Abstract
Bio-inspired cyclopeptidic heterodimers built on β-sheet-like hydrogen-bonding networks and bearing photoactive and electroactive chromophores on the outer surface have been prepared. Different cross-strand pairwise relationships between the side chains of the cyclic α,γ-peptides afford the heterodimers as three nonequivalent dimeric species. Steady-state and time-resolved spectroscopies clearly show an electron transfer process from π-extended tetrathiafulvalene, covalently attached to one of the cyclopeptides, to photoexcited [60]fullerene, located on the complementary cyclopeptide. The charge-separated state was stabilized for up to 1 μs before recombining and repopulating the ground state. Our current example shows that cyclopeptidic templates can be successfully used to form light-harvesting/light-converting hybrid ensembles with a distinctive organization of donor and acceptor units able to act as efficient artificial photosystems.
Keywords: cyclic peptide nanotubes, fullerenes, self-assembly
Peptide-based tubular systems are being widely used to mimic Nature's channel-forming structures (1). Stable architectures ranging from β-helices to stacked macrocycles can self-assemble through multiple hydrogen-bonding interactions between and/or within peptide units with backbones capable of adopting an appropriate curvature. However, although a variety of versatile strategies have been developed to produce peptide tubes (2, 3), comparatively little has been done in the way of controlled functionalization of their inner and/or outer surfaces so as to fit them for specific tasks (4).
Peptide nanotubes (PNs) or nanotube segments formed by the self-stacking of two or more cyclic peptides (2, 3, 5) are notable examples of the “bottom-up” approach to functional nanostructures. These self-assembled PNs have found application in biological and medical research and materials science (6–10). Their self-assembly is brought about by hydrogen bonding between their constituent cyclic peptides (CPs). The chirality of the amino acid residues of the CPs is such that the CP backbone forms an essentially flat ring with its CO and N
H groups oriented nearly perpendicular to the ring plane and its side chains radiating outward. This conformation allows each face of the CP to take part in a β-sheet-like hydrogen-bond array with another CP that, depending on the sequence of amino acid residues in the CPs, must be oriented either parallel or antiparallel to the first (11–13). Scheme 1 shows an example of the antiparallel stacking of CPs formed of alternating α- and γ-amino acid residues (α,γ-CPs). Note that the precise orientation of side chains in planes perpendicular to the CP rings depends on backbone interactions within each ring.
Scheme 1.
Self-assembling process of CPs. (Upper) Self-assembled PN formation by stacking of antiparallel CPs. (Lower) Chemical structure of α,γ-CPs 1a–4c, and dimeric nanotube segments formed from 1c, 2c, 3c, and 4c.
Within the nanobiomaterials field, one of the most actively pursued goals is the design of highly efficient and highly directional electron transfer mimics of the photosynthetic systems of plants and bacteria. In principle, self-assembled PNs bearing an appropriate array of photoactive and electroactive units might achieve this goal. Here, we describe the synthesis and physico-chemical properties of a class of nanotubular cyclopeptide heterodimers in which one CP bears an electron acceptor (C60), whose physico-chemical properties, its small reorganization energy in electron transfer reactions, and its ability to accept up to six electrons make this and other fullerenes attractive as potential electron-accepting units in functional assemblies and supramolecular arrays (14, 15), and the other an electron donor {2-[9-(1,3-dithiol-2-ylidene)anthracen-10(9H)-ylidene]-1,3-dithiole, exTTF} (16, 17).
As CPs we used α,γ-CPs based on 1c and 3c (Scheme 1), in which the all-trans conformation required for CP ring flatness is ensured by the alternation of d-α-amino acids with a (1R,3S)-3-aminocycloalkanecarboxylic acid (l-γ-Aca) (18, 19). It is known that heterodimerization is particularly strong between an α,γ-CP in which the γ-amino acid is γ-aminocyclohexanecarboxylic acid (γ-Ach) and an α,γ-CP in which the γ-amino acid is γ-aminocyclopentanecarboxylic acid (γ-Acp) (20). The target heterodimer 11 is formed by the self-assembly of γ-Acp-based CP 7, and the γ-Ach-based CP 10, decorated with C60 and exTTF units, respectively. It was expected that these two CPs would dimerize to form an equilibrium mixture of three 11-type species differing in the relative positions of their C60 and exTTF moieties, 11-X, 11-Y, and 11-Z (Scheme 2).
Scheme 2.
