Vibrational Analysis of Amino Acids and Short Peptides in Aqueous

12796

J. Phys. Chem. B 2009, 113, 12796–12803

Vibrational Analysis of Amino Acids and Short Peptides in Aqueous Media. V. The Effect of the Disulfide Bridge on the Structural Features of the Peptide Hormone Somatostatin-14 Beleń Hernańdez,† Claude Carelli,† Yves-Marie Coïc,‡ Joe¨l De Coninck,§ and Mahmoud Ghomi*,† Groupe de Biophysique Molećulaire (GBM), UFR SMBH, UniVersite´ Paris 13, 74 rue Marcel Cachin, 93017 Bobigny Cedex, France, Unite´ de Chimie des Biomolećules, URA 2128, De´partement de BSC, Institut Pasteur, 28 rue du Docteur Roux, 75724 Paris Cedex 15, France, and Centre de Recherche en Mode´lisation Molećulaire (CRMM), UniVersite´ de Mons-Hainaut, Materia NoVa, Parc Initialis, AVenue Copernic 1, B-7000 Mons, Belgium ReceiVed: May 20, 2009; ReVised Manuscript ReceiVed: July 15, 2009

To emphasize the role played by the S-S bridge in the structural features of somatostatin-14 (SST-14), newly recorded CD and Raman spectra of this cyclic peptide and its open analogue obtained by Cys f Ser substitution are presented. CD spectra of both peptides recorded in aqueous solutions in the 100-500 µM concentration range are strikingly similar. They reveal principally that random conformers constitute the major population in both peptides. Consequently, the S-S bridge has no structuring effect at submillimolar concentrations. In methanol, the CD spectrum of somatostatin-14 keeps globally the same spectral shape as that observed in water, whereas its open analogue presents a major population of helical conformers. Raman spectra recorded as a function of peptide concentration (5-20 mM) and also in the presence of 150 mM NaCl provide valuable conformational information. All Raman spectra present a mixture of random and β-hairpin structures for both cyclic and open peptides. More importantly, the presence or the absence of the disulfide bridge does not seem to influence considerably different populations of secondary structures within this range of concentrations. CD and Raman data obtained in the submillimolar and millimolar ranges of concentrations, respectively, lead us to accept the idea that SST-14 monomers aggregate upon increasing concentration, thus stabilizing β-hairpin conformations in solution. However, even at high concentrations, random conformers do not disappear. Raman spectra of SST-14 also reveal a concentration effect on the flexibility of the S-S linkage and consequently on that of its cyclic part. In conclusion, although the disulfide linkage does not seem to markedly influence the SST-14 conformational features in aqueous solutions, its presence seems to be necessary to ensure the flexibility of the cyclic part of this peptide and to maintain its closed structure in lower dielectric constant environments. 1. Introduction Somatostatin-14 (SST-14 or SRIF) is a natural cyclic 14mer peptide, closed by a disulfide linkage between two cysteines located at its 3rd and 14th positions (Figure 1). This peptide hormone is widely distributed throughout the central nervous system and peripheral tissues and exerts a variety of physiological effects, such as inhibition of growth-promoting hormones, growth factor secretion, angiogenesis, and so forth, through binding to specific cell surface receptors (SSTRs). SST14 produces a variety of effects which are mediated both indirectly through inhibition of hormones and growth factors as well as directly via SSTRs, present on cells, to inhibit mitogenic signaling of growth factor receptor kinases, leading to growth arrest and induction of apoptosis.1,2 Up to now, five different subtypes of somatostatin receptors (SSTR-1 to SSTR-5), with similar affinities for the endogenous ligand but different distributions in various tissues, have been characterized.3-6 For instance, in the case of acromegaly, that is, growth hormone (GH)-secreting adenomas7 or TSH-secreting * To whom correspondence should be addressed. E-mail: mahmoud.ghomi@ univ-paris13.fr. Tel: +33-1-48388928. Fax: +33-1-48387356. † Universite´ Paris 13. ‡ Institut Pasteur. § Universite´ de Mons-Hainaut.

