Possible Implications of Serine and Tyrosine Residues and

Nov 19, 2004 - Twelve selected peptides (1−12) incorporating Ser and Tyr residues in .... by 13C cross-polarization/magic-angle spinning (CP/MAS) NM...
0 downloads 0 Views 208KB Size
Download PDF
Biomacromolecules 2005, 6, 468-474

468

Possible Implications of Serine and Tyrosine Residues and Intermolecular Interactions on the Appearance of Silk I Structure of Bombyx mori Silk Fibroin-Derived Synthetic Peptides: High-Resolution 13C Cross-Polarization/Magic-Angle Spinning NMR Study Tetsuo Asakura,*,† Kosuke Ohgo,† Teppei Ishida,† Paola Taddei,‡ Patrizia Monti,‡ and Raghuvansh Kishore§ Department of Biotechnology, Tokyo University of Agriculture and Technology, Koganei, Tokyo 184-8588, Japan, Dipartimento di Biochimica “G. Moruzzi”, Sezione di Chimica e Propedeutica Biochimica, Centro di Studi sulla Spettroscopia Raman, Universita` di Bologna, Via Belmeloro 8/2, 40126 Bologna, Italy, and Institute of Microbial Technology, Sector 39-A, Chandigarh 160 036, India Received August 28, 2004; Revised Manuscript Received September 30, 2004

Bombyx mori silk fibroin molecule is known to exist in two distinct structural forms: silk I (unprocessed silk fibroin) and silk II (processed silk fibroin). Using synthetic peptides, we attempt to explore the structural role played by Ser and Tyr residues on the appearance of silk I structural form of the fibroin. Twelve selected peptides (1-12) incorporating Ser and Tyr residues in the (Ala-Gly)n copolypeptide, that is, the sequences mimicking the primary structure of B. mori silk fibroin molecule, have been investigated under the silk I state, employing high-resolution 13C cross-polarization/magic-angle spinning (CP/MAS) NMR spectroscopy. To acquire the silk I structural form, all the peptides were dissolved in 9 M LiBr and then dialyzed extensively against water, as established previously for the synthetic (Ala-Gly)15 copolypeptide and B. mori silk fibroin. The diagnostic line shape of the Ala Cβ peaks and the conformation-dependent 13C chemical shifts of Ala and Gly resonances are presented to analyze and characterize the structural features. The results indicate that the incorporation of one Ser and/or one Tyr residue(s) at selected position in the basic (Ala-Gly)15 sequence tend to retain predominantly the silk I structure. Conversely, the repeat pentameric and octameric Ala-Gly-Ser-Gly-Ala-Gly sequences, for example, (Ala-Gly-Ser-Gly-Ala-Gly)5 or (Ala-GlySer-Gly-Ala-Gly)8, preferred predominantly the silk II form. The peptide sequences incorporating Ser and Tyr residue(s) into repeat Ala-Gly-Ser-Gly-Ala-Gly sequences, however, adopted the silk II structure with certain content structural heterogeneity or randomness, more pronounced for specific peptides studied. Interestingly, the crystalline Cp fraction of B. mori silk fibroin, when mixed with (Ala-Gly-Ser-Gly-AlaGly)5 sequence in a 5:1 molar ratio, dissolved in 9 M LiBr, and dialyzed against distilled water, favor the silk I form. The finding tends to suggest that the less stable silk I form in (Ala-Gly-Ser-Gly-Ala-Gly)n sequences is likely to be induced and facilitated via intermolecular interactions with the Cp fraction, which predominantly prefers the silk I form under similar conditions; however, the hydrogen-bond formation involving OγH groups of the Ser residues may have some implications. Introduction The structural analysis of silk fibroin molecules receives considerable attention since they possess excellent mechanical properties such as strong and tough character and offer numerous biotechnological applications.1 Bombyx mori, a well-known silkworm that produces silk fibroin fibers at room temperature, has extensively been studied over decades. The silk fibroin molecule showed the predominance of five amino acids: ∼42.9% Gly, 30% Ala, 12.2% Ser, 4.8% Tyr, and 2.5% Val.2,3 Besides the fundamental (Gly-Ala)n se* To whom correspondence should be addressed: phone/fax (+81)-42383-7733; e-mail [email protected]. † Tokyo University of Agriculture and Technology. ‡ Universita ` di Bologna. § Institute of Microbial Technology.