Synthesis of CPs 7–10 and structure of heterodimer 11 are shown. Also shown is the schematic representation of the three possible registers of corresponding dimers (X, Y, Z) (dashed box) and hydrogen-bonding between the side chains of the serine residues of 8-Z (solid box).
The γ-Acp-based CP 7 was synthesized starting from the cyclic hexapeptide 5a, which was obtained as described (14–16). Hydrogenation of 5a (balloon pressure, 10% Pd/C) yielded peptide 5b, and coupling the latter with phenyl-C61-butyric acid (PCBA) (6) (21) [O-(7-azabenzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HATU), diisopropylethylamine (DIEA), CDCl3/dimethylformamide (DMF)] afforded cyclo[l-Lys(PCBA)-d-MeN-γ-Acp-(l-Leu-d-MeN-γ-Acp)2−] (7). The location of the amide proton NMR signals of 7 in nonpolar solvents (8.0–8.3 ppm) showed the formation of stable homodimers. The fact that the 1H NMR spectrum remained unaltered when concentration was lowered implied a dimerization constant >104 M−1, similar to those of analogous α,γ-CPs. Like 11, homodimers of 7 can form three different dimeric species (7-X, 7-Y, and 7-Z), but the complexity of the 1H NMR spectrum prevented accurate determination of their relative proportion; in this regard, whereas peptide 5a exists as an almost equimolar mixture of 5a-X, 5a-Y, and 5a-Z, peptide 8 exists as a 1:1:1.8 mixture of 8-X, 8-Y, and 8-Z (see ref. 23). FTIR spectroscopy corroborated the β-sheet nature of the hydrogen-bond system mediating dimerization (22), showing bands at 1,625, 1,531, and 3,303 cm−1, similar to those reported for related PNs (5, 18–20).
To prepare the exTTF-bearing CP 10 we started from the γ-Ach-based CP cyclo[l-Ser-d-MeN-γ-Ach-(l-Phe-d-MeN-γ-Ach)2−] (8 in Scheme 2) (23). The esterification reaction between 8 and the exTTF derivative 9 (24) in the presence of N,N′-diisopropylcarbodiimide (DIC) and 4-dimethylaminopyridine (DMAP) afforded the desired α,γ-CP conjugate 10 in high yield (92%). The formation of three stable homodimers in CDCl3 was shown by the location of six doublets at 8.6–8.8 ppm and three doublets at 8.3–8.4 ppm in the 1H NMR spectrum, which also showed 10 to exist as practically equal proportions of the three possible forms, 10-X, 10-Y, and 10-Z.
Addition of 1 eq of 10 to a 2 mM solution of 7 in CDCl3 afforded all three forms of heterodimer 11; their amide proton NMR signals appear at 8.1–8.7 ppm [supporting information (SI) Fig. 5]. The 1H NMR spectra of similar mixtures of 7 and 10 were unchanged by dilution down to 2 × 10−4 M, heating to 323 K, or addition of methanol, which implies a high affinity for these dimeric complexes. Although the complexity of the 1H NMR spectrum of 11 prevented accurate determination of the relative proportions of 11-X, 11-Y, and 11-Z, it is thought that, just as homodimer 8-Z is formed to a greater extent than 8-X and 8-Y because of hydrogen bonding between favorably oriented serine hydroxy groups (Scheme 2), van der Waals interactions between the C60 and exTTF units of 11-Z could facilitate the formation of this isomer in preference to 11-X and 11-Y, and through-space electron transfer (25).
Fig. 1 shows fluorescence spectra of a solution of 7 in dichloromethane that were recorded after addition of increasing amounts of 10. The fullerene characteristic peak at 700 nm is quenched upon dimerization, and the dependence of fluorescence intensity on the concentration of 10 implies an association constant of at least 106 M−1.
Fig. 1.
Fluorescence spectral changes (λexc = 340 nm) of 7 (298 K, dichloromethane, 4.4 × 10−6 M) upon addition of different amounts of 10 (0, 0.44, 0.88, 1.32, 1.76, 2.20, 2.64, 3.08, 3.52, 4.40, 5.28, and 10.56 × 10−6 M). (Inset) Displayed is the relationship of I/I0 versus 10 that was used to determine the association constant.
The electrochemical properties of 7, 10, and 11 were investigated at room temperature by cyclic voltammetry and differential pulse voltammetry (SI Table 1 and SI Fig. 6). The redox behavior of the electroactive groups of homodimers of 7 and 10 (C60 and exTTF) is largely preserved in heterodimer 11 (16, 17, 26), but oxidative scans of 11 nevertheless show a shift of approximately +50 mV relative to those of the exTTF/exTTF2+ couple of homodimers of 10. This shift may be a consequence of a weak intramolecular electronic interaction between the exTTF moiety and the fullerene-based peptide in the ground state.
