External Chirality-Triggered Helicity Control Promoted by Introducing a

We here aim to promote the NCDE on peptide helicity using two types of ... Chain: Allosteric Control of Asymmetry of the C-Terminal Site by External M...
0 downloads 0 Views 473KB Size
Download PDF
Biomacromolecules 2004, 5, 1231-1240

1231

External Chirality-Triggered Helicity Control Promoted by Introducing a β-Ala Residue into the N-Terminus of Chiral Peptides Yoshihito Inai* and Hisatoshi Komori Department of Environmental Technology and Urban Planning, Shikumi College, Graduate School of Engineering, Nagoya Institute of Technology, Gokiso-cho, Showa-ku, Nagoya 466-8555, Japan Received October 9, 2003; Revised Manuscript Received January 15, 2004

The noncovalent chiral domino effect (NCDE), defined as chiral interaction upon an N-terminus of a 310helical peptide, will provide a unique method for structural control of a peptide helix through the use of external chirality. On the other hand, the NCDE has not been considered to be effective for the helicity control of peptides strongly favoring a one-handed screw sense. We here aim to promote the NCDE on peptide helicity using two types of nonapeptides: H-β-Ala-∆ZPhe-Aib-∆ZPhe-X*-(∆ZPhe-Aib)2-OCH3 [∆ZPhe ) R,β-didehydrophenylalanine, Aib ) R-aminoisobutyric acid], where X* as the single chirality is L-leucine (1) or L-phenylalanine (2). NMR, IR, and CD spectroscopy as well as energy calculation revealed that both peptides alone form a right-handed 310-helix. The original CD amplitudes or signs in chloroform, irrespective of a strong screw-sense preference in the central chirality, responded sensitively to external chiral information. Namely added Boc-L-amino acid stabilized the original right-handed helix, while the corresponding D-isomer destabilized it or transformed it into a left-handed helix. These peptides were also shown to bind more favorably to an L-isomer from the racemate. Although similar helicity control was observed for analogous nonapeptides bearing an N-terminal Aib residue (Inai, Y.; et al. Biomacromolecules 2003, 4, 122), the present findings demonstrate that the N-terminal replacement by the β-Ala residue significantly improves the previous NCDE to achieve more effective control of helicity. Semiempirical molecular orbital calculations on complexation of peptide 2 with Boc-(L or D)-Pro-OH reasonably explained the unique conformational change induced by external chirality. Introduction A helix is chiral in nature.1 Thus, external chiral signals to a helical chain, in response to their chiral signs, would influence the original helical stability or screw sense. This phenomenon has been elegantly demonstrated in artificial, optically inactive helical chains, the helicity of which is induced by chiral stimuli such as additive, solvent, and light.2 In constant, little is known about examples that actively apply such noncovalent chiral interaction to structural modulation of biological polymers.2h Recently, we have proposed novel chiral interactions upon a 310-helical peptide as “noncovalent chiral domino effect (NCDE)”.3 The NCDE enables dynamic control of the helicity or helical stability of a peptide through mixing with a chiral molecule capable of binding to the N-terminus; i.e., either chiral sign of the external molecule promotes original helicity, whereas the opposite sign reduces it or reverses the screw sense.4 Thus, this effect might provide a special method for modulating asymmetric biological structures through the use of external chirality. However, the NCDE-driven helicity control was not so dramatic in chiral peptides that strongly prefer a one-handed helix. For instance, an N-terminal-free nonapeptide having * To whom correspondence should be addressed. E-mail: [email protected].

inai.

an L-leucine (L-Leu) residue in the central position adopts a right-handed helix, the high stability of which is not drastically influenced by the NCDE.4a Namely, although the addition of Boc-L-amino acid (Boc ) t-butoxycarbonyl) somewhat stabilizes the original right-handed helix, the corresponding D-isomer, which will promise helix destabilization, does not affect the original stability essentially. Also, an N-terminal-free decapeptide containing an L-L doublet at the C-terminus has a strong preference for a right-handed helicity, which is substantially retained by mixing with chiral species.4b In particular, neither chiral sign of additives gives rise to significant reduction in the original helicity. These two examples4 suggest that the NCDE might not reduce a screw-sense bias in a stable helical structure, where a covalent chiral factor determines the whole screw sense to obstinately resist external chirality. Very recently, amino acid substitution at the N-terminus of an achiral nonapeptide was shown to largely influence the chiral induction,5 i.e., CD amplitude induced for the nonapeptide almost doubled in the N-terminal replacement of R-aminoisobutyric acid (Aib) residue by β-alanine (βAla) residue. In addition, a higher sensitivity to the external chiral signal was observed in the β-Ala-terminal peptide.5 Theses facts well-characterize a unique nature of the NCDE that chiral complex produced around the N-terminus, on the basis of its steric fashion, controls the overall helicity. In

10.1021/bm0344001 CCC: $27.50 © 2004 American Chemical Society Published on Web 06/19/2004

1232

Biomacromolecules, Vol. 5, No. 4, 2004

Inai and Komori

Chart 1. H-β-Ala-∆ZPhe-Aib-∆ZPhe-X*-(∆ZPhe-Aib)2-OMe: X* ) L-leucine (L-Leu, 1) and L-phenylalanine (L-Phe, 2)

Figure 1. Competition between the internal chiral effect and the NCDE for helicity control in (A) H-β-Ala-∆ZPhe-Aib-∆ZPhe-X*-(∆ZPheAib)2-OMe [X* ) L-Leu (1) and L-Phe (2)] and (B) H-(Aib-∆ZPhe)2X*-(∆ZPhe-Aib)2-OMe4a [X* ) L-Leu (3) and L-Phe (4)]. Boc-amino acid was chosen as chiral carboxylic acid.

other words, chemical modification at the N-terminus belonging to an interactive site might promote a helicitycontrollable power of the NCDE. To improve the NCDE-based control of “chiral peptide’s helicity”, we here adopted two types of chiral nonapeptides 1 and 2 (Chart 1) possessing an N-terminal β-Ala residue noticed above. These are mainly composed of achiral helix-forming Aib6 and R,β-didehydrophenylalanine (∆ZPhe)7 residues. The centrally located L-residue will preferentially induce a onehanded screw sense through transferring its chiral information from the midpoint to both termini (Figure 1A). At this point, addition of chiral carboxylic acid, dependently on its chiral sign and power, will influence the original helicity, as illustrated in Figure 1A. This external chirality-induced helicity control is compared with that observed for the analogues where the only difference is the N-terminal Aib residue [Figure 1B]: H-(Aib-∆ZPhe)2-X*-(∆ZPhe-Aib)2-OMe [X* ) L-Leu (3) and L-phenylalanine (L-Phe) (4)].4a Consequently, the N-terminal replacement by β-Ala residue significantly promotes the previous NCDE to achieve effective control of helicity. Results and Discussion Helical Conformation of Peptides 1 and 2. A preferred conformation of the two β-Ala-terminal peptides was examined basically as described in ref 4a. Their FT-IR absorption spectra in chloroform ([1 or 2] ) 1 mM) yielded two peaks of amide I absorption band, i.e., 1658 and 1625 cm-1 for 1 and 1658 and 1626 cm-1 for 2. The complicated amide bands have been assigned to a helical sequence composed of saturated amino acid and ∆ZPhe residues,8 since the second peak at lower wavenumbers originates from partial resonance between carbonyl and styryl groups of ∆ZPhe residue.4,5,8a NOESY spectra of both peptides in CDCl3 at 293 K exhibited marked cross-peaks for the neighboring

Figure 2. Solvent-composition effects on NH chemical shifts of peptides 1 (A) and 2 (B) in CDCl3/(CD3)2SO systems in 200 MHz NMR: [peptide (1 or 2)] ) ca. 7 mM [prior to (CD3)2SO addition]; ca. 26 °C.

