6894
J. Phys. Chem. B 2009, 113, 6894–6897
Amide I Bands of Terminally Blocked Alanine in Solutions Investigated by Infrared Spectroscopy and Density Functional Theory Calculation: Hydrogen-Bonding Interactions and Solvent Effects Maeng-Eun Lee,†,⊥ So Yeong Lee,‡,⊥ Sang-Woo Joo,*,§ and Kwang-Hwi Cho*,| CAE Team, R&D Center, Samsung SDI Company, Limited, Yongin-si, Gyeinggi-do 446-577, Korea, Department of Pharmacology, College of Veterinary Medicine, Seoul National UniVersity, Seoul 151-742 Korea, and Department of Chemistry, Department of Bioinformatics and Life Science and CAMDRC, Soongsil UniVersity, Seoul 156-743 Korea ReceiVed: NoVember 18, 2008; ReVised Manuscript ReceiVed: February 23, 2009
Structural aspects of terminally blocked alanine trans-N-acetyl-L-alanyl-trans-N′-methylamide (Ac-Ala-NHMe) in several different solvents were compared by attenuated total reflection infrared (ATR-IR) spectroscopy and density functional theory (DFT) calculations. The amide I bands between 1600 and 1700 cm-1 appeared to change depending on media, indicating dissimilar hydrogen-bonding interactions among the peptides and solvent molecules. The minimum energy geometry in the isolated gas phase and aqueous environments were calculated at the B3LYP/6-311++G** theoretical level. In the solid state, Ac-Ala-NHMe is assumed to have an extended β-stranded structure (C5), whereas it is assumed to have a cyclic structure (C7eq or RL) in a nonpolar tetrahydrofuran (THF) solvent. The optimized backbone dihedral angles (Φ, Ψ) of Ac-Ala-NHMe plus four explicit water molecules were estimated to be -94° and +133°, respectively, indicating the polyproline II structure (PII). The energy differences between the most stable conformers were predicted to be larger for Ac-Ala-NHMe, which implies that more conformational ensemble structures should coexist for the gas phase than for the aqueous medium with explicit water molecules. I. Introduction Recently, interest in peptides has increased because of their importance in pharmaceutical and biological sciences.1 Hydrogen bonds can serve as a structure-determining role in folded proteins or peptides. A polar solvent such as water affects peptide conformations by weakening the intrapeptide hydrogen bonds and other intramolecular electrostatic interactions.2-10 Vibrational spectroscopic tools such as Raman scattering, Raman optical activity, and vibrational circular dichroism have been employed extensively to investigate the structures of peptides.11,12 Infrared (IR) spectroscopy has been widely used to determine the secondary structures of polypeptides in solutions. The amide I band in the wavenumber region between 1600 and 1700 cm-1 is of interest because peaks in this region are known to be affected by the peptides’ secondary structures.11,12 Terminally blocked alanine trans-N-acetyl-L-alanyl-trans-N′methylamide (Ac-Ala-NHMe) has been considered an ideal model system for experimental and theoretical studies. Dynamics and multidimensional structures of the peptides have been studied using the two-dimensional (2D) IR pump-probe13 and liquid crystal nuclear magnetic resonance (NMR)14-16 techniques. The 2D IR and NMR results yielded the polyproline II structure with backbone dihedral angles (Φ, Ψ) in the range of (≈ -70°, ≈ +120°) and (≈ -85°, ≈ +160°), respectively. Several molecular dynamics simulation packages such as * To whom correspondence should be addressed. E-mail:
[email protected] (K.-H.C.);
[email protected] (S.-W.J.). † Samsung SDI Company. ‡ Seoul National University. § Department of Chemistry, Soongsil University. | Department of Bioinformatics and Life Science and CAMDRC, Soongsil University. ⊥ These authors equally contributed to this work.
Figure 1. Minimum energy structures of Ac-Ala-NHMe with B3LYP/ 6-311++G** (a) without water in the gas phase and (b) with four water molecules in the gas phase starting from SCRF-optimized structures.