adenomas, the activations of SSTR-2 and SSTR-5 have been associated with GH and TSH suppression. In contrast, the activation of SSTR-2 and not SSTR-5 has been associated with treating prolactin (PRL)-secreting adenomas.8-11 Consequently, taking into account the importance of SST-14 receptors, many disorders in oncology, ophthalmology, and other fields of medicine could be treated through the use of selective SST analogues for SSTR subtypes responsible for the desired response. However, the main limitation of their therapeutic effectiveness is basically related to their receptor selectivity and their short half-life. During the recent years, numerous modified molecules with prolonged half-life and increased potency were developed.12-20 Among them, those based on cyclic peptides containing a S-S bridge (like SST-14) but with a shorter length and including D-amino acids, such as the octamers named octreotide and lanreotide (Figure 1), should be emphasized. Both of the mentioned octapeptides have shown their high affinity to SSTR-2, accompanied by a mediate affinity to SSTR-3 and SSTR-5.21,22 Since the earliest investigations on the structure-activity relationship of SST-14, the cyclic feature of SST-14 (and consequently, its analogues), maintained by a disulfide bridge, was supposed to play a pivotal role in the interaction with the corresponding cellular receptors. Thus, in a first step, it seemed

10.1021/jp904737v CCC: $40.75  2009 American Chemical Society Published on Web 08/26/2009

CD and Raman Data for Somatostatin-14 and Its Open Analogue

J. Phys. Chem. B, Vol. 113, No. 38, 2009 12797

Figure 1. Schematic representation of amino acid composition of SST-14 (A), open SST-14 (B), octreotide (C), and lanreotide (D). Note that open SST-14 is obtained by replacing in SST-14 the cysteines at the 3rd and 14th positions by serines (underlined).

interesting to us to verify whether the disulfide linkage is absolutely necessary to give to SST-14 particular conformational features in aqueous solution. To address this question, we have focused our attention on the comparison between the structural features of two 14-mer peptides, (i) SST-14 and (ii) one of its analogues, hereafter referred to as open SST-14, obtained by cysteine f serine replacement at the 3rd and 14th positions (Figure 1). The choice of serine in the design of open STT-14 arises from the similarity of its chemical composition with cysteine. Comparison has also been made between the CD spectra recorded in water (strongly polar solvent) and in methanol (low dielectric constant solvent) to emphasize preferentially the effect of aqueous media on the SST-14 conformational features. As described in our recent publications,23-27 especially in manuscripts II and III of the present series,25,26 an experimental protocol based on the joint use of two optical spectroscopic techniques, that is, circular dichroism (CD) and Raman scattering, has shown its capability to analyze the structural and dynamical properties of short minimalist LK and LR peptides, used as cationic vectors for DNA delivery into living cells. In the present work, we apply the same protocol to the study of somatostatin-14 and its open analogue. 2. Materials and Methods Somatostatin-14, or SST-14, Nter-AGCKNFFWKTFTSC-Cter (Figure 1) was purchased from Sigma-Aldrich. A second peptide, called open SST-14, obtained by the substitution of both cysteins of SST-14 (at the 3rd and 14th positions) by serines, that is, Nter-AGSKNFFWKTFTSS-Cter (Figure 1), was purchased from GENOSPHERE Biothechnologies, France. Both peptides SST-14 and open SST-14 contained TFA (trifluoracetate) as lysine counterions. Sodium chloride was from Merck (purity > 99.5%). Methanol was provided by Carlo Erba (ACS-ISO for analysis). Fresh pure water was obtained from a Millipore filtration system. D2O (100% purity) was provided by Euriso-top (Saclay, France).

Purity control of the peptides was assessed by RP-HPLC on an Agilent 1100 Series liquid chromatograph and monitored with a photodiode array detector by absorbance at 230 nm. A linear gradient of 22-32% solvent B (acetonitrile) in solvent A (0.08% aqueous trifluoroacetic acid) over 20 min (0.5% B/min) was applied at a 0.35 mL/min flow rate on a Symmetry300 C18 3.5 µm 2.1 × 100 mm column (Waters). Mass spectrometry was carried out on a quadrupole time-offlight (Q-TOF) micromass spectrometer (Waters, Manchester, UK) equipped with a Z-spray API source and calibrated with a phosphoric acid calibration solution. Capillary, sample cone and extraction cone voltages were set at 3 kV, 40 V, and 10 V, respectively. Source and desolvation temperatures were set at 80 and 250 °C, respectively. Data were acquired by scanning over the m/z range of 150-2000 at a scan rate of 1 s and an interscan delay of 0.1 s. Peptides were dissolved in a mixture of water/methanol/acetic acid 49.5:49.5:1 v/v/v at a concentration of 1 µg/µL and analyzed in positive ion mode by infusion at a flow rate of 5 µL/min. Three hundred spectra were combined, and the resultant raw multicharged spectra were processed using the MaxEnt 3 deconvolution algorithm embedded in the Masslynx software. As far as CD and Raman measurements are concerned, solutions were prepared by dissolving the lyophilized powder samples of the peptides in pure water containing or not containing 150 mM NaCl. Deuterated samples were prepared by dissolving the peptide directly in pure D2O. Peptide stock solutions at 20 mM were prepared and diluted in order to reach other concentrations, such as 10 and 5 mM for Raman spectroscopy and 500, 250, and 100 µM for CD spectroscopy. CD experiments above 500 µM are known as sample-consuming because they need short path length (0.01 mm) holders, from which analyzed solutions cannot be extracted for subsequent experiments. Thus, in the millimolar range, we only resorted to Raman spectroscopy, which is a powerful technique for analyzing peptide secondary structures (vide infra). Methanol solutions were prepared by diluting in MeOH the 20 mM stock solution (prepared in pure water) to reach the final concentrations