quence motif, the existence of unusual repeat sequences was independently noticed first by Mita et al.2 and later by Zhou et al.,3 who employed a shotgun strategy of cDNA sequencing of the fibroin gene coding for the heavy chain. An analysis of the primary structure of B. mori silk fibroin indicated that the sequence may be roughly divided into four modular motifs: module i contains a highly repetitive GAGAGS sequence (single-letter code of amino acids is used) and comprises the crystalline regions; module ii contains relatively less repetitive sequences with hydrophobic and/or aromatic residues, GAGAGY and GAGAGVGY, and makes up the semicrystalline regions; module iii is very similar to module i except for the presence of an AAS motif; and module iv constitutes the amorphous regions containing negatively charged, polar, bulky hydrophobic, and/or aro-

10.1021/bm049487k CCC: $30.25 © 2005 American Chemical Society Published on Web 11/19/2004

13C

Solid NMR Study of Silk Model Peptides

Biomacromolecules, Vol. 6, No. 1, 2005 469

Figure 1. Schematic representation of fine organization of the primary structure of B. mori silk fibroin heavy chain. R01‚‚‚R12 and A01‚‚‚A11 represent the unique arrangements of 12 repetitive and 11 amorphous regions, respectively. An approximate amino acid sequence of the R10 region is illustrated by i, ii, and iii. Module iv depicts a typical amino acid sequence in the amorphous region.

matic residues, for example, TGSSGFGPYVANGGYSGYEYAWSSESDFGT.3 The fine organization of the primary structure of B. mori silk fibroin, indicating repetitive and amorphous regions, is schematically presented in Figure 1. Many attractive properties of the silk material, expected to be associated with this natural fiber, presumably originate from the unique organization of such repetitive motifs in the fibroin molecule, which translate into higher-order structural orderliness under natiVe conditions. The two crystalline forms, silk I and silk II, are reported as the dimorphs of silk fibroin, based on extensive investigations by X-ray fiber diffraction,4-11 electron diffraction,8,10,11 conformational energy calculations,12,13 and infrared,14 Raman,15-17 and solid-state NMR spectroscopic techniques.18 Silk II is the structure of the fiber after spinning (after processing) and is essentially characterized as antiparallel β-sheet structure, stabilized by intersheet intermolecular hydrogen-bonding interactions,4 whereas silk I represents the conformation of the fibroin in the solid state before spinning (before processing). The latter sample is obtained as a film from the liquid silk stored in the gland of the silkworm or from an aqueous solution of regenerated fibroin. Despite a long history of interest in the less stable silk I structure, it largely remained poorly understood until recently. We recently attempted the characterization of the molecular structure of silk I form, using the synthetic peptide (Ala-Gly)15 sequence as a model for highly repetitive crystalline domain of the fibroin.19,20 The proposed model incorporated repeated type II β-turn structure, stabilized by a typical 4 f 1 intramolecular hydrogen bond. And the planar sheets are held together by intermolecular interactions involving the central amide bond of the β-turn, that is, the amide bond perpendicular to the intramolecular interaction. Widely accepted strategies have been employed to convert the silk fibroin and synthetic model peptides into either silk I or silk II structural forms. For example, if the silk fibroin or (Ala-Gly)15 peptide is dissolved in 9 M LiBr and dialyzed against distilled water, the resulting structure becomes silk I. On the other hand, if the silk I structural form is treated