AMBER, GROMOS, and CHARMM, however, have predicted dissimilar structures depending on the force fields.17 In this research we studied the solvation dynamics and hydrogen-bonding interaction for one of the simplest peptides by means of attenuated total reflection infrared (ATR-IR) spectroscopy in several different solvent environments. Simulation of the IR spectra of solvated peptides was performed to provide a better understanding of the hydrogen bonding of proteins in water using ab initio density functional theory (DFT) calculations. II. Experiments and Calculations The Ac-Ala-NHMe (97%) as depicted in Figure 1 was synthesized by Thermopeptide (New York, U.S.A.) or Peptron (Daejeon, Korea). The synthesized peptide was purified using
10.1021/jp810153w CCC: $40.75 2009 American Chemical Society Published on Web 04/17/2009
Amide I Band of Ac-Ala-NHMe in Different Solvents a Shimadzu prominence high-performance liquid chromatograph (HPLC) equipment and tested by a Hewlett-Packard 1100 series liquid chromatograph/mass spectrometer (LC-MS). The molecular weight of Ac-Ala-NHMe is 145, and the concentration of millimoles per liter is determined to be its mass dissolved in the solvent. The solubilities of Ac-Ala-NHMe in dimethyl sulfoxide (DMSO), tetrahydrofuran (THF), and water were estimated to be ∼104, 56, and 22 mM, respectively. All the chemicals otherwise specified were reagent-grade, and triply distilled water was used in making aqueous solutions. The infrared spectra were obtained using a Fourier transform infrared (FT-IR) spectrometer with a maximum resolution of 0.09 cm-1 (Thermo Nicolet 6700). The nominal resolution of our infrared measurements was estimated to be 4 cm-1. Data processing was carried out using the OMNIC v7.3 software. A portion of the peptide-dissolved liquid sample was transferred onto a heating ZnSe crystal of a Pike Technology Miracle external reflection accessory.18 The total 256 scans were averaged to obtain the spectrum. The spectral fitting of the overlapped peaks was carried out using a Peakfit v4.11 software. The elapsed time for 256 scan is less than 5 min. We attempted to obtain a good quality of infrared spectrum by appropriately subtracting the background spectrum and purging the air with the carbon dioxide free dry air frequently. We carried out molecular dynamics simulations in combinations with the DFT calculations at the B3LYP/6-311++G** level. All ab initio molecular orbital calculations were carried out using the Gaussian03 suite of programs.19 The geometry optimizations were performed on a total of 144 conformations. The range of the dihedral angles Φ and Ψ of the conformations are between -180 and 180° at the intervals of 30°. After obtaining the seven local minimum conformations out of 144 initial conformations, the structures are further minimized with B3LYP/6-311++G** level of theory in the gas phase and in the implicit water solvent using the SCRF (self-consistent reaction field) method. For each case, the seven minimum structures converged into five minimum conformations. For the gas-phase results, the calculation of infrared vibrational frequencies was carried out at B3LYP/6-311++G** level of the theory with the “Freq” option. The calculation represents the peptides in nonpolar environments. For the SCRF result, simple molecular dynamic simulation has been performed with explicit water in order to find the proper hydrogen-bond network between the peptide and water molecules. The minimum energy structures of Ac-Ala-NHMe with four water molecules were found starting from SCRF-optimized structures. With the explicit water molecules, the calculation of infrared vibrational frequencies was carried out at CPCM-B3LYP/6311++G** level of the theory with the “Freq” option. The results were compared with the results from the gas phase without water molecules. The calculation mimicked the peptides in aqueous solvent. The geometry optimizations were converged to certain structures. Simulations of the infrared spectra of solvated peptides were carried out to provide better understanding of the hydrogen bonding of proteins in solution. III. Results and Discussion III.A. IR Spectroscopy. Figure 2 shows the amide I stretching region of IR data for Ac-Ala-NHMe. Considering that the dielectric ε constants of THF, DMSO, and water are 8, 46, and 78, respectively, the frequency positions of the amide I stretching band at 1630 to ∼1670 cm-1, directly indicate the hydrogenbonding interaction of Ac-Ala-NHMe in solvent environments.17 The spectral features did not appear to change significantly in
J. Phys. Chem. B, Vol. 113, No. 19, 2009 6895
Figure 2. IR spectra of Ac-Ala-NHMe in (a) the powdered state or in solution of (b) THF, (c) DMSO, and (d) water.
