Tyr secondary amide peptide bonds in linear peptides, is due to a combination of increased kinetic stability of the cis conformation and decreased kinetic stability of the trans conformation of the Xaa-Pro peptide bond. For example, the average rate constant for cis-to-trans interchange for the Xaa-Pro peptide bond in 12 linear peptides, where Xaa ) Cys or Gly, is 0.047 s-1, as compared to the rate constants for trans-to-cis interchange in Table 2 that are some 8-43 times faster, while the average rate constant for trans-to-cis interchange for the 12 Xaa-Pro peptide bonds is 6.6 × 10-3 s-1, as compared to the values for ktc in Table 2 which range from 4 to 19 times slower.24,25 The increased kinetic stability of the trans isomer of the secondary amide peptide bonds, relative to that of the Xaa-Pro peptide bonds, is due to less steric interaction between the proton on the amide nitrogen of the secondary amide peptide bonds and the side chain of the adjacent amino acid as compared to steric interactions between the side chain of Xaa and the δ-CH2 group of the proline ring in the trans conformation of the Xaa-Pro peptide bond. Likewise, the increased kinetic stability of the cis conformation of the Xaa-Pro peptide bond is due to less steric interaction between the side chain of Xaa and the constrained proline ring, as compared to steric interactions between the two side chains of the Phe-Ala, Ala-Phe, TyrAla, and Ala-Tyr secondary amide peptide bonds. Conclusions The results presented here significantly increase the quantitative data available on the kinetics and equilibria of cis-trans isomerization of secondary amide bonds in peptides. The results indicate the degree of conformational heterogeneity along the backbones of peptides, and presumably unfolded proteins, and the time scales of interchange among the conformational isomers. A significant finding is that, although cyclization by disulfide bond formation imposes conformational constraints on the peptide backbones, cyclization has relatively little affect on the population of the sterically less-favored cis conformation and the rates of cis-trans isomerization of the Xaa-Yaa secondary amide peptide bonds. This suggests that the kinetics and equilibria of cis-trans isomerization of the Xaa-Yaa peptide bonds are governed primarily by factors localized to the dipeptide sequence.

3392

J. Phys. Chem. B, Vol. 114, No. 9, 2010

Acknowledgment. This work was supported in part by the University of California, Riverside. M.I. was supported by the Medical Scholars Summer Research Program, University of California, Riverside. The authors thank Dan Borchardt for assistance with the NMR measurements. References and Notes (1) Schiene-Fischer, C.; Fischer, G. J. Am. Chem. Soc. 2001, 123, 6227– 6231. (2) Li, P.; Chen, X. G.; Shulin, E.; Asher, S. A. J. Am. Chem. Soc. 1997, 119, 1116–1120. (3) Scherer, G.; Kramer, M. L.; Schutkowski, M.; Reimer, U.; Fischer, G. J. Am. Chem. Soc. 1998, 120, 5568–5574. (4) Nelson, D. L.; Cox, M. M. Lehninger Principles of Biochemistry; Worth Publ.: New York, 2000; pp161-163. (5) Ramachandran, G.; Sasisekharan, V. AdV. Protein Chem. 1968, 23, 283–438. (6) Weiss, M. S.; Jabs, A.; Hilgenfeld, R. Nat. Struct. Biol. 1998, 5, 676. (7) Stewart, D.; Sarkar, A.; Wampler, J. J. Mol. Biol. 1990, 214, 253– 260. (8) MacArthur, M.; Thornton, J. J. Mol. Biol. 1991, 218, 397–412. (9) Jabs, A.; Weiss, M. S.; Hilgenfeld, R. J. Mol. Biol. 1999, 286, 291– 304. (10) Pal, D.; Chakrabarti, P. J. Mol. Biol. 1999, 284, 271–288. (11) Pappenberger, G.; Aygu¨n, H.; Engels, J. W.; Reimer, U.; Fischer, G.; Kiefhaber, T. Nat. Struct. Biol. 2001, 8, 452–458.

Nguyen et al. (12) Odefey, C.; Mayr, L.; Schmid, F. X. J. Mol. Biol. 1995, 245, 69– 78. (13) Mayr, L. M.; Willbold, D.; Rosch, P.; Schmid, F. X. J. Mol. Biol. 1994, 240, 288–293. (14) Jeng, M.-F.; Campbell, A. P.; Begley, T.; Holmgren, A.; Case, D. A.; Wright, P. E.; Dyson, H. J. Structure 1994, 2, 853–868. (15) King, D. S.; Fields, C. G.; Fields, G. B. J. Peptide Protein Res. 1990, 115, 1497–1498. (16) Pearson, D. A.; Blanchette, M.; Baker, M. L.; Guindon, C. A. Tetrahedron Lett. 1989, 30, 2739–2742. (17) Shi, T.; Rabenstein, D. L. J. Am. Chem. Soc. 2000, 122, 6809– 6815. (18) Bothner-By, A.; Stephens, R. L.; Lee, J.; Warren, C. D.; Jeanloz, R. W. J. Am. Chem. Soc. 1984, 106, 811–813. (19) Bax, A.; Davis, D. G. J. Magn. Reson. 1985, 63, 207–213. (20) Krishnamurthy, V. V. Magn. Reson. Chem. 1997, 35, 9–12. (21) Kaerner, A.; Rabenstein, D. L. Magn. Reson. Chem. 1998, 36, 601– 607. (22) Robinson, G.; Kuchel, P. W.; Chapman, B. E. J. Magn. Reson. 1985, 63, 314–319. (23) Mariappan, S. V. S.; Rabenstein, D. L. J. Magn. Reson. 1992, 100, 183–188. (24) Shi, T.; Spain, S. M.; Rabenstein, D. L. Angew. Chem., Int. Ed. 2006, 45, 1780–1783. (25) Shi, T.; Spain, S. M.; Rabenstein, D. L. J. Am. Chem. Soc. 2004, 126, 790–796.

JP1000286