with formic acid and air-dried, the structure would be silk II. The high-resolution 13C NMR characteristics, that is, the diagnostic chemical shifts and line shape analysis of the Gly CR, Ala CR, Ala Cβ, and Gly and Ala CdO resonances, of the two discrete structural forms have been exploited for detailed conformational interpretations.18,21 In view of the relative significance of various constituent amino acids of B. mori silk fibroin, besides Gly and Ala residues, the role of Ser and Tyr residues may be considered critical in contributing to many superior properties that are inherently associated with silk fibers.22,23 In our previous paper, we examined the influence of the Tyr residue, by altering its position along the peptide chain, on the structural properties of selected model peptides with pure (Ala-Gly)15 sequence.24 The solid-state 13C NMR spectra of the Tyr-containing peptides permitted detailed analysis of intermolecular arrangement of the polypeptide chains, particularly with respect to its local environment. The localized disruptive influence of the Tyr residues on the preferred molecular conformation and organization of the basic (Ala-Gly)15 sequence have been discussed. Moreover, the structural characteristics of different peptides of varying chain lengths, with Ser and Tyr residues contained in the Ala-Gly copolypeptides, in silk II structural form, for example, after formic acid treatment, have also been studied by 13C crosspolarization/magic-angle spinning (CP/MAS) NMR25 and Raman spectroscopies.17 As part of our systematic investigation, to further assess the role of Ser and Tyr residues on the appearance of silk I structure of B. mori silk fibroin, employing high-resolution 13 C CP/MAS NMR, we examined the structural properties of 12 synthetic peptides (1-12) containing Ser and/or Tyr residues incorporated in the basic (Ala-Gly)n sequence. Especially, (Gly-Ala-Gly-Ser-Gly-Ala)n, which is the main sequence of the Cp fraction, still takes the silk II form after being dissolved in 9 M LiBr and then dialyzed against water. Syntheses of several peptides where other sequences containing Ser and/or Tyr residues were added to the repeated AGSGAG sequences in the chain were tried. The mixing of

470

Biomacromolecules, Vol. 6, No. 1, 2005

Asakura et al.

the peptide (AGSGAG)5 and the Cp fraction was also tried to check whether silk I can be formed by intermolecular interactions between the fraction and the peptide. For monitoring the structure after mixing with the Cp fraction by 13C CP/MAS NMR, 13C isotope labeled peptide was used. Experimental Section All the twelve peptides (1-12) reported here were synthesized employing solid-phase Fmoc chemistry on a fully automated Pioneer peptide synthesis system (Applied Biosystems Ltd.). The procurement of all the relevant chemicals and reagents, details of the procedures followed for chemical synthesis, cleavage of the peptide from the poly(ethylene glycol)-polystyrene (PEG-PS) resin, and their isolation have been described elsewhere.19,24 The sequences of the 12 peptides reported in this paper are (AG)15 (1), (AG)7SG(AG)7 (2), (AG)7YG(AG)7 (3), (AG)12YG(AG)2 (4), (AG)7SG(AG)4YG(AG)2 (5), (AGSGAG)5 (6), (AGSGAG)8 (7), G(AG)2SGAA(SGAGAG)3SGAG (8), G(AG)2SGAA(SGAGAG)6YG (9), (AGSGAG)4AGYGAG (10), (AGSGAG)2AGYGAGYG(AG)3YGAG)3 (11), and AGSG(AG) 2 SGA[1- 1 3 C]G[3- 1 3 C]A[2- 1 3 C]GSG[2- 1 3 C]AG[1-13C]AGSG(AG)2SGAG (12). The single-letter code has been used to indicate the sequences. Of the 12 peptides investigated, the chemical synthesis of two peptides, 8 and 9, has been described previously.19,24,25 To confirm specific structural features, the 13C isotope-enriched peptide 12 was synthesized for the 13C CP/MAS NMR analysis. TheCp fraction of B. mori silk fibroin used was prepared from regenerated silk fibroin solution, as reported previously.14,18 The primary structure of the Cp fraction was originally reported by Strydom et al.26 as G(AG)2SGAAG[SGAGAG]8Y. The correct sequence analysis is discussed in a later section. To obtain the silk I structural form of these peptides, each peptide was dissolved in a minimal amount of 9 M LiBr solution and then dialyzed extensively against distilled water. The precipitate resulting was collected by filtration or lyophilized in the case of soluble peptides. It may be noted that this is a widely accepted method for the preparation of B. mori silk fibroin with silk I structure. The high-resolution 13C CP/MAS NMR measurements were conducted on a Chemagnetics CMX-400 spectrometer operating at 100 MHz for 13C nucleus. CP was employed for sensitivity enhancement with high-power 1H decoupling during the signal acquisition interval. A 1H 90° pulse width of duration 5 µs was used with a 1 ms contact time and a 3 s repetition time. Approximately 15K FIDs were added to generate the spectra. The chemical shifts were represented in parts per million (ppm) with respect to external reference adamantane and converted to tetramethylsilane (TMS) chemical shift reference. Results and Discussion 1. 13C CP/MAS NMR Spectral Analysis of Peptides 1-5. For the sake of comparison, the 13C CP/MAS NMR spectra of peptide 1 in its predominant silk I form, that is,