Figure 3. IR spectrum of Ac-Ala-NHMe in the solid state and the decovoluted spectra. Each component may be assigned to our DFTcalculated peak spectra for the gas phase.
the nonpolar solvent THF. The multiple structures, as shown in Figure 2b, indicate the peptide in a nonpolar THF solvent should be somewhat similar to the environment in the gas phase. In DMSO and distilled water, the spectral positions are solvatochromically red-shifted and observed at 1664 and 1632 cm-1, respectively. These results indicate that the amide I bands should provide information on dissimilar hydrogen-bonding interactions between the peptides and solvent molecules. In the gas phase and THF, the amide stretching bands were seen to separate into several bands. The doublet feature was assumed to originate from heterogeneity of peptide N-methyl acetamide solvation structures.2,3 Figure 3 shows the fitted peak of Ac-Ala-NHMe in its solid state. The two amide I bands appeared at 1635 and 1665 cm-1 in the solid spectrum of which may be ascribed to the C-terminal and N-terminal vibrational frequencies, respectively. The carbonyl oxygen of a peptide molecule can form two hydrogen bonds with protic solvent molecules. The two peaks at 1642 and 1629 cm-1 could be assigned to the acetyl (N-terminal) and amino (C-terminal) ends of the peptide.9 The transition dipole strengths for an amide I mode were 0.40 debye for the 1642 cm-1 and 0.38 debye for the 1629 cm-1.9 The intensity of the band at 1635 cm-1 was observed to be larger than that at 1665 cm-1. To better understand the origin of this splitting and make appropriate peak assignments of the amide bands, we performed a DFT calculation. III.B. DFT Calculation. The geometries of Ac-Ala-NHMe are exhibited in Table 1. The predicted energy differences in
6896
J. Phys. Chem. B, Vol. 113, No. 19, 2009
Lee et al.
TABLE 1: Optimized Geometries of Ac-Ala-NHMe in (a) the Gas Phase and (b) the Aqueous Environments (a) Gas Phase (B3LYP/6-311++G**) (Φ, Ψ)
secondary structurea
energy (hartree)
relative energy (kcal/mol)
(-83.4, 76.1) (-154.7, 160.0) (73.1, 57.0) (-112.5, 13.1) (-165.7, -44.6)
C7eq C5 RL R2 R′
-496.01233195 -496.01013322 -496.00746057 -496.00451090 -496.00367574
0.00 THF 1.01 solid 2.43 THF 2.60 6.45
(b) with Water (B3LYP/6-311++G**) (Φ, Ψ)
secondary structurea
energy (hartree)
relative energy (kcal/mol)
(-93.9, 133.2) (64.4, -120.6) (76.7, -120.2) (-118.7, 26.1) (62.0, 32.3)
PII RD RD R2 RL
-801.915282 -801.908537 -801.900092 -801.895294 -801.883725
0.00 4.23 9.53 12.54 19.80
a
Figure 4. IR spectrum of Ac-Ala-NHMe in THF and the decovoluted spectra. Each component may be assigned to our DFT-calculated peak spectra for the gas phase.
Based on ref 21.
the gas phase are close to one another. In our calculated IR vibrational intensities in the gas phase, the CdO band at higher wavenumber (∼1660 cm-1) is smaller than that at lower wavenumber (∼1640 cm-1) and should correspond to only the second most stable conformer in Table 1. For this second most stable conformer, the CdO band intensity at higher wavenumber (∼1660 cm-1) is calculated as 4.9 times smaller than that of the lower wavenumber (∼1640 cm-1). This conformer suggests an extended β structure. The two amide stretching peak features should indicate an extended β structure. In a solid state, the neighboring peptide strands should interact with one another, resulting in an extended β structure. The prediction was also found to be in agreement with the multiple structures shown in Figures 4 and 5. In the THF solution, the amide I stretching region of Ac-Ala-NHMe mainly could be fitted to the four spectral features shown in Figure 4. The vibrational frequencies of IR experiments and DFT calculations are compared in Table 2. Considering that the spectral feature of Ac-Ala-NHMe in THF appeared to be similar to that in the powdered sample, the hydrogen bonding should be an important factor to determine the spectral line of the amide I band. The two vibrational bands at 1673 and 1634 cm-1 in THF for the amide I stretching regions can be assigned to the CdO stretching bands for the N-terminal and C-terminal site,
Figure 5. IR spectrum of Ac-Ala-NHMe in aqueous environments and the decovoluted spectra. Each component may be assigned to our DFT-calculated peak spectra for the four associated water molecules.