12798

J. Phys. Chem. B, Vol. 113, No. 38, 2009

Hernańdez et al.

Figure 3. Comparison between the CD spectra of SST-14 (circles) and open SST-14 (squares) observed in the 195-300 nm region in methanol at 250 µM peptide concentration.

Raman spectra of SST-14 and assigned to the photodecomposition of the Trp residue,28 was observed in our experiments. The solution remained clear without any color change during the whole experiment. We have previously shown24 the negligible contribution of TFA to the amide I and III regions, taken as reference for secondary structural assignments. The analysis of Raman spectra was performed by curve fitting. Gaussian + Lorentzian functions, with the Lorentzian contribution kept equal to or greater than 50%, were employed in band decompositions.24-27 Postprocessing of Raman spectra was performed using the GRAMS/32 package (Galactic Industries). Figures shown in this paper were drawn using the SIGMAPLOT (Systat Software Inc., Point Richmond, CA) package. 3. Results Figure 2. CD spectra of SST-14 (A) and open SST-14 (B) in aqueous solutions observed in the 195-300 nm spectral region at a 250 µM peptide concentration. Circles: CD spectrum in pure water. Crosses: CD spectrum from the aqueous solution containing 150 mM NaCl. Note that in each case, the spectral shape is independent of the peptide concentration in the 100-500 µM range.

included in the 100-500 µM range for recording CD spectra. Dichroic signals from peptide samples were analyzed on a JASCO J-810 spectrophotometer. Samples were placed in suprasil quartz cells with a 1 mm path length. Each spectrum recorded in the 190-300 nm region corresponds to the average of five scans with a speed of 100 nm/min (5 min of accumulation). CD spectra were baseline corrected, and the measured ellipticity for each sample was normalized and expressed in deg cm2 dmol-1. For recording Raman spectra, 30 µL of a peptide solution placed in suprasil quartz cells (5 mm path length) was excited with the 488 nm line of an Ar+ laser (Stabilite model 201704S, Spectra Physics), and scattered photons were collected at a right angle on a Jobin-Yvon T64000 spectrograph in a single configuration with a 1200 grooves/mm holographic grating and a holographic notch filter. The effective spectral slit width was set to ∼5 cm-1. Stokes Raman spectra were analyzed at room temperature in the 1750-400 cm-1 region (40 min of accumulation). No large fluorescence background developed under laser light at 488 nm, as indicated in the previously reported

HPLC Analysis and Electrospray Ionization Mass Spectrometry. UV chromatograms showed a good purity for both peptides (SST-14: >99%; open SST-14: 93%). Furthermore, the experimental data from mass spectroscopy were consistent with the expected masses (SST-14 monoisotopic [M + H]+ 1637.72 d, observed 1637.76 d; open SST-14 monoisotopic [M + H]+ 1648.81 d, observed 1648.85 d). CD Spectra. The first CD spectra of SST-14 were described in 1976,29 accompanied by a subsequent report with additional data from the D-Phe6, D-Phe7, D-Trp8, and D-Phe11 diastereomers of the same peptide.30 This set of data recorded in the 100-160 µM concentration range, depending on the analyzed spectral region, permitted assignment of the observed peaks. Although no evident characteristic marker from β-turns and/or β-sheets was present in theses spectra, a rigid β-hairpin structure (with Trp8 located on the top of the turn) stabilized by both an intramolecular H-bond between two intramolecular β-strands as well as a S-S linkage between the two cysteines was proposed by the authors.29 This preliminary structure, giving a nonflexible image of the SST-14 secondary structure, was considered in many subsequent discussions on the structural features of this peptide hormone. Figures 2 and 3 display the CD spectra of SST-14 and open SST-14 in aqueous solutions and in methanol, respectively. It is to be noted that the CD spectral shape is concentrationindependent in the 100-500 µM range. This is the reason why we show here only the spectra at 250 µM (middle concentration