Figure 2. 13C CP/MAS NMR spectra of (AG)15 with silk I form (a); (AG)15 with silk II form (b); peptide 2, (AG)7SG(AG)7 (c); peptide 3, (AG)7YG(AG)7 (d); peptide 4, (AG)12YG(AG)2 (e); and peptide 5, (AG)7SG(AG)4YG(AG)2 (f). For silk I form, the peptides (AG)15 (a) and 2-5 (c-f) were first dissolved in 9 M LiBr and then dialyzed against water. For silk II form (b), the peptide (AG)15 was treated with the formic acid and air-dried.

after dissolution in 9 M LiBr followed by dialysis against water, and silk II form, that is, after formic acid treatment,27 are shown in Figure 2, panels a and b, respectively. Apparently, the chemical shifts and characteristic line shapes of the Gly CR, Ala CR, Gly CdO, Ala CdO, and Ala Cβ resonance are quite distinct in the two structural forms. The 13 C NMR spectrum of (AG)7SG(AG)7, peptide 2, wherein the Ser moiety occupies the central position of the (AG)15 sequence, in silk I form exhibits remarkably similar spectral characteristics (Figure 2c) to those observed for 1 under identical experimental conditions (Figure 2a). It is noteworthy that the appearance of peaks assignable to the single Ser residue is not so emergent. From the spectral features, it seems apparent that the Ser residue with short polar side

13C

Solid NMR Study of Silk Model Peptides

chains is easily accommodated in silk I structural form, without any significant disruption of the repeated type II β-turn structure preferred by the basic (AG)15 sequence. As reported previously, the influence of OγH functionality of the Ser residue may be localized within the β-turn structure via an intramolecular hydrogen bond with the CdO group of the Gly residue.28 On the contrary, the occurrence of a hydrophobic aromatic Tyr residue in the analogous central position of the (AG)15 sequence, that is, (AG)7YG(AG)7, peptide 3, tends to promote significantly the silk II structure, as indicated from the 13C NMR spectrum shown in Figure 2d.24 The inference receives corroboration from the fact that when the Tyr residue is located near the C-terminus of the (AG)15 copolypeptide, as in (AG)12YG(AG)2, peptide 4, remarkably distinct 13C NMR spectral characteristics were observed (Figure 2e). Interestingly, the conformation of peptide 4 is strongly driven toward the silk I structural form. Therefore, it may be presumed that in B. mori silk fibroin the bulky Tyr residue can disrupt the silk I structure, if the Tyr residue(s) is flanked or located near relatively short stretches of (AG)n sequences. On the other hand, a sufficient long segment of the (AG)n sequence in close proximity to the Tyr residue, may preserve the silk I structure. Apparently, in view of silk I and silk II structural forms of B. mori fibroin molecule, the influence of Tyr and Ser residues on the conformational preferences appears to be opposite: if the single Ser residue is located in the center of (AG)15 sequence or the Tyr is positioned in the proximity of the C-terminus of the (AG)15 copolypeptide, the observed 13 C NMR spectra reflected the features typical of silk I structural form. In view of this conjecture and to provide further experimental support, we synthesized (AG)7SG(AG)4YG(AG)2, peptide 5, which essentially retains the position of Ser and Tyr residues in the (AG)15 sequence. Expectedly, the observed 13C NMR spectrum of the peptide (Figure 2f), after dissolution in 9 M LiBr and dialysis against distilled water, was predominantly representative of silk I structural form but nevertheless indicates the existence of a small amount of β-sheet structure. 2. 13C CP/MAS NMR Spectral Analysis of the Cp Fraction and Peptides 6-11. The primary structure of the Cp fraction was originally reported by Strydom et al.26 as G(AG)2SGAAG[SGAGAG]8Y. Recently Zhou et al.3 presented the complete sequence and fine organization of B. mori fibroin heavy-chain gene. The result permits necessary correction of the primary structure of the Cp fraction as G(AG)2SGAA[SGAGAG]8Y, and it may be treated as an averaged sequence since the longer and shorter sequences with slightly different compositions are expected to coexist in the Cp fraction if the cleavage by chymotrypsin occurs at the C-terminal side of the Tyr residues. It should be noted that the main part of the Cp fraction is anyway the repeated SGAGAG sequences. Figure 3 panels a and b show the 13C CP/MAS NMR spectra of the Cp fraction of B. mori silk fibroin, after dissolution in 9 M LiBr and then dialysis against water (silk I form) and after formic acid treatment (silk II form) published previously.18 The distinct spectral characteristics represent the anticipated structural features of silk I and silk II forms.