NH pairs [NiH-Ni+1H resonances; NN(i,i+1)] within the segment of Aib(3) to Aib(9). These sequential NOEs are assigned to a 310- helix or an R-helix.9a Meanwhile, NOE of CRH X*(5)-NH ∆ZPhe(6) [RN(5,6)] was fairly weak. This suggests a ψ5 value around 120°9b,c that leads to an extended conformation.6g,9 (Likewise, a minimum distance for RN(5,6) was obtained at ca. 120° (ψ5) through the ψ5 variation in each theoretical geometry of peptides 1 and 2, which will be given in Figure 4. Herein the RN(5,6) distance was within ca. 2.7 Å in ψ5 ) 60-180°. A similar relationship between the ψ5 value and RN(5,6) distance is seen in peptides 3 and 4.4a) The combination of strong NN(i,i+1) and weak RN(5,6) is another criterion for evidence of a helical structure.6g,9 A similar NOE pattern was observed in peptides 3 and 4.4a For discrimination between 310- and R-helices, solvent variation in NH chemical shifts was investigated basically according to refs 3a,c, 5, 6c,f,g, 7a, and 10. In both peptides (Figure 2), Aib(3)’s NH in CDCl3 is markedly shifted to a lower magnetic field by addition of strong hydrogen-accepting (CD3)2SO, whereas six NH’s resonances of ∆ZPhe(4) to Aib(9) residues are substantially unaffected due to intramolecular hydrogen bonding. N-Terminal and ∆ZPhe(2) NH’s resonances were not clearly observed in this condition. However, we can mention that the occurrence of hydrogenbonding ∆ZPhe(4) to Aib(9) NH’s and of free Aib(3) NH meets criteria for a 310-helix [(i+3)fi], but not an R-helix [(i+4)fi].11 The preference for a 310-helix in both peptides should arise from helix-forming nature inherent in Aib and ∆ZPhe residues.6,7,11 Also, analogous peptides 3 and 4 were shown to form a 310-helix in chloroform,4a thus supporting that N-terminal replacement of Aib by β-Ala does not influence a helical propensity of (Aib-∆ZPhe)-based peptides. Their preferred helix senses are identified from their CD spectra (Figure 3). Both peptides in chloroform afford a strong exciton splitting at around 284 nm; the split amplitudes are ca. 29 for 1 and ca. 18 for 2. The split pattern consisting of a negative first Cotton effect and a positive

Helicity Control of a Chiral Peptide by NCDE

Biomacromolecules, Vol. 5, No. 4, 2004 1233

Figure 3. CD (top) and UV absorption (bottom) spectra of peptides 1 and 2 in chloroform: [1] ) 0.13 mM; [2] ) 0.14 mM.

second Cotton effect (- to +) can be unambiguously assigned to a right-handed helix, according to refs 5 and 12. The screw sense preference originates from the centrally located L-residue that usually favors right-handedness.6f,13 The same helix sense is also found in peptides 3 and 4, where there is no significant difference between split CD amplitudes of the β-Ala- and Aib-terminal peptides, i.e., 3 (ca. 30) and 4 (ca. 18).4a Thus, the terminal replacement by β-Ala is again proven to retain the overall helicity and screw-sense bias, which originate primarily from the remaining common octapeptide sequences. Semiempirical molecular orbital calculation (the AM1 method14) strongly supports the experimentally proposed right-handed 310-helix. The energy-minimized helices (Figure 4) are characterized by the segment -∆ZPhe-Aib-∆ZPhe-X*∆ZPhe-Aib-∆ZPhe- with average torsional angles of 〈φ〉 ) -41.1°, 〈ψ〉 ) -37.4°, and 〈ω〉 ) 179.6° for 1, and 〈φ〉 ) -41.5°, 〈ψ〉 ) -38.1°, and 〈ω〉 ) 179.8° for 2. The following helix parameters are also evaluated: number of residues per turn (NR) ) 3.16 (1) and 3.21 (2); translation per residues along helix axis (TR) ) 1.92 (1) and 1.95 (2).15 These helix parameters are quite similar to those averaged for Aib-based 310-helices.11 Also analogous nonapeptide Boc-(Aib-∆ZPhe)4Aib-OMe in the crystalline state was recently proven to form a 310-helix with similar helix parameters.16 In addition, similar helical backbones are obtained for peptides 3 and 4.4a Therefore, peptides 1 and 2 are experimentally and theoretically shown to have a strong tendency to form a right-handed 310-helix. Control of Helicity through External Chirality. Peptide 1 or 2 alone adopts a right-handed 310-helix in chloroform due to its central L-residue. While to a chloroform solution of peptide 1 is added Boc-(L or D)-Pro-OH (Pro ) proline) as chiral triggers, the original CD pattern (- to +) is retained as shown in Figure 5. The CD amplitude, however, depends significantly on the external chiral sign; that is, the amplitude increases with Boc-L-Pro-OH (Figure 5A) while decreasing with the D-isomer (Figure 5B). Obviously, the original helical stability replies to the external chiral sign. Meanwhile, the addition of Boc-D-Pro-OH to analogous peptide 3 bearing an N-terminal Aib residue instead of the β-Ala residue did not reduce its original right-handed helicity.4a Thus, the

Figure 4. Side (top) and top (bottom) views of AM1-based14 energyminimized structures of peptides 1 (left) and 2 (right). Both structures adopt a right-handed 310-helix essentially, as mentioned in the text.

Figure 5. CD (top) and UV absorption (bottom) spectra of peptide 1 in chloroform containing (A) Boc-L-Pro-OH or (B) Boc-D-Pro-OH: [1] ) 0.13 mM; [Boc-Pro-OH] ) 0-10 mM.

N-terminal modification with a β-Ala residue can amplify the NCDE on helicity control. A more dramatic CD change was observed for peptide 2 having an L-Phe residue in the central position. As shown in Figure 6, the addition of Boc-L-Pro-OH markedly increases the original CD amplitude with its split pattern held, thus indicating that the original right-handed helicity is promoted by the chiral stimulus. In contrast, the D-isomer gives rise to a unique transition from the right-handed helix to a lefthanded helix (Figure 6B). Herein, the original CD amplitude of (- to +) decreases with an initial increase in [Boc-LPro-OH] (0-0.5 mM). Subsequently a reversed pattern (+ to -) appears at around 1 mM, and the amplitude then increases with further addition of Boc-D-Pro-OH (3-10 mM). It is apparent that the external chiral signal reduces

1234

Biomacromolecules, Vol. 5, No. 4, 2004

Inai and Komori Table 1. Split CD Values Observed for H-β-Ala-∆ZPhe-Aib-∆ZPheX*-(∆ZPhe-Aib)2-OMe with Carboxylic Acida

peptide (X*) 1 (L-Leu)

Figure 6. CD (top) and UV absorption (bottom) spectra of peptide 2 in chloroform containing (A) Boc-L-Pro-OH or (B) Boc-D-Pro-OH: [2] ) 0.14 mM; [Boc-Pro-OH] ) 0-10 mM.