J. Phys. Chem. B, Vol. 113, No. 38, 2009 12799

Figure 4. Raman spectra obtained from aqueous solutions of SST-14 (A) and open SST-14 (B). The traces in (A) correspond to the spectra recorded in H2O (in blue) and in D2O (in red). The trace in (B) corresponds to the Raman spectrum in H2O. The peptide concentration was 20 mM, and the exciting laser wavelength was selected to be 488 nm. Raman bands including TFA contributions are marked by asterisks.

used in this work). Although in the 220-280 nm region, we observe a similar spectral shape for SST-14 (Figure 2A) as that previously published,28,29 the present spectra also show a deep mimimum at ∼200 nm, which has never been documented up to now. The presence of this minimum is of pivotal importance in the structural assignments of SST-14 in water. It clearly shows that random SST-14 conformers are predominant in aqueous solutions. We recall that random conformers are generally characterized by a negative CD signal at ∼198 nm. Surprisingly, open SST-14 in aqueous solution gives rise to a similar CD spectrum as that observed for SST-14 (Figure 2B). This new result shows that both peptides, at least within the 100-500 µM concentration range, are basically randomly structured and manifest no major population of conformers containing β-turns or β-sheets. We also mention that βI-turn provides a CD signal with a double minima (208-222 nm), resembling that of an R-helix conformation; βII-turn is characterized by a positive signal in the low wavelength region (190-210 nm), and β-sheet (β-strand) structures provide a negative CD signal at ∼218 nm.31 In methanol, the situation becomes rather different (Figure 3). Although SST-14 keeps a similar CD signal shape as that observed in aqueous solutions (Figure 2) with, however, a better resolved shoulder at ∼214 nm, assignable to β-type conformers, open SST-14 provides a new CD signal, characteristic of an R-helical structure (Figure 3). Nevertheless, the low intensity of the positive CD signal at ∼200 nm, along with the higher intensity of the negative signal at the 208 nm band compared to that at 222 nm (Figure 3), leads us to conclude that there still remains a random population in open SST-14. The structuring effect of methanol, as well as many other low dielectric constant solvents, leading generally to the formation of helical structures, has been discussed previously.32,33

Raman Spectra. The first Raman spectra of SST-14 were published in 198028 as obtained from highly concentrated (60 mM) H2O and D2O solutions. Although the most intense bands observed in these spectra could be carefully assigned, the whole spectral resolution, especially in the amide I and amide III regions, was not sufficient for further conformational analysis. Being really surprised by the results obtained from the new CD spectra (see above), we have analyzed the Raman spectra of SST-14 and its open analogue as a function of peptide concentration (5-20 mM) and also in the presence of NaCl. In Figure 4, we display the Raman spectra of both peptides at 20 mM. Note that the amide III bands (1350-1250 cm-1 region, mainly from the backbone N-H bending) vanish completely upon deuteration. In order to highlight the subtle changes appearing in the most characteristic spectral regions, we have focused in Figures 5-7 on the amide I, amide III, and S-S linkage regions, respectively. A tentative band decomposition in these regions allowed us to clarify our discussion in the following section. 4. Discussion The most striking effect presented here is, in fact, the similarity of the CD spectra of both peptides in aqueous solutions. This confirms that both peptides are basically randomly structured. In other words, the existence of the S-S bridge does not seem to induce a perceptible structuring effect in the closed part of SST-14 within the concentrations used in CD spectroscopy. In methanol, open SST-14 seems to prefer a helical structure in methanol. Raman spectra recorded at much higher concentrations (at least 10-40 times higher that those used for CD spectros-

12800


Hernańdez et al.

Figure 5. Focus on the amide I region (1725-1625 cm-1) of the Raman spectra of SST-14 (A) and open SST-14 (B) recorded in H2O samples (Figure 4). See also Table 1. The band decomposition of this spectral region is also presented (see section 2 for details). Circles correspond to the sum of the calculated components.