Biomacromolecules, Vol. 6, No. 1, 2005 471

Figure 3. 13C CP/MAS NMR spectra of Cp fraction with silk I form (a); Cp fraction with silk II form (b); peptide 6, (AGSGAG)5 (c); peptide 7, (AGSGAG)8 (d); peptide 8, G(AG)2SGAA(SGAGAG)3SGAG (e); and peptide 9, G(AG)2SGAA(SGAGAG)6YG (f). All the peptides 6-9 were first dissolved in 9 M LiBr and then dialyzed against water.

Figure 3c shows the observed 13C CP/MAS NMR spectrum of 30-residue (AGSGAG)5, peptide 6, after it is dissolved in 9 M LiBr and then dialyzed against water. The spectral pattern clearly correlates with silk II structural forms, presented by either (Ala-Gly)15 sequence or the Cp fraction or the silk fibroin. The result was somewhat inconsistent. We presume that the Cp fraction and peptide 6 in fact incorporate relatively much shorter stretches of repeat AlaGly sequences, as compared to peptide 2, and regular appearance of the Ser residues in Ala-Gly copolymer may have a tendency to destabilize the silk I form. To substantiate the possible deduced conclusion, a longer 48-residue peptide (AGSGAG)8, peptide 7, was synthesized and its 13C NMR spectrum was compared under similar conditions. As shown in Figure 3d, the observed spectrum was slightly broader as compared to peptide 6, although the diagnostic chemical shifts of the main peaks are very similar to silk II form. We attribute this spectral broadening to a slight increase in the heterogeneity of the preferred structure(s). Okuyama and coworkers29 pointed out that relative population of silk I and silk II structures of the Cp fraction also depends on the pH of the water used for dialysis, after the Cp fraction was dissolved in 9 M LiBr solution. For example, pH lower than

472

Biomacromolecules, Vol. 6, No. 1, 2005

Asakura et al.

Figure 4. 13C CP/MAS NMR spectra of peptide 6, (AGSGAG)5 (a); peptide 10, (AGSGAG)4AGYGAG (b); and peptide 11, (AGSGAG)2AGYGAGYG(AG)3YGAG (c). The peptides 6, 10, and 11 were first dissolved in 9 M LiBr and then dialyzed against water.

5 is revealed to induce silk I structure. Nevertheless, despite a change of the pH of the dialysis medium, the structure of peptides 6 and 7 remained largely unaffected (data not shown). We noticed the existence of an uncommon AAS segment in the primary structure of the Cp fraction.3,30 Therefore, it may be of some interest to assess the structural consequences of such a segment, and for that reason, we investigated two synthetic peptides: G(AG)2SGAA(SGAGAG)3SGAG, peptide 8, and G(AG)2SGAA(SGAGAG)6YG, peptide 9, of varying chain lengths, mimicking the sequence of the Cp fraction. The observed 13C NMR spectra of peptides 8 and 9 (9 M LiBr treatment and dialysis against distilled water) are shown in Figure 3, panels e and f, respectively. The spectral features reflected the characteristics of silk II structural form. Thus, it seems apparent that the generation of silk I structure in synthetic peptides, under the conventional conditions, may not be straightforward and attributable explicitly to the AAS sequence. Moreover, the chain length alone does not appear to be the primary factor for the induction and appearance of the silk I form, and more complex mechanism(s) need to be visualized for elucidating and understanding the structural transformation between silk I and silk II structures. More refined analysis of the designed sequences, with respect to relative content and position of the Ser and/or Tyr residues in the basic (Ala-Gly)n copolymer, may be mandatory to search for a plausible explanation. To get further insight into the role played by Ser and/or Tyr residues, on the appearance of silk I form, structural analysis of two additional peptides, (AGSGAG)4AGYGAG, peptide 10, and (AGSGAG)2AGYGAGYG(AG)3YGAG,