The above evidence of donor–acceptor interaction between the photo and redox active side chains of 11 (see Fig. 1) was corroborated by time-resolved transient absorption measurements in which the fullerene moiety was photoexcited at 387 nm by 150-fs laser pulses. Whereas the spectra of homodimers of 7 exhibited the characteristic transient fluorescence maximum at ≈900 nm (SI Fig. 7) correponding to the singlet excited state of fullerene, followed by a slow transition to a transient maximum at 720 nm corresponding to the triplet state, the spectra of 11 demonstrated that the fullerene singlet excited state decays very fast (i.e., 0.055 ns) and forms the radical ion pair state, C60•−/exTTF•+. The spectral characteristics observed at 680 and 1,040 nm resemble the fingerprints of the one-electron oxidized exTTF (i.e., exTTF•+) and the one-electron reduced fullerene (i.e., C60•−), respectively (Fig. 2) (27–30). On the time scale of up to 1,600 ps, no notable decay of the radical ion pair state features is observed, which rules out a fast intrahybrid charge recombination. In fact, charge separation caused by photoexcitation of exTTF in either heterodimer 11 or homodimers of 10 is negligible, and an efficient electron transfer is precluded by the short lifetime (≈0.86 ps) of the transient species with a spectral peak at 605 nm (SI Fig. 8) (31).
Fig. 2.
Femtosecond flash photolysis of heterodimer 11. (a) Differential absorption spectra (visible and NIR) obtained upon femtosecond flash photolysis (387 nm) of heterodimer 11 (≈5.0 μM) in nitrogen-saturated dichloromethane with several time delays between 0 and 200 ps at room temperature. Arrows indicate the features of the radical ion pair state. (b) Time-absorption profiles of the spectra shown in a at 660 nm, monitoring the intrahybrid charge separation process.
The above charge separation process is presumably limited to heterodimer 11-Z (Fig. 3), because the distances between C60 and exTTF in 11-X and 11-Y (>15 Å) are unlikely to power an intrahybrid electron transfer reaction. However, these latter isomers appear to undergo singlet-to-triplet processes on a time scale similar to that of homodimers of 7 (τ = 1.4 ns). This hypothesis is supported by fluorescence lifetime measurements of 11, which show an ≈1:2 ratio of two processes with lifetimes of <0.1 and ≈1.5 ns, respectively.
Fig. 3.
Computer-generated structure of heterodimer 11-Z, showing photo-induced electron transfer facilitated by the proximity of the donor and acceptor units (exTTF and C60, respectively).
The formation of two distinct photoproducts was corroborated by the results of nanosecond-scale experiments in which charge recombination was investigated after photoexcitation of heterodimer 11 at 355 nm. The transient spectra obtained in these experiments (Fig. 4a) show features at ≈360 and 720 nm that are attributable to the triplet excited state of fullerene, which presumably correspond to 11-X and 11-Y, and peak at 660 and 1,040 nm that are attributed to exTTF•+ and C60•−, respectively. The decay profiles of these spectra imply lifetimes of 25 μs for the triplet state (which decays to the singlet ground state) and 0.86 μs for the radical ion pair state.
Fig. 4.
Nanosecond flash photolysis of heterodimer 11. (a) Differential absorption spectrum (visible and NIR) obtained upon nanosecond flash photolysis (355 nm) of heterodimer 11 (≈5.0 μM) in nitrogen-saturated dichloromethane with a 50-ns time delay at room temperature. Arrows indicate the features of the radical ion pair state. (b) Time-absorption profiles of the spectrum shown in a at 1,040 nm, monitoring the charge recombination process.
In conclusion, we have prepared a bio-inspired nanohybrid, 11, in which a CP bearing an electron-donor unit (exTTF) is coupled by a β-sheet-like hydrogen-bond system to another bearing a photoactive electron-acceptor unit (C60). Photoexcitation of the fullerene chromophores to their 1.76-eV excited state is followed, in one of the three forms of 11, by a charge separation process generating a 1.15-eV radical ion pair state. On average, this state perdures for at least 1.5 μs, which is comparable to the performance of other noncovalent C60-based hybrids (in which electron transfer occurs basically via through-space interactions) (14, 15, 31–33) and is superior to that of typical covalent C60–exTTF conjugates (25–28). The structure of 11 in principle allows its extension to form a nanotubular battery along which electron donors and electron acceptors alternate and also suggests the possibility of designing molecular switches based on dynamic control of interconversion between the electroactive isomer 11-Z and the inactive isomers 11-X and 11-Y.