the original right-handed helicity, finally leading to a lefthanded helix via equimolar amounts of both helical chains. Although peptide 4 capped by an N-terminal Aib residue also undergoes destabilization of the original right-handed helix through the addition of Boc-D-Pro-OH, it does not take a left-handed helix even under the presence of a large excess of the D-isomer.4a Meanwhile, a helix-to-helix transition induced by external chirality is found in H-(Aib-∆ZPhe)4L-Leu-OMe (5), where the original left-handed helix is converted into a right-handed one by the addition of BocL-Pro-OH.4b Under similar conditions (solvent, temperature, and peptide concentration), onset concentration of chiral BocPro-OH for helix inversion is much lower in peptide 2 (ca. 1 mM) than in peptide 5 (ca. 18 mM4b). Therefore, the N-terminal replacement by β-Ala residue effectively enhances not only the NCDE on helicity control but sensitivity to external chiral signal. The Boc-(L or D)-Pro-OH-induced screw-sense directions of peptide 2 agree with those observed in the achiral or chiral peptides.3-5 Similar helicity control for peptides 1 and 2 is achieved with other chiral Boc-amino acids [alanine (Ala), Leu, valine (Val), and Phe], as shown in Table 1 and Figures 7 and 8. Herein, Boc-L-amino acid gives rise to promotion of the original right-handed helicity, whereas the helicity reduction or helix inversion is induced by the corresponding D-isomer. Changes in the CD amplitudes are attributed to varying amounts of the right-handed helix among all conformers. In this regard, we previously suggested that the CD change originates mainly from various molar ratios of a right-handed helix to a left-handed helix.4a This tentative assignment should be confirmed by the fact that a large excess of BocD-amino acid finally induces a left-handed helix via a screwsense balance from the original right-handed helicity (Figure 6B). Chiral guest molecules might operate on not only an N-terminal sequence leading to the NCDE but also the other part of the chain. The latter interaction, however, is not significant in induction or control of helicity on the basis of the following reasons. (i) Although analogous peptides 3 and 4 undergo similar helicity control in reply to chiral sign of Boc-amino acids, the helicity of the corresponding N-Bocprotected peptides is not almost influenced by an extremely large excess (ca. 1 × 103 fold) of the same chiral molecules.4a

2 (L-Phe)

added acid none Boc-Gly-OH Boc-DL-Pro-OH Boc-L-Pro-OH Boc-L-Val-OH Boc-L-Leu-OH Boc-L-Ala-OH Boc-L-Phe-OH Boc-D-Pro-OH Boc-D-Val-OH Boc-D-Leu-OH Boc-D-Ala-OH Boc-D-Phe-OH none Boc-Gly-OH Boc-DL-Pro-OH Boc-L-Pro-OH Boc-L-Val-OH Boc-L-Leu-OH Boc-L-Ala-OH Boc-L-Phe-OH Boc-D-Pro-OH Boc-D-Val-OH Boc-D-Leu-OH Boc-D-Ala-OH Boc-D-Phe-OH

first Cotton effect

second Cotton effect

∆/wavelengthb

∆/wavelengthb

-9.5/303 -9.0/301 -10.9/302 -12.6/304 -12.1/303 -12.3/302 -11.9/302 -12.6/302 -4.5/302 -4.1/301 -3.9/301 -6.0/301 -7.5/301 -5.5/301 -5.7/303 -5.9/300 -9.6/304 -10.2/302 -10.7/302 -9.4/303 -10.0/303 +3.5/307 +3.0/308 +4.3/307

+19.0/269 +17.7/270 +21.6/270 +25.9/271 +25.5/271 +25.8/269 +24.7/269 +c +8.8/270 +9.1/269 +8.7/269 +11.8/269 +c +12.3/271 +11.5/270 +13.3/267 +21.0/270 +21.1/269 +22.0/270 +19.1/270 +c -3.9/277 -3.3/279 -5.1/269

d

-2.0/296

+c

a CD spectra were recorded using a JASCO J-600 in chloroform: [peptide (1 or 2)] ) 0.13-0.14 mM and [Boc-amino acid] ) 6 mM. b ∆ (M-1 cm-1); wavelength (nm). A more precise expression is adopted here for ∆ values below a decimal point. c Overlapped with CD signal of Boc(L or D)-Phe-OH. d Complicated split CD pattern of (+ to - to +) as shown in Figure 8B.

Figure 7. CD spectra of peptide 1 in chloroform containing (A) BocL-amino acid (Ala, Leu, Val, Phe, and Pro) or (B) the corresponding Boc-D-amino acid: [1] ) 0.13 mM; [Boc-amino acid] ) 6 mM.

Similar insensitivity to external chiral signals is observed for other N-Boc-protected chiral and achiral sequences.3,4b (ii) As for an optically inactive 310-helical H-β-Ala-(∆ZPheAib)4-OMe (6), the amino head and two free NHs of its N-terminal sequence (H-β-Ala-∆ZPhe-Aib-) are proven to bind a Boc-amino acid molecule.5 In contrast, all NH groups of the remaining segment -(∆ZPhe-Aib)3-OMe are shielded due to the (i+3)fi intramolecular hydrogen bonds.5 These facts strongly support that some interaction between Bocamino acid molecules and a 310-helical backbone other than

Helicity Control of a Chiral Peptide by NCDE

Figure 8. CD spectra of peptide 2 in chloroform containing (A) BocL-amino acid (Ala, Leu, Val, Phe, and Pro) or (B) the corresponding Boc-D-amino acid: [2] ) 0.14 mM; [Boc-amino acid] ) 6 mM.

its N-terminal sequence, even if present, is essentially “nonspecific” with respect to chirality. Concentration Dependence of External Chirality on Helicity Control. For more qualitative evaluation of the external chiral effect, the split CD amplitude of peptides 1 and 2 is plotted against chiral additive concentration (Figure 9, parts A and B, respectively). In each figure, there is a sharp increase or decrease in the CD amplitudes at [Boc-(D or L)-Pro-OH] ) 0-5 mM, while each amplitude reaches essentially a maximum or minimum around 6 mM. In both peptides with Boc-D-Pro-OH, the saturation values are directed to the original positions by further addition above 20 mM. A similar tendency is found in porphyrin tweezer17a or achiral peptide bearing an N-terminal β-Ala residue.5 Probably, a large excess of chiral guest might convert the 1:1 host-guest complex into a 1:2 complex,5,17 thereby reducing a chiral power capable of destabilizing the original helix. On the whole, Boc-D-Pro-OH decreases the original righthanded helicity or converts it into a left-handed helicity, whereas the L-isomer strongly promotes the original helicity. When achiral Boc-Gly-OH (Gly ) glycine) is regarded as a criterion for nonchiral effect, Boc-D-Pro-OH yields a larger amount of CD change from the criterion than does the L-isomer. This might be relevant to a screw-sense imbalance at the initial state, i.e., a right-handed helix, prior to helicity control by the L-species, is already being dominated by the internal chirality. The relations4a in peptides 3 and 4 versus Boc-(D and L)Pro-OH are superimposed therein. In a comparison between the β-Ala and Aib termini, two notable features can be seen. First, helicity of the β-Ala-terminal peptides is sufficiently promoted by a much lower concentration of Boc-L-Pro-OH, as being compared to that of the Aib-terminal peptides. The apparent binding constant (Kapp, M-1) for Boc-L-Pro-OH was estimated from the nonlinear fitting18 to be 3.4 × 103 for 1, 3.5 × 103 for 2, 38 for 3, and 26 for 4. Considerably higher binding affinity is achieved in the β-Ala-terminal peptides, suggesting that such a more flexible N-terminus is suitable for capturing the chiral molecule. A similar improvement in sensitivity to external chirality is seen in achiral nonapeptide by its N-terminal replacement of Aib by β-Ala.5 Second, CD amplitudes of the β-Ala-terminal peptides decrease dramatically with addition of Boc-D-Pro-OH, whereas those of the