peptide 11, were undertaken. It must be realized that the highly repetitive (AGSGAG)n sequence of module i and relatively less repetitive (AGYGAG)n sequence of module ii coexist in the repetitive region of B. mori silk fibroin, as shown in Figure 1. The 13C CP/MAS NMR spectra of peptides 10 and 11 (after 9 M LiBr and dialysis) are illustrated in Figure 4, panels b and c, respectively. To elucidate the structural features, the 13C NMR spectrum of (AGSGAG)5, under similar conditions, is also shown (Figure 4a). As already discussed, peptide 6 evidently takes the silk II structure. Although the spectral features of peptides 10 and 11 reflect the characteristics of silk II structure, the relative content of structural heterogeneity or randomness was more pronounced for peptide 11. The influence of one or more Tyr residues on the basic (AGSGAG)n sequences, to induce a disruptive effect on the preferred conformations is indeed apparent. Remarkably, a similar conclusion was drawn from experiments performed with formic acid treatment.25 As already recognized, after 9 M LiBr treatment and dialysis, peptide 6 takes silk II structure in the solid state, whereas the (AG)n copolymer and Cp fraction take the silk I structure under similar experimental conditions. We suspect that the nature of hydrogen bonding, involving the OγH group of the Ser side chain, might be critical to select silk I or silk II structures during the course of dialysis process against water. Peptide 2, with one Ser residue in the central position of (AG)n sequence, takes the silk I structure. As reported previously, the Ser side chain residue is capable of forming an intramolecular hydrogen bond involving the OγH group and the carbonyl group of the Gly residue, presumably stabilizing the silk I form.28 The incorporation of a single

13C

Solid NMR Study of Silk Model Peptides

Tyr residue, toward the C-terminus in peptide 2 (i.e., peptide 5), does not perturb the spectral pattern of silk I form. However, multiple existence of (AGSGAG) sequence appears to modulate the pattern of the hydrogen-bonding interactions involving OγH group of the Ser side chains and the backbone CdO oxygen. The 2H NMR measurements of uniaxially aligned [3,3-2H2]Ser-labeled B. mori silk fiber with silk II form indicated that the dominant conformer of the Ser side chain is gauche+, and this orientation could be a good candidate for facilitating intermolecular hydrogen bonds with the carbonyl groups on adjacent peptide chains, strongly favoring the β-sheet structure.31 The proposal received strong support from molecular mechanics calculations, which indicated that such backbone to side-chain intermolecular hydrogen bonding facilitates β-sheet conformation.25 Nevertheless, it is of considerable interest that the Cp fraction with mainly (AGSGAG)n sequence still takes the silk I structure after dialysis. At present, we are unable to offer a precise explanation. However, the Cp fractions, which are obtained by chymotryptic digestion of silk fibroin, may be much more heterogeneous than the synthetic ones. Heterogeneity may arise from differences in primary structure and size. Considering the primary structure of the silk fibroin heavy chain3 and the cleavage specificity of chymotrypsin, a bit more than 300 cleavage sites are available along the chain. If all of them are used, peptides with length ranging from a few residues to 100 residues are formed. Since the Cp fractions precipitate during hydrolysis, owing to their hydrophobic character, it is likely that not all cleavage sites are attacked by the enzyme before precipitation, resulting in longer peptides present in the precipitate fraction. In such longer peptide chains, amino acid residues different from Gly, Ala, Ser, and Tyr residues (Val, Thr, Asp, Glu, Phe, etc.) might be present in small but not negligible amounts. Such heterogeneity in size and composition of the chains seems to be a significant role in the structural determination of the Cp fractions. In addition, the presence of phenolic OH and COO- groups of the C-terminal Tyr residue in the Cp fraction might trigger the silk I structure since the structure of the Cp fraction depends largely on the pH of the water during the process of dialysis.29 3. 13C CP/MAS NMR Spectral Analysis of the Mixture of Cp Fraction and Peptide 1. We previously reported a detailed 13C CP/MAS NMR study of peptide mixtures: peptide (AG)7YG(AG)7 or (AG)7FG(AG)7 was mixed at a molecular level with (AG)15 in 1:3 ratio.24 After dissolution in 9 M LiBr solution and dialysis against water, the observed NMR spectral features of the samples showed the characteristics of silk II peaks for Ala and Gly residues, however, without complete disappearance of the silk I structure, possibly originated exclusively from excess amounts of (AG)15 sample. It was proposed that the predominant molecular environment and the nature of intermolecular interaction, induced by the presence of the central Tyr or Phe residues, is able to overcome the silk I form of (AG)15. Since the preferred conformations adopted by the fibrous fibroin generally depend on the nature of intra- and intermolecular interactions, the structural study of OγH-containing