Figure 6. Focus on the amide III region (1325-1225 cm-1) of the Raman spectra of SST-14 (A) and open SST-14 (B) recorded in H2O samples (Figure 4). The band decomposition of this spectral region is also presented (see section 2 for details). Circles correspond to the sum of the calculated components.

copy; see above) reveal new and interesting information. First of all, the existence of β-strands (which were negligible in the 100-500 µM concentration range; see the CD results above) is clearly evidenced in both peptides through the amide I and amide III regions (see Table 1 and Figures 5 and 6). For instance, the components at ∼1665 (amide I) and 1236 cm-1 (amide III) are both the known structural markers of β-strands.24-26 On the other hand, other markers at ∼1678 and 1652 cm-1 (amide I, Figure 5) along with three others at ∼1304, 1281, and 1253 cm-1 (amide III, Figure 6) prove the existence of different types of β-turns (βI, βII, and βVIII).34-37 The simultaneous presence of β-strand and β-turn markers confirm the existence of β-hairpins.38 Finally, the components at ∼1689 (amide I) and 1263 cm-1 (amide III) are both the fingerprints of random chains. In both peptides, a general decrease of random conformers versus an increase of β-hairpins is found on the basis of the Raman markers as a function of concentration (Table 1). We can conclude that both peptides contain a comparable amount of β-hairpin and random conformers at a concentration selected in the 5-20 mM range. The addition of NaCl to the peptide solutions at 20 mM does not alter considerably the conformational equilibrium in aqueous solutions.

Our structural results obtained by means of Raman spectra corroborate those based on the infrared absorption data recorded in concentrated SST-14 samples (6-120 mM).39 Analysis of the amide I region in infrared spectra has shown a gradual decrease of random conformers (without complete vanishing) accompanied by an increase of β-hairpin species. The same work37 also reported appreciable results obtained by the use of electron microscopy. The observed patterns clearly evidenced the aggregation of SST-14 monomers upon increasing concentration, leading to the formation of organized nanofibrils with variable diameters. Consequently, our presently described results lead us to admit that the S-S bridge does not have a considerable effect on the SST-14 conformational features. The question raised here is obviously related to the exact role of the S-S linkage in SST14. To bring an answer to this question, we have focused on the Raman bands mainly arising from the S-S bond stretching vibrations (Figure 7) by considering the conclusions derived from the previous works devoted to disulfide linkage in peptides and proteins.39-42 It has been shown that the corresponding Raman bands are sensitive to the conformation in the vicinity of the S-S linkage between the two linked cysteines. Briefly,


J. Phys. Chem. B, Vol. 113, No. 38, 2009 12801 TABLE 1: Different Possible Secondary Structures Appearing in an Aqueous Solution of SST-14 and Its Open Analoguea concentration components

20 mM/ secondary 5 mM 10 mM 20 mMb 150 mM NaCl structure SST-14 27 37 22 14 100

1689 1678 1665 1652 sum of areas

48 30 12 10 100

31 38 17 14 100

1687 1676 1661 1648 sum of areas

37 30 19 14 100

Open SST-14 32 30 31 32 25 24 12 14 100 100

25 37 24 14 100

random β-turn β-strand β-turn

26 31 26 17 100

random β-turn β-strand β-turn

a Different contributions (with a sum normalized to 100 for each sample) are determined by means of band decomposition of the amide I region. Each contribution represents a normalized band area. Contributions correspond, in fact, to average values within an accuracy of (5 for each of them. b See Figure 5 for band decomposition.

Figure 7. Raman spectra of SST-14 as a function of peptide concentration recorded in H2O and displayed in the 575-450 cm-1 region. The band decomposition of this spectral region by means of three main components located at 510, 525, and 540 cm-1 corresponding to the gauche-gauche-gauche, gauche-gauche-trans, and transgauche-trans conformations of the S-S bridge, respectively, is also presented (see section 2 for details). Circles correspond to the sum of the calculated components. The population of each conformer as calculated on the basis of the corresponding band area could be estimated as follows: (Bottom, 5 mM) 510 (60%), 525 (6%), and 540 (33%); (Middle, 10 mM) 510 (70%), 525 (7%), and 540 (23%); (Top, 20 mM) 510 (90%), 525 (3%), and 540 (7%). These contributions correspond in fact to average values estimated within an accuracy of 5% for each of them.