Biomacromolecules, Vol. 5, No. 4, 2004 1235

Figure 9. Boc-amino acid-concentration effects on CD amplitudes of peptides 1 (A) and 2 (B) in chloroform containing various amounts of Boc-L-Pro-OH (solid red), Boc-D-Pro-OH (solid blue), Boc-Gly-OH (solid black), or Boc-DL-Pro-OH (D/L ) 50/50; solid green): [1] ) 0.13 mM; [2] ) 0.14 mM. Amplitude value is defined as difference between maximum intensities of first and second Cotton effects. Its positive sign corresponds to split CD pattern of (- to +), while negative sign represents that of (+ to -). (Theoretical interpretation of split-CD amplitudes as well as ∆ values has been deeply described in ref 12a,b.) The broken green curve stands for the arithmetic mean of the two values for Boc-L-Pro-OH and Boc-D-Pro-OH. For comparison, the titration curves of peptide 3 (A) or 4 (B) with Boc-L-Pro-OH (broken red) and Boc-D-Pro-OH (broken blue) are superimposed: these curves were obtained from nonlinear fitting18 of the original data.4a

corresponding Aib-terminal peptides keep essentially constant or somewhat decrease.4a It should be noted that there is no significant difference between the original CD amplitudes of β-Ala-terminal and Aib-terminal peptides. This means that an initial, strong bias to a right-handed screw sense is essentially the same in both peptides, suggesting that both backbones are prone to maintain the original helicity against external chirality. Nevertheless, in the β-Ala-terminal peptides (particularly in 2), the external chiral effect overwhelms the internal one. Therefore, the incorporation of a β-Ala residue into the N-terminus gives rise to remarkable promotion of the NCDE on helicity control (Figure 1). Another noteworthy phenomenon was observed in complexation between β-Ala-terminal peptide (1 or 2) and racemic guest. As shown in Figure 9, the CD amplitude in the presence of Boc-DL-Pro-OH (D/L ) 50/50; solid green) is somewhat larger over a wide range of Boc-Pro-OH concentration than the arithmetic mean (broken green) of the two amplitudes induced by the D- and L-isomers. Apparently, the right-handed helical chain of both peptides binds more preferentially to the L-isomer from the racemate. This direction of the preferential binding should be relevant to the fact that the L-isomer tends to promote the original right-handed helicity. A similar chiral discrimination is found in peptides 3 and 4,4a and N-terminal-free peptide having a C-terminal L-L doublet.4b These findings should arise from underlying functionality inherent in an N-terminal 310-helical segment,5 which possesses a prearranged structure for chiral binding. Although many helical backbones are capable of chiral discrimination and recognition upon their main chains and side chains,19 our chiral discrimination originates from one terminus of a helical chain. Theoretical Interpretation of Helicity Control. The helicity control originates from dynamic chiral complexation performing on the N-terminal segment of a 310-helix. The mechanism for the NCDE is already proven in an optically

1236

Biomacromolecules, Vol. 5, No. 4, 2004

Inai and Komori

Figure 10. Theoretical complex structures of peptide 2-Boc-Pro-OH, obtained from energy minimization of combinations of (A) right-handed helix (gray carbons)-L-isomer (blue carbons), (B) left-handed helix (orange carbons)-L-isomer, (C) right-handed helix-D-isomer (green carbons), and (D) left-handed helix-D-isomer. In all cases, the segment of -CO-∆ZPhe-Aib-∆ZPhe-Phe-∆ZPhe-Aib-NH- retains a 310-helix essentially. Energy-minimizations were performed by using the AM1 method in MOPAC97.14 The most stable complex is A, while the second low energy one is D. For complexation in A and D, the intermolecular hydrogen bonding is illustrated with the blue arrows, whereas the ionic interaction is indicated with the red arrows. For the detailed order of complexation stability, see Figure 11.

inactive 310-helical peptide 6.5 In brief, the amino terminus and two free NH’s of H-β-Ala-∆ZPhe-Aib- capture a Bocamino acid molecule cooperatively through three-point interactions, which produce a chiral environment leading to helicity control. A similar mechanism should be applied to chiral complexation in peptides 1 and 2, because they have the same N-terminal 310-helical sequence (H-β-Ala-∆ZPheAib-), and undergo induction or promotion of the same helicity as peptide 6 through a chiral sign of Boc-amino acid.5 We here estimated theoretical complex structures of peptide 2 versus Boc-Pro-OH and their energy values using the semiempirical MO calculation.14 A peptide 2 molecule was assumed to adopt either a left-handed helix or a righthanded helix but not to take other conformers.20 First, energy minimization of peptide 2 alone was performed to yield two conformers essentially assignable to right-handed and lefthanded 310-helices. The right-handed helix (Figure 4, the right side figure) was estimated to be about 1.3 kcal mol-1 more stable than the left-handed one, being consistent with the experimentally observed screw-sense preference. The energyminimized left-handed helix was characterized by helix parameters15 of NR ) 3.19 and TR ) 1.95. Second, a BocL-Pro-OH (or Boc-D-Pro-OH) molecule was placed on the N-terminal sequence of the right-handed helix (or the lefthanded helix) at relatively close positions, in which energy minimization will lead to the three-point interactions,5 i.e., salt-bridge formation between the amino terminus and the carboxyl group, and two hydrogen bonds of the ∆ZPhe(2) NH versus the urethane carbonyl O’s and of the Aib(3) NH versus either of carboxylate O’s. Such placement could be readily produced only through conformational alternation of the β-Ala residue (θ ) (66°).5,22,23 Each of the two complexes was energy-minimized with variation of all bond lengths, bond angles, and torsion angles, except for the three ammonium N-H bond lengths that were set to 1.025 Å24 to maintain the ionic state of -COO-‚NH3+-. From these energy-minimized complexes, initial conformations for the remaining two diastereomeric complexes were generated

primarily through configurational inversion around CR atom of Boc-Pro-OH. Subsequently they were energy-minimized as mentioned above. The resulting four energy-minimized complexes are depicted in Figure 10. They support substantially the threepoint interactions upon the N-terminal sequence of H-β-Ala∆ZPhe-Aib-, thus implying that each complex is regarded as being one of local energy minima. However, there is a striking difference among their stabilities. When a Boc-LPro-OH molecule operates on peptide 2, a right-handed 310helix binds more preferentially to the L-isomer (Figure 10A) than does a left-handed one (Figure 10B). The instability of the left-handed helix with the L-isomer (Figure 10B) should arise from a sterically intricate interaction in which the Boc group stands close to the ∆ZPhe(2) residue, with the Pro pyrrolidine face directed to the helical axis. This steric strain should vanish in the diastereomeric pairing (Figure 10A), which allows the Boc group to be placed at a free space over the N-terminal segment, with the pyrrolidine face parallel to the helical rod. Conversely, a Boc-D-Pro-OH isomer binds more favorably to a left-handed helix (Figure 10D) than a right-handed one (Figure 10C). The energy gap between both helices can also be explained from a different steric effect in N-terminal complexation. Namely, complex formation of a right-handed helix with the D-isomer (Figure 10C) enforces spatial proximity of the Boc group to the ∆ZPhe(2) residue, whereas the diastereomeric coupling (Figure 10D) shifts the Boc group to a space of the N-terminal segment. For a closer analysis, an energy diagram of the four complexes is illustrated together with that of peptide 2 alone in Figure 11. Prior to complexation, the right-handed helix is predicted to be about 1.3 kcal mol-1 more stable than the left-handed one, as mentioned above. When each helix binds to a Boc-L-Pro-OH molecule, the original energy gap significantly expands to ca. 4.9 kcal mol-1 with a preference for a right-handed helix held. Thus, the experimental fact that the L-isomer stabilizes the original right-handed helix

Helicity Control of a Chiral Peptide by NCDE

Biomacromolecules, Vol. 5, No. 4, 2004 1237

of the original helicity. While some oligopeptides were found to stabilize an R-helical structure,25 our NCDE with chiral amino acid derivatives is attributed to terminal chiral recognition leading to chirality transfer on a helical chain. As previously mentioned,5 the NCDE originates from a 310-helical N-terminal motif that involves three functional groups (amino head and two free amide NH’s). These functionalities give rise to asymmetric coordination with an external chiral molecule, and subsequently, “chirality” produced at the N-terminus stimulates the overall helicity as well as screw-sense bias. In this regard, the present study demonstrates that one can modulate the helicity-controllable factor of and sensitivity to external chirality effectively through artificial modification upon the N-terminal sequence of a 310-helical peptide. Figure 11. Relative energy diagram of the theoretical complexes of peptide 2-Boc-Pro-OH shown in Figure 10. The energy level of peptide 2 alone is also illustrated. In each state of complexation and isolation, the most stable one is set to 0 kcal mol-1. RH and LH stand for right-handed and left-handed helices respectively, and L and D represent chirality of Boc-Pro-OH.