Detailed descriptions of the synthesis and characterization of key compounds, including NMR spectra (1H and 13C, NOESY and/or ROESY) and FTIR spectra of peptides 5a, 7, 10, and 11; electrochemical data for compounds 7, 10, and 11; and differential absorption spectra are available in NMR Spectra in SI Text.
Supplementary Material
Acknowledgments
This work was supported by the Ministry of Education and Science of Spain under Projects SAF2004-01044 (jointly with the European Regional Development Fund), CTQ2005-02609/BQU, and P-PPQ-000225-0505; the Xunta de Galicia under Project PGIDT04BTF209006PR; SFB583, DFG (60517/4-1), Fonds der Chemischen Industrie, and the Office of Basic Energy Sciences of the U.S. Department of Energy under Notre Dame Radiation Laboratory Project NDRL 4715; an Formación de Profesorado Universitario grant from the Ministry of Education and Science of Spain (to R.J.B.); and a Ramón y Cajal contract from the Ministry of Education and Science of Spain (to M.A.H.).
Abbreviations
- PN
peptide nanotube
- CP
cyclic peptide
- γ-Ach
γ-aminocyclohexanecarboxylic acid
- γ-Acp
γ-aminocyclopentanecarboxylic acid
- exTTF
2-[9-(1,3-dithiol-2-ylidene)anthracen-10(9H)-ylidene]-1,3-dithiole.
Footnotes
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
This article contains supporting information online at www.pnas.org/cgi/content/full/0609506104/DC1.
References
- 1.Ghadiri MR, Granja JR, Buehler LK. Nature. 1994;369:301–304. doi: 10.1038/369301a0. [DOI] [PubMed] [Google Scholar]
- 2.Brea RJ, Granja JR. In: Dekker Encyclopedia of Nanoscience and Nanotechnology. Schwarz JA, Contescu CI, Putyera K, editors. New York: Dekker; 2004. pp. 3439–3457. [Google Scholar]
- 3.Bong DT, Clark TD, Granja JR, Ghadiri MR. Angew Chem Int Ed. 2001;40:988–1011. [PubMed] [Google Scholar]
- 4.Balbo Block MA, Hecht S. Angew Chem Int Ed. 2005;44:6986–6989. doi: 10.1002/anie.200502436. [DOI] [PubMed] [Google Scholar]
- 5.Clark TD, Buriak JM, Kobayashi K, Isler MP, McRee DE, Ghadiri MR. J Am Chem Soc. 1998;120:8949–8962. [Google Scholar]
- 6.Martin CR, Kohli P. Nat Rev. 2003;2:29–37. doi: 10.1038/nrd988. [DOI] [PubMed] [Google Scholar]
- 7.Gao X, Matsui H. Adv Mater. 2005;17:2037–2050. doi: 10.1002/adma.200401849. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ghadiri MR, Granja JR, Buehler LK. Nature. 1994;369:301–304. doi: 10.1038/369301a0. [DOI] [PubMed] [Google Scholar]
- 9.Sánchez-Quesada J, Kim HS, Ghadiri MR. Angew Chem Int Ed. 2001;40:2503–2506. [PubMed] [Google Scholar]
- 10.Fernández-López S, Kim HS, Choi EC, Delgado M, Granja JR, Khasanov A, Kraehenbuehl K, Long G, Weinberger DA, Wilcoxen K, Ghadiri MR. Nature. 2001;412:452–455. doi: 10.1038/35086601. [DOI] [PubMed] [Google Scholar]
- 11.Vollmer MS, Clark TD, Steinem C, Ghadiri MR. Angew Chem Int Ed. 1999;38:1598–1601. doi: 10.1002/(SICI)1521-3773(19990601)38:11<1598::AID-ANIE1598>3.0.CO;2-J. [DOI] [PubMed] [Google Scholar]
- 12.Couet J, Samuel JDJS, Kopyshev A, Santer S, Biesalski M. Angew Chem Int Ed. 2005;44:3297–3301. doi: 10.1002/anie.200462993. [DOI] [PubMed] [Google Scholar]