a possible combination of the three torsion angles defined along the three successive chemical bonds in the -Cβ-S-S-Cβsegment gives rise to three well-known characteristic Raman markers located at ∼510 (gauche-gauche-gauche), ∼525 (gauche-gauche-trans), and ∼540 cm-1 (trans-gauchetrans).42 Raman spectra recorded as a function of the SST-14 concentration (5-20 mM, Figure 7) show distinctly the presence of two major conformers, that is, gauche-gauche-gauche and gauche-gauche-trans, at lower concentrations (5 and 10 mM), revealed by the two components at 510 and 540 cm-1. A gradual vanishing of the mode at 540 cm-1 upon increasing concentration can also be observed in these spectra. In other words, for the concentrations up to ∼15 mM, the S-S linkage can fluctuate basically between gauche-gauche-gauche and gauchegauche-trans conformations. Above 15 mM, the S-S linkage appears to be maintained in the gauche-gauche-gauche conformation. By means of molecular modeling (results not shown), we have tried to visualize the effect of the transgauche-trans to gauche-gauche-gauche conformational transition. It appeared that this transition leads to a decrease of the distance between the CR atoms of the two linked cysteines and consequently to the contraction of the peptide chain. We can deduce that the S-S bridge and particularly its conformational transition may play a role in the flexibility (relaxation and stress) of the closed cyclic part of SST-14. The most rigid conformation, that is, gauche-gauche-gauche, is gradually preferred upon increasing concentration, presumably because molecular aggregation is increased. Moreover, on the basis of these new Raman data, one can now understand better why in the previously published Raman spectra,28 recorded at a much higher concentration (60 mM), only the ∼510 cm-1 Raman mode was observed. Upon this observation, the authors had confirmed the presence of rigid structured conformers (β-hairpins).28 It should be noted that the inherent flexibility of the cyclic part of SST-14 has been previously mentioned by NMR data.43-45 For instance, it has been indicated that SST-14 does not adopt a unique stable structure in aqueous solution but several “rapidly interconverting” conformers. Especially the most recent NMR analysis45 supports the idea that intercon-

12802


verting β-turns may appear in the whole range of the cyclic part, that is, in going from Cys3 to Cys14 residues. It should be emphasized that the concentrations used for NMR spectroscopy are of the same magnitude as those used here for Raman spectroscopy, and the propensity of SST-14 to aggregate was also mentioned in relation to severe NMR line broadenings at concentrations above 15 mM.43

Hernańdez et al. Ref 13982ZL), covering the travels and scientific exchanges between the partners during the period 2007-2008. The scientific collaboration between the two French laboratories located at the Institut Pasteur and at the University Paris 13 is protected by a Research Contract (Contrat de Collaboration de Recherche) since January 2009. References and Notes

5. Concluding Remarks First of all, the rigid β-hairpin structure model, proposed previously for the closed part of SST-14, might be reconsidered, at least within submillimolar concentrations. At these concentrations, SST-14 might be extremely flexible, thus explaining its modest affinity22 toward its receptors (see section 1 for details). To elaborate on SST-14 peptidic analogues with shorter sequences, the S-S bridge was generally maintained, whereas the cyclic part of SST-14 was shortened to a limited, but efficient, number of amino acids. For instance, octreotide and lanreotide (Figure 1) have both only four amino acids in their closed part, that is, an average size for a β-turn. Obviously, this way of operating, leading to a considerable decrease of the conformational flexibility, may explain the higher affinity of these two analogues toward a limited number of SST-14 receptors (see section 1 for details). Upon increasing the concentration and in the course of the aggregation process, as evidenced by small-angle X-ray diffraction,38 the interaction between the cyclic parts of SST14 monomers may lead them to undergo a random f β-hairpin transition. However, this transition is never complete, as evidenced by the presence of both random and β-hairpin markers in the present Raman spectra (recorded up to 20 mM), as well as in the recently published infrared spectra (up to 120 mM).38 A recent work based on the combination of several physical techniques (FT-IR, Raman, WAXS, SAXS, and electron microscopy) could bring information about the role of the S-S bridge in the conformational features and aggregation properties of lanreotide (Figure 1D). In this framework, comparison was made between the native compound (lanreotide) and its reduced (obtained by S-S to S-H reduction) and open (obtained by Cys to Ala substitution) analogues.46 The obtained data have shown that the reduced and open analogues prefer to self-assemble into organized linear β-sheets, different from the nanofibrils based on the cyclic β-hairpin aggregates formed in the native compound. On the basis of the presently described results, one cannot accept the same conclusion for SST-14, in which the S-S linkage does not seem to have a major effect on its conformational features (see the comparison between the data obtained from the closed and open SST-14 above). We attribute this difference to the chain length and amino acid composition of the two peptides SST-14 and lanreotide. The presence of the S-S bridge seems however to be necessary to ensure the flexibility of its cyclic part and to maintain the stability of its closed structure in lower dielectric constant environments. Further investigations are necessary in order to (i) evidence whether the disulfide linkage can be considered as the major element for the SST-14 folding and (ii) extend the discussion on the structure/activity relationship of SST-14 and its peptidic analogues. Acknowledgment. The cooperation between the French (GBM) and Belgian (CRMM) laboratories was supported by a PHC (Partenariat Hubert Curien)-Tournesol France-Belgique,