respectively, from our current DFT calculations. These features could be attributed to the most and the third most stable conformers for the C7eq and the RL structure as listed in Tables 1 and 2. Considering that the energy difference between the most and the second most stable structures is predicted to be as small as ∼1 kcal/mol, there should be a mixtures of the two conformers in a nonpolar THF solution. For the most stable conformer, the CdO band intensity at a higher wavenumber (∼1674 cm-1) is calculated to be only 2.1 times larger than that at a lower wavenumber (∼1639 cm-1). For the third most stable conformer, the CdO band intensity at a higher wavenumber (∼1681 cm-1) is calculated to be only 3.5 times larger than that at a lower wavenumber (∼1657 cm-1). In our fitted spectra, the C-terminal CdO band intensity at ∼1636 cm-1 appeared to be comparable with that of ∼1672 cm-1 for the N-terminal CdO band. The band intensity at
TABLE 2: Observed and Calculated Vibrational Frequencies of Amide I Bands of Ac-Ala-NHMe IR experimenta solid
THF
DFT calculationb water
gas
a
1693 1689e 1662
1697
CdO (N-, C-terminal) + H2O (1,2,3) bending CdO (N-terminal)e CdO (N-terminal) + H2O (1,3) bending CdO (N-terminal)f CdO (N-terminal)g H2O (3,4) bending CdO (C-terminal)e H2O (1,2) bending CdO (C-terminal) + H2O (1,4) bending CdO (C-terminal)f CdO (C-terminal)g
1660 1674f 1662g
1665g 1650 1650e
1652 1657e
1650 1633 1636f
assignmentd
1681e
1672f
1635g
waterc
1651 1637 1639 f 1644g
a Unit: cm-1. b Level: B3LYP/6-311++G**. c The number in water is calculated from the structure drawn in Figure 1b; scale factor: 0.995. Mainly from the current DFT calculation. e The third most stable conformer in the gas phase; scale factor: 0.970. f The most stable conformer in the gas phase; scale factor: 0.960. g The second most stable conformer in the gas phase; scale factor: 0.955. d
Amide I Band of Ac-Ala-NHMe in Different Solvents ∼1650 cm-1 appeared to be comparable with that of ∼1689 cm-1. This difference may be due to the dissimilar solvent interactions for the N-terminal and C-terminal CdO stretching bands. It is expected that the N-terminal band intensities may be more interacted and thus reduced than the C-terminal band after the interaction with the nonpolar solvent. By introducing four explicit water molecules, the optimized energy differences were determined to differ more than the cases in the gas phase. The optimized backbone dihedral angles (Φ, Ψ) of Ac-Ala-NHMe plus four explicit water molecules were estimated to be -94° and +133°, respectively. The results appear to be consistent with the values of -70° ( 25° and +120° ( 25° from the previous 2D IR experiment indicating the polyproline II structure. As listed in Table 2, the predicted vibrational positions of the amide I band at 1697, 1660, 1652, 1651, and 1637 cm-1 with four water molecules for the most stable conformer appeared to match well with the experimental values of 1693, 1662, 1650, 1650, and 1633 cm-1 in aqueous solutions by introducing appropriate scaling factors. The estimated intensities of the five vibrational bands at 1637, 1651, 1652, 1660, and 1697 cm-1 are 4.8, 1.0, 1.7, 2.4, and 2.2, respectively. The several vibrational band between 1650 and 1660 cm-1 overlapped considerably. Our results indicate that IR measurements and quantum chemical calculations can predict and explain the structures of the simplest terminal-based alanine in a successful manner. The calculated results are similar