is theoretically interpreted as the energy gap expanded by complexation with the L-species. In contrast, when the D-species couples with each original helix, the resulting left-handed helical complex is energetically preferred, thereby leading to helix inversion experimentally observed. The energy gap between both handed helices combined with the D-isomer is estimated to be ca. 1.7 kcal mol-1, much smaller than the aforementioned gap (ca. 4.9 kcal mol-1) for complexation with the L-species. This implies that contribution of a second-low energy complex to the lowest-energy complex becomes more prominent in the case of D-isomer. This theoretical prediction should reflect a substantial difference between the L- and D-isomer-induced CD amplitudes, because each second-low energy complex produces split CD signs opposite to the corresponding lowestenergy complex. In fact, the maximum CD amplitude driven by the D-isomer (ca. 12) is considerably smaller than that by the L-isomer (ca. 32), as shown in Figure 9. These theoretical simulations demonstrate that the helicity control induced by the NCDE should originate from external chirality-driven variation in energy gap between diasteromeric complexes. Concluding Remarks We here have attempted to amplify the NCDE through N-terminal chemical modification of a 310-helix. Consequently, the replacement of Aib residue by β-Ala residue was shown to successfully promote the helix-controlling factor of external chiral signal. The NCDE thus can be widely applied to helicity control of chiral peptides that prefer a one-handed helix even if its screw-sense bias is considerably strong. Herein the diastereomeric paring of the target helix with a small chiral guest, in response to its chiral sign and power, produces a new energy gap between left-handed and right-handed helices, thereby leading to either remarkable promotion or reduction (in more effective cases, inversion)

Experimental Section Materials. All amino acids and coupling reagents were commercially available, while β-phenylserine for our present and previous works has been prepared basically by a standard method.26a The Boc-amino acid as guest molecule was prepared with (Boc)2O or was commercially available. Thinlayer chromatography (TLC) was performed on precoated silica plates in the following solvent systems: (A) ethyl acetate, (B) methanol, (C) chloroform-methanol (9:1), and (D) 1-butanol-acetic acid-water (7:2:1).26b Synthesis. N-Boc-protected peptides 1 and 2 were synthesized by ring-opening of Boc-β-Ala-∆ZPhe azlactone5 with H-Aib-∆ZPhe-X*-(∆ZPhe-Aib)2-OMe,4a basically according to refs 4 and 5. Based on a manner similar to refs 3a and 5, their Boc-deprotection with formic acid and subsequent neutralization yielded peptides 1 and 2, which were recovered from precipitation from chloroform/hexane. The characterization data for the final peptides and the corresponding Boc-protected peptides are follows. A somewhat detailed description for NMR integral is given here. For FT-IR data, only major bands relevant to peptide bonds are provided. Boc-β-Ala-∆ Z Phe-Aib-∆ Z Phe- L -Leu-(∆ Z Phe-Aib) 2 OMe. RfA 0.42; RfB 0.89; RfC 0.81; RfD 0.62. MS (MALDITOF) (m/z) [M + Na]+ (calcd for C63H77N9O12Na as its monoisotopic mass, 1174.56): found, 1175.48. 600 MHz 1H NMR [δ, in CDCl3 containing 5 vol % (CD3)2SO at 293 K]: 9.50 + 9.46 [2H, s + s, NH ∆ZPhe(4) + NH ∆ZPhe(2)], 9.08 [1H, s, NH ∆ZPhe(6)], 8.64 [1H, s, NH ∆ZPhe(8)], 8.21 + 8.13 [2H, s + s, NH Leu(5) + NH Aib(3)], 7.80 [2H, s, NH Aib(7) and NH Aib(9)], 7.55-7.13 + 6.91 [ca. 24H (after subtraction of CHCl3 signal), m + s, 4×(phenyl H and CβH) ∆ZPhe], 5.30 [1H, s, NH β-Ala(1)], 4.19 [1H, m, CRH Leu(5)], 3.69 (3H, s, COOCH3), 3.31 + 3.06 [2H, s + s, CβH2 β-Ala(1)], ca. 2.55-2.37 [ca. 2H (after removal of H2O signal), CRH2 β-Ala(1)], 2.0-1.25 [ca. 30H, m, CβH2 Leu(5), CγH Leu(5), 6 × CH3 Aib, and 3 × CH3 Boc], and 0.91 + 0.82 [3H + 3H, d + d, 2 × CδH3 Leu(5)]. FT-IR (cm-1, in KBr): 3283, 1660, 1626, 1531. Boc-β-Ala-∆ Z Phe-Aib-∆ Z Phe- L -Phe-(∆ Z Phe-Aib) 2 OMe. RfA 0.43; RfB 0.89; RfC 0.79; RfD 0.62. MS (MALDITOF) (m/z) [M + Na]+ (calcd for C66H75N9O12Na as its