- 13.Horne WS, Ashkenasy N, Ghadiri MR. Chem Eur J. 2005;11:1137–1144. doi: 10.1002/chem.200400923. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Guldi DM, Rahman GMA, Ehli C, Sgobba V. Chem Soc Rev. 2006;35:471–487. doi: 10.1039/b511541h. [DOI] [PubMed] [Google Scholar]
- 15.Martín N. Chem Commun. 2006:2093–2104. doi: 10.1039/b601582b. [DOI] [PubMed] [Google Scholar]
- 16.Yamashita Y, Kobayashi Y, Miyashi T. Angew Chem Int Ed Engl. 1989;28:1052–1053. [Google Scholar]
- 17.Bryce MR, Moore AJ, Hasan M, Ashwell GJ, Fraser AT, Clegg W, Hursthouse MB, Karaulov AI. Angew Chem Int Ed Engl. 1990;29:1450–1452. [Google Scholar]
- 18.Amorín M, Castedo L, Granja JR. J Am Chem Soc. 2003;125:2844–2845. doi: 10.1021/ja0296273. [DOI] [PubMed] [Google Scholar]
- 19.Amorín M, Castedo L, Granja JR. Chem Eur J. 2005;11:6543–6551. doi: 10.1002/chem.200500478. [DOI] [PubMed] [Google Scholar]
- 20.Brea RJ, Amorín M, Castedo L, Granja JR. Angew Chem Int Ed. 2005;44:5710–5713. doi: 10.1002/anie.200501555. [DOI] [PubMed] [Google Scholar]
- 21.Hummelen JC, Knight BW, Le Peq F, Wudl F, Yao J, Wilkins CL. J Org Chem. 1995;60:532–538. [Google Scholar]
- 22.Haris PI, Chapman D. Biopolymers (Peptide Sci) 1995;37:251–263. doi: 10.1002/bip.360370404. [DOI] [PubMed] [Google Scholar]
- 23.Amorín M, Villaverde V, Castedo L, Granja JR. J Drug Del Sci Tech. 2005;15:87–92. [Google Scholar]
- 24.Herranz MA, Martín N, Campidelli S, Prato M, Brehm G, Guldi DM. Angew Chem Int Ed. 2006;45:4478–4482. doi: 10.1002/anie.200504354. [DOI] [PubMed] [Google Scholar]
- 25.Pérez EM, Sánchez L, Fernández G, Martín N. J Am Chem Soc. 2006;128:7172–7173. doi: 10.1021/ja0621389. [DOI] [PubMed] [Google Scholar]
- 26.Echegoyen LE, Herranz MA, Echegoyen L. In: Encyclopedia of Electrochemistry. Bard AJ, Stratmann M, Scholz F, Pickett CJ, editors. Vol 7. Weinheim, Germany: Wiley; 2006. pp. 145–201. [Google Scholar]
- 27.Martín N, Sánchez L, Guldi DM. Chem Commun. 2000:113–114. [Google Scholar]
- 28.Díaz MC, Herranz MA, Illescas BM, Martín N, Godbert N, Bryce MR, Luo C, Swartz A, Anderson G, Guldi DM. J Org Chem. 2003;68:7711–7721. doi: 10.1021/jo034432e. [DOI] [PubMed] [Google Scholar]
- 29.Giacalone G, Segura JL, Martín N, Guldi DM. J Am Chem Soc. 2004;126:5340–5341. doi: 10.1021/ja0318333. [DOI] [PubMed] [Google Scholar]
- 30.Sánchez L, Sierra M, Martín N, Guldi DM, Wienk MM, Janssen RAJ. Org Lett. 2005;7:1691–1694. doi: 10.1021/ol050127z. [DOI] [PubMed] [Google Scholar]
- 31.Sánchez L, Sierra M, Martín N, Myles AJ, Dale TJ, Rebek J, Jr, Seitz W, Guldi DM. Angew Chem Int Ed. 2006;45:4637–4641. doi: 10.1002/anie.200601264. [DOI] [PubMed] [Google Scholar]
- 32.Sessler JL, Jayawickramarajah J, Gouloumis A, Torres T, Guldi DM, Maldonado S, Stevenson KJ. Chem Commun. 2005:1892–1894. doi: 10.1039/b418345b. [DOI] [PubMed] [Google Scholar]
- 33.McClenaghan ND, Grote Z, Darriet K, Zimine M, Williams RM, De Cola L, Bassani DM. Org Lett. 2005;7:807–810. doi: 10.1021/ol047527r. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.