to the previous results with MP2/cc-pVTx//MP2/6-31G** calculation,20 where C7eq and C5 are the most stable structures for the gas phase and in water, respectively. C7eq in the gas phase is consistent with our results, whereas C5 in aqueous solution is somewhat conflicting. This discrepancy derives from us using explicit water molecules with the CPCM solvation model, whereas the previous calculation used only an implicit solvation model.20 Because our result agrees well with the spectral data, we presume that the minimum energy conformation in the polar solvent is not C5 but PII. Our data are also in agreement with the previous 2D IR and NMR results13-16 that yielded the PII structure with backbone dihedral angles (Φ, Ψ) in the range of (≈ -70°, ≈ +120°) and (≈ -85°, ≈ +160°), respectively. IV. Summary and Conclusions We examined solvent-dependent hydrogen-bonding interactions for one of the simplest peptides Ac-Ala-NHMe by comparing IR experiments and DFT calculations. The optimized backbone dihedral angles (Φ, Ψ) of Ac-Ala-NHMe plus four explicit water molecules were estimated to be -94° and +133°, respectively. The calculated vibrational frequencies of the amide
J. Phys. Chem. B, Vol. 113, No. 19, 2009 6897 I band were determined to match well with the ATR-IR experimental values. Acknowledgment. This work was supported by the Nano R&D program through the Korea Science and Engineering Foundation funded by the Ministry of Education, Science and Technology (2008-04308), the Soongsil University Research Fund, and the Korea Research Foundation Grant (KRF-2006331-C00138). References and Notes (1) Barth, A.; Zscherp, C. Q. ReV. Biophys. 2002, 35, 369–430. (2) Woutersen, S.; Mu, Y.; Stock, G.; Hamm, P. Chem. Phys. 2001, 266, 137–147. (3) Eaton, G.; Symons, M. C. R.; Rastogi, P. P. J. Chem. Soc., Faraday Trans. 1 1989, 85, 3257–3271. (4) Bour, P.; Keiderling, T. A. J. Chem. Phys. 2003, 119, 11253–11262. (5) Colley, C. S.; Griffiths-Jones, S. R.; George, M. W.; Searle, M. S. Chem. Commun. 2000, 593–594. (6) Dado, G. P.; Gellman, S. H. J. Am. Chem. Soc. 1994, 116, 1054– 1062. (7) Kubelka, J.; Turner, D. R. J. Phys. Chem. B 2007, 111, 1834– 1845. (8) Plass, M.; Griehl, C.; Kolbe, A. J. Mol. Struct. 2001, 570, 203– 214. (9) Torii, H.; Tasumi, M. J. Chem. Phys. 1992, 96, 3379–3387. (10) Kang, Y. K.; Scheraga, H. A. J. Phys. Chem. B 2008, 112, 5470– 5478. (11) Mantsch, H. H.; Chapman, D. Infrared Spectroscopy of Biomolecules; Wiley-Liss: New York, 1996. (12) Gremlich, H.-U.; Yan, B. Infrared and Raman Spectroscopy of Biological Materials; Marcel Dekker: New York, 2000. (13) Kim, Y. S.; Wang, J.; Hochstrasser, R. M. J. Phys. Chem. B 2006, 110, 18834–18843. (14) Madison, V.; Kopple, K. D. J. Am. Chem. Soc. 1980, 102, 4855– 4863. (15) Poon, C-D.; Samulski, E. T.; Weise, C. F.; Weisshaar, J. C. J. Am. Chem. Soc. 2000, 122, 5642–5643. (16) Mehta, M. A.; Fry, E. A.; Eddy, M. T.; Dedeo, M. T.; Anagnost, A. E.; Long, J. R. J. Phys. Chem. B 2004, 108, 2777–2780. (17) Arnautova, A. Y.; Jagielska, A.; Scheraga, H. A. J. Phys. Chem. B 2006, 110, 5025–5044. (18) Lim, J. K.; Lee, Y.; Lee, K.; Gong, M-s.; Joo, S. W. Chem. Lett. 2007, 36, 1226–1227. (19) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; HeadGordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 03, revision A.1; Gaussian, Inc.: Pittsburgh, PA, 2003. (20) Wang, Z.-X.; Duan, Y. J. Comput. Chem. 2004, 25, 1699–1716. (21) Ramakrishnan, C.; Ramanchandran, G. N. Biophys. J. 1965, 5, 909–933.
JP810153W