1238

Biomacromolecules, Vol. 5, No. 4, 2004

monoisotopic mass, 1208.54): found, 1209.46. 600 MHz 1H NMR [δ, in CDCl3 containing 5 vol % (CD3)2SO at 293 K]: 9.59 [1H, s, NH ∆ZPhe(4)], 9.42 [1H, s, NH ∆ZPhe(2)], 9.23 [1H, s, NH ∆ZPhe(6)], 8.64 [1H, s, NH ∆ZPhe(8)], 8.22 + 8.15 [2H, s + s, NH Phe(5) + NH Aib(3)], 7.81 + 7.78 [2H, s + s, NH Aib(9) + NH Aib(7)], 7.57-6.86 [ca. 29H (after subtraction of CHCl3 signal), m, 4 × (phenyl H + CβH) ∆ZPhe and phenyl H Phe(5)], 5.29 [1H, s, NH β-Ala(1)], 4.39 [1H, s, CRH Phe(5)], 3.69 (3H, s, COOCH3), 3.273.13 + 3.03 [3H + 1H, m + s, CβH2 Phe(5) and CβH β-Ala(1) + CβH β-Ala(1)], ca. 2.46-2.32 [ca. 2H (after removal of H2O signal), CRH2 β-Ala(1)], 1.62 + 1.61 + 1.57 + 1.57 + 1.37 + 1.34 (18H, 6 × s, 6 × CH3 Aib), and 1.41 (9H, s, 3 × CH3 Boc). FT-IR (cm-1, in KBr): 3286, 1661, 1626, 1530. H-β-Ala-∆ Z Phe-Aib-∆ Z Phe- L -Leu-(∆ Z Phe-Aib) 2 OMe (1). RfA 0; RfB 0-0.24; RfC 0-0.21; RfD 0.35. MS (MALDI-TOF) (m/z) [M + Na]+ (calcd for C58H69N9O10Na as its monoisotopic mass, 1074.51): found, 1075.22. 600 MHz 1H NMR (δ, in CDCl3): 9.35 [1H, s, NH ∆ZPhe(4)], 9.11 [1H, s, NH ∆ZPhe(6)], 8.79 [1H, s, NH ∆ZPhe(8)], 8.43 [1H, s, NH Aib(3)], 8.18 [1H, s, NH Leu(5)], 7.85 [1H, s, NH Aib(9)], 7.79 [1H, s, NH Aib(7)], 7.53-7.16 + 6.87 [ca. 24H (after subtraction of CHCl3 signal), m + s, 4× (phenyl H and CβH) ∆ZPhe], 4.14 [1H, s, CRH Leu(5)], 3.61 (3H, s, COOCH3), 2.96-2.75 [2H, m, CβH2 β-Ala(1)], 2.462.29 [2H, m, CRH2 β-Ala(1)], 1.96-1.17 [ca. 21H (after removal of H2O signal), m, CβH2 Leu(5), CγH Leu(5), and 6 × CH3 Aib], and 0.87 + 0.78 [3H + 3H, d + d, 2 × CδH3 Leu(5)]. FT-IR (cm-1, in chloroform): 3297, 1658, 1625, 1534; (cm-1, in KBr): (3358), 3321, 3271, 1661, 1628, 1530. H-β-Ala-∆ZPhe-Aib-∆ZPhe-L-Phe-(∆ZPhe-Aib)2-OMe (2). RfA 0; RfB 0-0.25; RfC 0-0.27; RfD 0.36. MS (MALDITOF) (m/z) [M + Na]+ (calcd for C61H67N9O10Na as its monoisotopic mass, 1108.49): found, 1109.22. 600 MHz 1H NMR (δ, in CDCl3): 9.35 [1H, s, NH ∆ZPhe(4)], 9.20 [1H, s, NH ∆ZPhe(6)], 8.74 [1H, s, NH ∆ZPhe(8)], 8.37 [1H, s, NH Aib(3)], 8.19 [1H, s, NH Phe(5)], 7.85 [1H, s, NH Aib(9)], 7.77 [1H, s, NH Aib(7)], 7.55-6.86 [ca. 29H (after subtraction of CHCl3 signal), m, 4 × (phenyl H + CβH) ∆ZPhe and phenyl H Phe(5)], 4.33 [1H, s, CRH Phe(5)], 3.61 (3H, s, COOCH3), 3.15 [2H, d, CβH2 Phe(5)], 2.95-2.76 [2H, m, CβH2 β-Ala(1)], 2.40-2.27 [2H, m, CRH2 β-Ala(1)], 1.54 + 1.53 + 1.52 + 1.50 + 1.27 + 1.27 [ca. 18H (after removal of H2O signal), 6 × s, 6 × CH3 Aib]. FT-IR (cm-1, in chloroform): 3292, 1658, 1626, 1533; (cm-1, in KBr): 3289, 1660, 1626, 1531. Spectroscopic Measurement. Although CD and UV analyses were basically conducted in a manner similar to a series of our previous studies, the detailed description has not been provided yet. CD and UV data were simultaneously obtained at ambient room temperature (typically, around 1821 °C) by using a JASCO J-600 spectrometer. In the present analysis, the following parameters for J-600 were usually employed: step interval for data acquisition, 0.2 nm; accumulation number, 2-3; scanning rate, 50 nm/min; time constant, 1 s. UV absorption spectra were also recorded by using a JASCO V-550 spectrometer usually with a step of 0.2 nm. The other conditions were as follows: [peptide (1

Inai and Komori

or 2)] ) 0.13-0.14 mM and [Boc-amino acid] ) 0-65 mM in chloroform (with a quartz cell of 1 mm optical path length). Chloroform was distilled over CaSO4 before use. Typically, the following procedure for sample preparation was chosen to minimize peptide-concentration change due to the high volatilization of chloroform in the handling process. First, a chloroform solution of peptide 1 or 2 was prepared with a prescribed concentration, which was determined on the basis of max (maximal value per ∆ZPhe residue in an absorption band around 280 nm) ) 1.8 × 104.4,5 Subsequently, a volume of the solution was added to an amount of Boc-amino acid, which was dissolved to give a desirable guest concentration. (A guest-peptide solution was diluted only with the original peptide solution for preparation of more dilute guest concentration.) Here, total volume change accompanied with the mixing of the guest molecule was neglected on the basis of the fact that there was no significant volume expansion of chloroform with and without Boc-Pro-OH under our conditions. CD and UV spectral data acquired from the J-600 spectrometer were usually smoothed by an appropriate mathematical method for reduction of statistical noise. ∆ and  expressed with respect to ∆ZPhe concentration were used for ordinates of CD and UV absorption spectra, respectively. Other spectroscopic measurements were performed in a manner similar to refs 4 and 5. Namely, 1H NMR data were acquired by using a Bruker DPX-200 (200 MHz) spectrometer at ambient room temperature (or ca. 26 °C). For more precise analysis, a Bruker DRX-600 (600 MHz) spectrometer was employed with typical conditions as follows: [peptide (1 or 2)] ) 2.4-2.6 mM in CDCl3 or CDCl3/(CD3)2SO at 293 K. CDCl3 and (CD3)2SO commercially available were used without purification. NOESY spectra were measured on the Bruker DRX-600 by using a Bruker standard pulse program (noesytp),27 usually with a mixing time of 200 or 400 ms, 2048 data points (TD) in F2 with 4 or 8 scans (NS) and 256 points (TD) in F1. (These NMR parameters are included in the Bruker-XWINNMR software.) FT-IR absorption spectra were obtained from a JASCO FT/IR-430 spectrometer in KBr and in chloroform. MALDI-TOF mass spectra of final nonapeptides were acquired on a VoyagerDE RP spectrometer [PerSeptive Biosystems; Applied Biosystems (at present)] in reflector mode by using a 1,8dihydroxy-9(10H)-anthracenone matrix and NaI salt for preparation of the sample. In our conditions, Boc-protected peptides 1 and 2 showed some prominent fragmentation peaks, possibly relevant to the Boc moiety, together with the expected parent peak. A similar tendency had also been observed in our previous studies, which had not mentioned that yet. Conformational Energy Calculations. Right-handed helices of Aib-terminal peptide 3 and 4 were already energyminimized4a by using semiempirical MO calculations (AM1 method in MOPAC97).14 Initial conformations for the AM1 calculation were obtained from molecular mechanics with the modified PEPCON.28 The AM1-based geometry-optimized right-handed helices were used for the corresponding initial conformations of the β-Ala-terminal peptides (1 and 2) after each N-terminal

Biomacromolecules, Vol. 5, No. 4, 2004 1239

Helicity Control of a Chiral Peptide by NCDE

H-Aib residue was replaced by H-β-Ala residue with a trans-trans conformation (θ ) 180° and ψ ) 180°).23 The minimization was performed for the variables of all bond lengths, bond angles, and torsion angles, with a MOPAC97 keyword of MMOK14b for correction of the rotational barrier about amide bonds.14c,d For each structural optimization, the EF-keyword14b-e was used with “LET DDMIN ) 0.0”,14b-d [Although we noticed, in the proof stage, that the “LET DDMIN ) 0.0” had not been used for the right-handed helix of peptide 2 alone (Figure 4, right), its original energy value was essentially unaffected by the use of this keyword.] Additional comments regarding MOPAC keywords14b-d for our studies will be reported elsewhere. In addition, a lefthanded 310-helix of peptide 2 was obtained through a similar procedure mentioned above.4a,14,28,29 Theoretical complex structures of peptide 2-Boc-Pro-OH were also obtained from the AM1-based MO calculation.14 The detailed calculation procedure was described in the preceding section and ref 5. The molecular modeling and molecular graphics were made on software mentioned in ref 5.