(1) Patel, Y. C. Front Neuroendocrinol. 1999, 20, 157–198. (2) Sharma, K.; Patel, Y. C.; Srikant, C. B. Mol. Endocrinol. 1999, 13, 82–90. (3) Reisine, T.; Bell, G. I. Endocr. ReV. 1995, 16, 427–442. (4) Lamberts, S. W. J. Endocr. ReV. 1988, 9, 417–436. (5) Lamberts, S. W. J.; Krenning, P. E.; Reubi, J. C. Endocr. ReV. 1991, 12, 450–482. (6) Schonbrunn, A. Ann. Oncol. 1999, 10, S17-S21. (7) Melmed, S.; Sterneberg, R.; Cook, D.; Klibanski, A.; Chanson, P.; Bonert, V.; Vance, M. L.; Rhew, D.; Kleinberg, D.; Barkan, A. J. Clin. Endocrionol. Metab. 2005, 90, 4405–4410. (8) Shimon, I.; Yan, X.; Taylor, J. E.; Bitonte, R. A.; Kim, S.; Morgan, B.; Coy, D. H.; Culler, M. D.; Melmed, S. J. Clin. InVest. 1997, 99, 789– 798. (9) Shimon, I.; Yan, X.; Taylor, J. E.; Weiss, M. H.; Culler, M. D.; Melmed, S. J. Clin. InVest. 1997, 100, 2386–2392. (10) Shimon, I. Endocrine 2003, 20, 265–269. (11) Hofland, L.; Lamberts, S. Front Hormone Res. 2004, 32, 235–252. (12) Bauer, W.; Briner, U.; Doepfner, W.; Haller, R.; Huguenin, R.; Marbach, P.; Petcher, T. J.; Pless, J. Life Sci. 1982, 31, 1133–1140. (13) Murphy, W. A.; Lance, V. A.; Moreau, S.; Moreau, J.; Coy, D. H. Life Sci. 1987, 40, 2515–2522. (14) Chanson, P.; Warnet, A. Metabolism 1992, 41, 62–65. (15) Gancel, A.; Vuillermet, P.; Legrand, A.; Catus, F.; Thomas, F.; Kuhn, J. M. Clin. Endocrinol. 1994, 40, 421–428. (16) Iglesias, P.; Diez, J. J. J. Endocrinol. InVest. 1998, 21, 775–778. (17) Shimatsu, A.; Murabe, H.; Kamoi, K.; Suzuki, Y.; Nakao, K. Endocr. J. 1999, 46, 113–123. (18) Danila, D. C.; Haidar, J. N.; Zhang, X.; Katznelson, L.; Culleer, M. D.; Klinbanski, A. J. Clin. Endocrinol. Metab. 2001, 86, 2976–2981. (19) Caron, P.; Arlot, S.; Bauters, C.; Chanson, P.; Kuhn, J. M.; Pugeat, M.; Marechaud, R.; Teutsch, C.; Vidal, E.; Sassano, P. J. Clin. Endocrinol. Metab. 2001, 86, 2849–2853. (20) Lamberts, S.; van der Lely, A.; Hofland, L. Eur. J. Endocrinol. 2002, 146, 701–705. (21) Weckbecker, G.; Lewis, I.; Albert, R.; Schmid, H. A.; Hoyer, D.; Bruns, C. Nat. ReV. Drug DiscoVery 2003, 2, 999–1017. (22) Pawlikowski, M.; Melen´-Mucha, G. Curr. Opin. Pharmacol. 2004, 4, 608–613. (23) Boukhalfa-Heniche, F. Z.; Hernańdez, B.; Gaillard, S.; Coïc, Y. M.; Huynh-Dinh, T.; Lecouvey, M.; Seksek, O.; Ghomi, M. Biopolymers 2004, 73, 727–34. (24) Hernańdez, B.; Boukhalfa-Heniche, F. Z.; Coı¨c, Y. M.; Seksek, O.; Ghomi, M. Biopolymers 2006, 81, 8–19. (25) Guiffo Soh, G.; Hernańdez, B.; Coïc, Y. M.; Boukhalfa-Heniche, F. Z.; Ghomi, M. J. Phys. Chem. B 2007, 111, 12563–12572. (26) Guiffo Soh, G.; Hernańdez, B.; Coïc, Y. M.; Boukhalfa-Heniche, F. Z.; Fadda, G.; Ghomi, M. J. Phys. Chem. B 2008, 112, 1282–1289. (27) Tagounits, A.; Briane, D.; Hernańdez, B.; Coïc, Y. M.; Ghomi, M.; Cao, A. Int. J. Biomed. Pharm. Sci. 2007, 1, 135–139. (28) Han, S. L.; Rivier, J. E.; Scheraga, H. A. Int. J. Pept. Protein Res. 1980, 15, 355–364. (29) Holladay, L. A.; Puett, D. Proc. Natl. Acad. Sci. U.S.A. 1976, 112, 1199–1202. (30) Holladay, L. A.; Rivier, J.; Puett, D. Biochemistry 1977, 16, 4895– 4900. (31) Perczel, A.; Fasman, G. D. Protein Sci. 1992, 1, 378–395. (32) Dong, A.; Matsuura, J.; Manning, M. C.; Carpenter, J. F. Arch. Biochem. Biophys. 1998, 355, 275–281. (33) Arunkumar, A. I.; Kumar, T. K. S.; Yu, C. Int. J. Biol. Macromol. 1997, 21, 223–230. (34) Bandekar, J.; Krimm, S. Proc. Natl. Acad. Sci. U.S.A. 1979, 76, 774–777. (35) Ishizaki, H.; Balaram, P.; Nagaraj, R.; Venkatachalapathi, Y.; Tu, A. T. Biophys. J. 1981, 36, 509–517. (36) Mikhonin, A. V.; Bykov, S. V.; Myshakina, N. S.; Asher, S. A. J. Phys. Chem. B 2006, 110, 1928–1943. (37) Thomas, G. J., Jr.; Prescott, B.; Urry, D. W. Biopolymers 2004, 26, 921–934. (38) van Grondelle, W.; Iglesias, C.; Coll, E.; Artzner, F.; Paternostre, M.; Lacombe, F.; Cardus, M.; Martinez, G.; Montes, M.; Cherif-Cheikh, R.; Vale´ry, C. J. Struct. Biol. 2007, 160, 211–223.