Biomacromolecules, Vol. 6, No. 1, 2005 473

Figure 5. 13C CP/MAS NMR spectra of peptide 12, AGSG(AG)2SGA[1-13C]G[3-13C]A[2-13C]GSG[2-13C]AG[1-13C]AGSG(AG)2SGAG (a); the peptide 12 + Cp fraction in a molar ratio of 1:5 (b); Cp fraction alone (c); and difference spectrum (d), that is, spectrum b - spectrum c. ssb represents the spinning sideband. All peptides were treated in a similar fashion; that is, they are first dissolved in 9 M LiBr and then dialyzed against water.

Ser peptides, mixed with native Cp fraction, was examined. The observed 13C CP/MAS NMR spectrum of the 13C isotope-labeled peptide AGSG(AG)2SGA[1-13C]G[3-13C]A[2-13C]GSG[2-13C]AG[1-13C]AGSG(AG)2SGAG, peptide 12 (after dissolution in 9 M LiBr and dialysis against water), shown in Figure 5a, clearly indicates that the structure is silk II; the spectral characteristics are the same as those of nonlabeled peptide 6. Interestingly, when peptide 12 was mixed with Cp fraction in a molar ratio of 1:5 and dissolved in 9 M LiBr and then dialyzed against water, the apparent 13 C CP/MAS NMR spectrum (shown in Figure 5b) displayed the 13C chemical shifts for four 13C-labeled sites: 16.8 ppm for Ala Cβ, 50.9 ppm for Ala CR, 176.4 ppm for Ala CdO, 43.5 ppm for Gly CR, and 170.9 ppm for Gly CdO carbons, which originated from peptide 12.18 These characteristic chemical shifts are in good agreement with those of the Cp fraction in silk I form as illustrated in Figure 5c. Thus, the structure of (AGSGAG)5 sequence radically changed from silk II to silk I, presumably due to reorientation of various intermolecular interactions of the peptide, induced by the preferred conformation of the Cp fraction. Further evidence to substantiate this fact was obtained by subtracting the 13C NMR spectrum of Cp fraction with silk I form (Figure 5c) from that of the spectrum observed for the mixture of peptide

474

Biomacromolecules, Vol. 6, No. 1, 2005

12 (Figure 5a) and the Cp fraction (Figure 5b). Evidently, the difference spectrum (Figure 5d) reveals the characteristics of silk I spectrum, which are quite distinct from the spectrum of silk II form (Figure 5a). It is noteworthy that the results signifying induced structural changes clearly correlate with our previous experiments where the peptides (AG)7YG(AG)7 and (AG)7FG(AG)7 were mixed with (AG)15 sequence.24 Thus, the preferred conformation adopted by the fibrous fibroin may be altered via intermolecular interactions and the side-chain orientations of Ser residues (peptide 6); that is, the pattern of hydrogen-bond formation through OγH groups may control the creation of silk I structure, nevertheless, in connection with the surrounded molecules with silk I form of the Cp fraction. Acknowledgment. T.A. acknowledges supports from the Insect Technology Project, Japan, and Agriculture Biotechnology Project, Japan. R.K. was supported by the Japan Society for the Promotion of Science Invitation Fellowship. References and Notes (1) Asakura, T.; Kapl. D. L. In Encyclopedia of Agricultural Science; Arutzen, C. J., Ed.; Academic Press: New York, 1994; Vol. 4, pp 1-11. (2) Mita, K.; Ichimura, S.; James, C. T. J. Mol. EVol. 1994, 38, 583592. (3) Zhou, C.; Confalonieri, F.; Medina, N.; Zivanovic, Y.; Esnault, C.; Yang, T.; Jacquet, M.; Janin, J.; Duguet, M.; Perasso, R.; Li, Z. Nucleic Acids Res. 2000, 28, 2413-2419. (4) Marsh, R.; Corey, R. B.; Pauling, L. Biochim. Biophys. Acta 1995, 16, 1-34. (5) Fraser, R. D. B.; MacRae, T. P. In Conformations of Fibrous Proteins and Related Synthetic Polypeptides, Academic Press: New York, 1973. (6) Takahashi, Y.; Gehoh, M.; Yuzuriha, K. Int. J. Biol. Macromol. 1999, 24, 127-138. (7) Konishi, T.; Kurokawa, M. Sen’I Gakkaishi 1968, 24, 550-554. (8) Okuyama, K.; Takanashi, K.; Nakajima, Y.; Hasegawa, Y.; Hirabayashi, K.; Nishi, N. J. Seric. Sci. Jpn. 1998, 57, 23-30.