(8)

(9)

(10) (11) (12)

(13) (14)

Acknowledgment. This work was partly supported by the Ministry of Education, Culture, Sports, Science, and Technology of Japan under a grant (No.12650883) to Y.I. The authors sincerely thank Professor M. Kawai, Nagoya Institute of Technology for the use of the CD apparatus. References and Notes (1) Nakano, T.; Okamoto, Y. Chem. ReV. 2001, 101, 4013-4038. (2) (a) Yashima, E.; Matsushima, T.; Okamoto, Y. J. Am. Chem. Soc. 1997, 119, 6345-6359. (b) Schlitzer, D. S.; Novak, B. M. J. Am. Chem. Soc. 1998, 120, 2196-2197. (c) Majidi, M. R.; Kane-Maguire, L. A. P.; Wallace, G. G. Polymer 1994, 35, 3113-3115. (d) Yashima, E.; Maeda, K.; Okamoto, Y. Nature 1999, 399, 449-451. (e) Khatri, C. A.; Pavlova, Y.; Green, M. M.; Morawetz, H. J. Am. Chem. Soc. 1997, 119, 6991-6995. (f) Prince, R. B.; Barnes, S. A.; Moore, J. S. J. Am. Chem. Soc. 2000, 122, 2758-2762. (g) Li, J.; Schuster, G. B.; Cheon, K.-S.; Green, M. M.; Selinger, J. V. J. Am. Chem. Soc. 2000, 122, 2603-2612. (h) Kozlov, I. A.; Orgel, L. E.; Nielsen, P. E. Angew. Chem., Int. Ed. 2000, 39, 4292-4295. The original helix sense of a chiral polymer was shown to be reversed by addition of small chiral molecules. For the details, see: (i) Yashima, E.; Maeda, Y.; Okamoto, Y. J. Am. Chem. Soc. 1998, 120, 8895-8896. (j) Morino, K.; Maeda, K.; Yashima, E. Macromolecules 2003, 36, 1480-1486. (3) (a) Inai, Y.; Tagawa, K.; Takasu, A.; Hirabayashi, T.; Oshikawa, T.; Yamashita, M. J. Am. Chem. Soc. 2000, 122, 11731-11732. (b) Inai, Y. Recent Research DeVelopments in Macromolecules; Research Signpost: India, 2002; Chapter 2. (c) Inai, Y.; Hirano, T. ITE Lett. Batt. New Technol. Med. 2003, 4, 485-488. (4) (a) Inai, Y.; Komori, H.; Takasu, A.; Hirabayashi, T. Biomacromolecules 2003, 4, 122-128. (b) Inai, Y.; Ishida, Y.; Tagawa, K.; Takasu, A.; Hirabayashi, T. J. Am. Chem. Soc. 2002, 124, 2466-2473. (5) Inai, Y.; Ousaka, N.; Okabe, T. J. Am. Chem. Soc. 2003, 125, 81518162. (6) (a) Benedetti, E.; Bavoso, A.; Di Blasio, B.; Pavone, V.; Pedone, C.; Crisma, M.; Bonora, G. M.; Toniolo, C. J. Am. Chem. Soc. 1982, 104, 2437-2444. (b) Karle, I. L.; Balaram, P. Biochemistry 1990, 29, 6747-6756. (c) Toniolo, C.; Bonora, G. M.; Bavoso, A.; Benedetti, E.; Di Blasio, B.; Pavone, V.; Pedone, C. Macromolecules 1986, 19, 472-479. (e) Venkatraman, J.; Shankaramma, S. C.; Balaram, P. Chem. ReV. 2001, 101, 3131-3152. (f) Pengo, B.; Formaggio, F.; Crisma, M.; Toniolo, C.; Bonora, G. M.; Broxterman, Q. B.; Kamphuis, J.; Saviano, M.; Iacovino, R.; Rossi, F.; Benedetti, E. J. Chem. Soc., Perkin Trans. 2 1998, 1651-1657. (g) Vijayalakshmi, S.; Rao, R. B.; Karle, I. L.; Balaram, P. Biopolymers 2000, 53, 84-98. (7) (a) Pieroni, O.; Fissi, A.; Pratesi, C.; Temussi, P. A.; Ciardelli, F. J. Am. Chem. Soc. 1991, 113, 6338-6340. (b) Ciajolo, M. R.; Tuzi, A.; Pratesi, C. R.; Fissi, A.; Pieroni, O. Biopolymers 1990, 30, 911920. (c) Rajashankar, K. R.; Ramakumar, S.; Chauhan, V. S. J. Am.

(15)

(16) (17)

(18)

(19) (20)

Chem. Soc. 1992, 114, 9225-9226. (d) Jain, R.; Chauhan, V. S. Biopolymers 1996, 40, 105-119. (e) Ramagopal, U. A.; Ramakumar, S.; Sahal, D.; Chauhan, V. S. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 870-874. (a) Inai, Y.; Sakakura, Y.; Hirabayashi, T. Polym. J. 1998, 30, 828832. (b) Kennedy, D. F.; Crisma, M.; Toniolo, C.; Chapman, D. Biochemistry 1991, 30, 6541-6548. For IR studies of other ∆ZPhecontaining peptides, see: (c) Gupta, A.; Mehrotra, R.; Tewari, J.; Jain, R. M.; Chauhan, V. S. Biopolymers 1999, 50, 595-601. (a) Wu¨thrich, K.; Billeter, M.; Braun, W. J. Mol. Biol. 1984, 180, 715-740. (b) Billeter, M.; Braun, W.; Wu¨thrich, K. J. Mol. Biol. 1982, 155, 321-346. For a related comprehensive review, see: (c) Wu¨thrich, K. NMR of Proteins and Nucleic Acids; John Wiley and Sons: New York, 1986. Pitner, T. P.; Urry, D. W. J. Am. Chem. Soc. 1972, 94, 1399-1400. Toniolo, C.; Benedetti, E. Trends Biochem. Sci. 1991, 16, 350-353 and references therein. (a) Harada, N.; Chen, S. L.; Nakanishi, K. J. Am. Chem. Soc. 1975, 97, 5345-5352. (b) Harada, N.; Nakanishi, K. Circular Dichroic Spectroscopy. Exciton Coupling in Organic Stereochemistry; University Science Books: Mill Valley, CA, 1983. For CD studies on ∆ZPhe-containing peptides, see: (c) Pieroni, O.; Fissi, A.; Jain, R. M.; Chauhan, V. S. Biopolymers 1996, 38, 97-108. Aib-based pentapeptides containing an internal L-residue were shown to adopt a right-handed 310-helix.6f (a) The AM1 method in MOPAC97 was used: (a) Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J. P. J. Am. Chem. Soc. 1985, 107, 3902-3909. For MOPAC97, see: (b) Stewart, J. J. P. MOPAC97; Fujitsu Ltd.: Tokyo, Japan, 1998. For MOPAC keywords, see also: (c) Stewart, J. J. P. MOPAC 2000 Manual; Fujitsu Ltd.: Tokyo, Japan, 1999. (d) Stewart, J. J. P. MOPAC 93 Manual, revision number 2; Fujitsu Ltd.: Tokyo, Japan, 1993. For the EF keyword-based routine, see also: (e) Baker, J. J. Comput. Chem. 1986, 7, 385-395. The values were calculated from CR positions of segment -∆ZPheAib-∆ZPhe-X*-∆ZPhe-Aib-∆ZPhe-. For a general mathematical expression for obtaining helix parameters, see: Sugeta, H.; Miyazawa, T. Biopolymers 1967, 5, 673-679. Inai, Y.; Oshikawa, T.; Yamashita, M.; Tagawa, K.; Hirabayashi, T. Biopolymers 2003, 70, 310-322. (a) Kurta´n, T.; Nesnas, N.; Li, Y.-Q.; Huang, X.; Nakanishi, K.; Berova, N. J. Am. Chem. Soc. 2001, 123, 5962-5973. For the pioneering concepts of “host-guest” and “supramolecular” complexations, see: (b) Cram, D. J. Science 1988, 240, 760-767. (c) Lehn, J.-M. Science 2002, 295, 2400-2403. The initial titration data (0-10 mM for 1 and 2; 0-65 mM for 3 and 4) were used for the nonlinear fitting, in which the root-meansquare deviation between experimental and calculated data was minimized by the Simplex method. For the Simplex algorithm, see: Nelder, J. A.; Mead, R. Comput. J. 1965, 7, 308-313. Okamoto, Y.; Yashima, E. Angew. Chem., Int. Ed. Engl. 1998, 37, 1020-1043. This treatment should be valid based on the aforementioned experimental evidence for a 310-helix of peptide 2. Basically, we also neglected various species where left-/right-handed helical segments appear with any composition along a single chain, because the chain length for maintaining the same helicity in (Aib-∆ZPhe)-based peptides is expected to exceed our current length (9 residues) from the following facts. (i) Achiral Boc-(Aib-∆ZPhe)4-Aib-OMe adopts a 310-helix in the crystal, in which each peptide chain conserves either helicity without any helix inversion at internal positions.16 (ii) A decapeptide bearing a C-terminal L-L doublet, Boc-(Aib-∆ZPhe)4L-Leu2-OMe4b (7), folds into a right-handed 310-helix in solution. The C-terminal chirality transfers covalently to the N-amino head along the achiral sequence, thereby inducing right-handedness for the whole chain. Chirality-periodically incorporated sequence, Boc(L-Leu-∆ZPhe)4-L-Leu-OMe21a (8), also forms a right-handed 310helix. Intriguingly, both split CD amplitudes are essentially the same: ca. 50 for 7;4b ca. 52 for 8.21a Apparently, the C-terminal chiral signal of peptide 7 should move successfully to the N-terminus along the preceding achiral segment. Thus, (Aib-∆ZPhe)-based backbones have a potency to maintain only a one-handed helix through a chiral terminus, if its chiral information succeeds to first achiral segment in most effective manners. (iii) An insightful variable-temperature NMR study demonstrated that an Aib-decapeptide in solution adopts either a left-handed or right-handed 310-helix dynamically.21b Similar alternative helicity was also suggested from low-temperature NMR spectra of peptide 6 with Boc-L-amino acid.5