CD and Raman Data for Somatostatin-14 and Its Open Analogue (39) van Wart, H. E.; Lewis, A.; Scheraga, H. A.; Saeva, F. D. Proc. Natl. Acad. Sci. U.S.A. 1973, 70, 2619–2623. (40) Ozaki, Y.; Mizuno, A.; Itoh, K.; Iriyama, K. J. Biol. Chem. 1987, 262, 15545–15551. (41) Schlu¨cker, S.; Liang, C.; Strehle, K. R.; DiGiovanna, J. J.; Kraemer, K. H.; Levin, I. W. Biopolymers 2006, 82, 615–622. (42) Ackermann, K. R.; Koster, J.; Schlu¨cker, S. Chem. Phys. 2009, 355, 81–84. (43) Hallenga, K.; van Binst, G.; Scarso, A.; Michel, A.; Knappenberg, M.; Dremier, C.; Brison, J.; Dirkx, J. FEBS Lett. 1980, 119, 47–52.

J. Phys. Chem. B, Vol. 113, No. 38, 2009 12803 (44) Buffington, L. A.; Garsky, V.; Rivier, J.; Gibbons, W. A. Biophys. J. 1982, 41, 299–304. (45) Kaerner, A.; Weaver, K. H.; Rabenstein, D. L. Magn. Reson. Chem. 1996, 34, 587–594. (46) Vale´ry, C.; Pouget, E.; Pandit, A.; Verbavatz, J. M.; Bordes, L.; Boisde´, I.; Cherif-Cheikh, R.; Artzner, F.; Paternostre, M. Biophys. J. 2008, 94, 1782–1795.

JP904737V

Vibrational Analysis of Amino Acids and Short Peptides in Aqueous

Recommend Documents