Asakura et al. (9) Anderson, J. P. Biopolymers 1998, 45, 307-321. (10) Lotz, B.; Keith, H. D. J. Mol. Biol. 1971, 61, 201-215. (11) He, S.; Valluzzi, R.; Gido, S. P. Int. J. Biol. Macromol. 1999, 24, 187-195. (12) Lotz, B.; Cesari, F. C. Biochimie 1979, 61, 205-214. (13) Fossey, S. A.; Nemethy, G.; Gibson, K. D.; Scheraga, H. A. Biopolymers 1991, 31, 1529-1541. (14) Asakura, T.; Kuzuhara, A.; Tabeta, R.; Saito, H. Macromolecules 1985, 18, 1841-1845. (15) Monti, P.; Taddei, P.; Freddi, G.; Asakura, T.; Tsukada, M. J. Raman Spectrosc. 2001, 32, 12-14. (16) Monti, P.; Taddei, P.; Freddi, G.; Ohgo, K.; Asakura, T. Biopolymers 2003, 72, 329-338. (17) Taddei, P.; Asakura, T.; Yao, J.; Monti, P. Biopolymers 2004, in press. (18) Yao, J.; Nakazawa, Y.; Asakura, T. Biomacromolecules 2004, 5, 680-688. (19) Asakura, T.; Ashida, J.; Yamane, T.; Kameda, T.; Nakazawa, Y.; Ohgo, K.; Komatsu, K. J. Mol. Biol. 2001, 306, 291-305. (20) Asakura, T.; Yamane, T.; Nakazawa, Y.; Kameda, T.; Ando, K. Biopolymers 2001, 58, 521-525. (21) Asakura, T.; Demura, M.; Date, T.; Miyashita, N.; Ogawa, K.; Williamson, M. P. Biopolymers 1997, 41, 193-203. (22) Kameda, T.; Asakura, T. Annu. Rep. NMR Spectrosc. 2002, 46, 101149. (23) Asakura, T.; Sugino, R.; Yao, J.; Takashima, H.; Kishore, R. Biochemistry 2002, 41, 4415-4424. (24) Asakura, T.; Suita, K.; Kameda, T.; Afonin, S.; Ulrich, A. S. Magn. Reson. Chem. 2004, 42, 258-66. (25) Yao, J.; Ohgo, K.; Sugino, R.; Kishore, R.; Asakura, T. Biomacromolecules 2004, 5, 1763-1769. (26) Strydom, D. J.; Haylett, T.; Stead, R. H. Biochem. Biophys. Res. Commun. 1977, 79, 932-938. (27) Asakura, T.; Yao, J.; Yamane, T.; Umemura, K.; Ulrich, A. S. J. Am. Chem. Soc. 2002, 124, 8794-8795. (28) Yamane, T.; Umemura, K.; Asakura, T. Macromolecules 2002, 35, 8831-8838. (29) Nakajima, Y.; Takanashi, K.; Okuyama, K.; Hirabayashi, K. J. Seric. Sci. Jpn. 1987, 56, 83-84. (30) Asakura, T.; Sugino, R.; Okumura, T.; Nakazawa, Y. Protein Sci. 2002, 11, 1873-1877. (31) Kameda, T.; Ohkawa, Y.; Yoshizawa, K.; Naito, J.; Ulrich, A. S.; Asakura, T. Macromolecules 1999, 32, 7166-7171.

BM049487K