1240

Biomacromolecules, Vol. 5, No. 4, 2004

(21) (a) Inai, Y.; Hasegawa, K.; Hirabayashi, T.; Yokota, K. Polym. J. 1996, 28, 238-245. (b) Hummel, R.-P.; Toniolo, C.; Jung, G. Angew. Chem., Int. Ed. Engl. 1987, 26, 1150-1152. (22) According to ref 5, the θ value of β-Ala residue was set to +66° for a right-handed helix, and to -66° for a left-handed one. (23) For definition of torsional angles of β-amino acid residues, see: (a) Cheng, R. P.; Gellman, S. H.; DeGrado, W. F. Chem. ReV. 2001, 101, 3219-3232. See also: (b) Karle, I. L.; Pramanik, A.; Banerjee, A.; Bhattacharjya, S.; Balaram, P. J. Am. Chem. Soc. 1997, 119, 9087-9095. For the usual definition, see: (c) IUPAC-IUB Commission on Biochemical Nomenclature, Biochemistry 1970, 9, 34713479. (24) This value was obtained from the AM1 calculation14 of H2+-β-AlaNH-CH3, while such value was set to ca. 1.000 Å in the previous study.5 (25) (a) Kuroda, Y.; Maeda, Y.; Nakagawa, T. J. Am. Chem. Soc. 2000, 122, 12596-12597. (b) Maeda, Y.; Nakagawa, T.; Kuroda, Y. J. Pept. Sci. 2003, 9, 106-113. (26) For example, see: (a) Carrara, G.; Weitnauer, G. Gazz. Chim. Ital. 1949, 79, 856-862; SciFinder Scholar, American Chemical Society, CAN 44:38039. For a general guideline for peptide synthesis, see: (b) Izumiya, N.; Kato, T.; Aoyagi, H.; Waki, M. Principle and Practice of Peptide Synthesis. (Pepuchido Gosei no Kiso to Jikken); Maruzen Co., Ltd.: Tokyo, Japan, 1985; SciFinder Scholar, American Chemical Society, CAN 103:37731. (27) Bodenhausen, G.; Kogler, H.; Ernst, R. R. J. Magn. Reson. 1984, 58, 370-388. (28) In ref 4a, PEPCON-based energy-minimization of peptides 3 and 4 was performed with varying torsional angles of main chains (φ, ψ) and side chains (χ1,1Aib, χ1,2Aib, χ2∆Phe; χ1Leu, χ2Leu, χ3,1Leu, χ3,2Leu for 3; χ1Phe, χ2Phe for 4) through the Simplex method18 that is incorporated into the original PEPCON. For the original PEPCON based on the ECEPP28a parameters, see: (a) Momany, F. A.; McGuire, R. F.; Burgess, A. W.; Scheraga, H. A. J. Phys. Chem. 1975, 79, 23612381. (b) Beppu, Y. Comput. Chem. 1989, 13, 101. (c) Sisido, M. Pept. Chem. 1992, 1991, 105-110. Aib residue with its ECEPP parameters28d is incorporated into the original version28e: (d) Paterson,

Inai and Komori Y.; Rumsey, S. M.; Benedetti, E.; Ne´methy, G.; Scheraga, H. A. J. Am. Chem. Soc. 1981, 103, 2947-2955. (e) Sisido, M.; Ishikawa, Y.; Harada, M.; Itoh, K. Macromolecules 1991, 24, 3999-4003. For stable conformers of natural residues, see: (f) Zimmerman, S. S.; Pottle, M. S.; Ne´methy, G.; Scheraga, H. A. Macromolecules 1977, 10, 1-9. For ECEPP-based energy minimization from local minima, see also: (g) Ishikawa, Y.; Hirano, Y.; Yoshimoto, J.; Oka, M.; Hayashi, T. Polym. J. 1998, 30, 256-261. For examples of the modified version for β-aryldehydroresidues, see: (h) Inai, Y.; Ito, T.; Hirabayashi, T.; Yokota, K. Biopolymers 1993, 33, 1173-1184. (i) Inai, Y.; Kurashima, S.; Hirabayashi, T.; Yokota, K. Biopolymers 2000, 53, 484-496. (j) Inai, Y.; Oshikawa, T.; Yamashita, M.; Hirabayashi, T.; Hirako, T. Biopolymers 2001, 58, 9-19. (k) Inai, Y.; Hirabayashi, T. Biopolymers 2001, 59, 356-369. (l) Inai, Y.; Oshikawa, T.; Yamashita, M.; Hirabayashi, T.; Kurokawa, Y. Bull. Chem. Soc. Jpn. 2001, 74, 959-966. In our related papers, 10 (φ),28m,n 8 (ψ),28o and 6.4 (χ2)28p have been used for torsional barriers (kcal mol-1) about dihedral angles28a in β-aryldehydroresidues. These parameters are based on the following refs: (m) Ajo`, D.; Granozzi, G.; Tondello, E. J. Mol. Struct. 1977, 41, 131-137. (n) Uma, K.; Balaram, P. Indian J. Chem. 1989, 28B, 705-710. (o) Ajo`, D.; Granozzi, G.; Tondello, E.; Del Pra`, A.; Zanotti, G. J. Chem. Soc. Perkin 2 1979, 927-929. (p) Ajo`, D.; Casarin, M.; Granozzi, G. J. Mol. Struct. (THEOCHEM) 1982, 86, 297-300. For CdC barrier relevant to χ1, see also: (q) Burkert, U.; Allinger, N. L. Molecular Mechanics; American Chemical Society: Washington, DC, 1982. (29) The first process for obtaining the initial conformation of peptide 2 was based on PEPCON and subsequent MO calculation of peptide 4. For initial conformations of peptide 4 for the PEPCON, stable conformers of Phe residue28f were combined with achiral residues in a left-handed 310-helix (60°, 30°), of which (φ, ψ) belongs to the enantiomer of a standard right-handed 310-helix,28d or to a type III′ β-turn assignable to initiation of a left-handed 310-helix:29a (a) Venkatachalam, C. M. Biopolymers 1968, 6, 1425-1436.

BM0344001