DOI: 10.1002/chem.200901504
Sequence-Specific Self-Sorting of the Binding Sites of a Ditopic Guest by
Cucurbituril Homologues and Subsequent Formation of a
Hetero[4]pseudorotaxane
Gizem Celtek,
[a]Mge Artar,
[a]Oren A. Scherman,
[b]and Dçns Tuncel*
[a]Biomimetic systems are of great importance because they resemble and offer models to more complex reactions in-volved in biological chemistry.[1] Although self-sorting by
recognition of self from nonself is a common phenomenon in biological systems, the examples are scarce in the synthet-ic systems.[2]Among many other synthetic receptors,
calixar-enes, cyclodextrins and more lately cucurbit[n]uril have been widely employed as biomimetic receptors; the host– guest chemistry and molecular recognition phenomena cou-pled by high-fidelity self-sorting involved with these hosts have been studied extensively.[3, 4]Self-sorting has also been
utilized in the construction of a hetero[3]rotaxane, which is based on two similar crown ethers and a diammonium alkyl axle containing two binding sites.[5]
Cucurbit[n]uril (CB[n], n = 5–10), a family of macrocycles synthesized by acid catalyzed condensation of glycoluril with formaldehyde has a hydrophobic cavity and two identi-cal carbonyl containing portals.[6] They have been used in
the synthesis of a wide variety of supramolecular assemblies including molecular machines and switches.[7]
In this paper, a guest containing two distinct binding sites, which differ chemically and geometrically from each other, was prepared through a click reaction. The selectivity and recognition behavior of cucurbit[n]uril homologues (n = 6,7,8) toward each of the binding sites of these guests were
studied. These binding sites are comprised of a flexible and hydrophobic dodecyl spacer and a five-membered triazole ring terminated with ammonium ions. The formation of 1:2, 1:1 and 1:1 complexes between the guest and CB homo-logues, namely CB6, 7, and 8 respectively have been con-firmed by 1H NMR spectroscopy and mass spectrometry.
Moreover, it was shown that CB6, CB7 and CB8 have the ability to recognize and self-sort the recognition sites ac-cording to their chemical nature, size and shape. The forma-tion of hetero[4]pseudorotaxane containing both CB6 and CB8 through sequence-specific self-sorting was also ob-served (Scheme 1).
In our previous paper, Axle A was synthesized in a CB6-catalyzed [3 + 2] addition, as a part of the [3]pseudoro-taxane.[8] However, here an alternative method, which
en-sures that the Axle A is free from CB6, was utilized and ac-cordingly Axle A was prepared through CuI-catalyzed
click-chemistry[9]between diazide and propargylammonium
chlo-ride monomers according to the reaction depicted in Scheme 2 in almost quantitative yield. The axle was charac-terized by1H,13C NMR, COSY and elemental analysis (see
the Supporting Information).
In this axle, there are two recognition sites; the dodecyl alkyl chain and the two diammonium triazoles. The com-plexation behavior of CB6, CB7 and CB8 with Axle A was first investigated by 1H NMR titration then the resulting
pseudorotaxanes were isolated and characterized by 1D, 2D-NMR experiments and MALDI-MS.
CB6 prefers to bind the diaminotriazole sites rather than the dodecyl spacer as previously reported.[8] The driving
force for this type of complexation is mainly due to size se-lectivity as triazole has an appropriate size to fill the cavity of the CB6 as well as ion-dipole interactions between am-monium ions and the oxygen atoms of the carbonyl portals. Figure 1 b shows the 1H NMR spectrum of
[3]pseudoro-taxane of CB6 with Axle A. It is evident that triazole units are threaded by CB6 s and as a result the signal from the tri-azole proton is shifted 1.7 ppm upfield (from d = 8.2 to 6.5 ppm); however, there are not significant changes in the
[a] G. Celtek, M. Artar, Dr. D. Tuncel
Chemistry Department and Institute of Materials Science and Nanotechnology Bilkent University 06800 Ankara (Turkey) Fax: (+ 90) 3122664068 E-mail: dtuncel@fen.bilkent.edu.tr [b] Dr. O. A. Scherman
Melville Laboratory for Polymer Synthesis Chemistry Department
University of Cambridge Cambridge, CB2 1EW (UK)
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.200901504.
2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Chem. Eur. J. 2009, 15, 10360 – 10363
signals for dodecyl protons indicating this part is not encap-sulated by CB6.
Host–guest chemistry and recognition properties of CB8 toward the same guest were also investigated by1H NMR
ti-tration (see the Supporting Information). The axle was dis-solved in D2O and 0.5 equiv increments of CB8 up to
2.1 equiv were added as solid. Although CB8 is not soluble in water, it dissolves in the aqueous solution of Axle A. Just after each addition and dissolving CB8, a1H NMR spectrum
was recorded at room temperature. The 1H NMR spectrum
after addition of 0.5 equiv of CB8 shows two sets of signals, assigned to encapsulated and unencapsulated protons of the guest. This indicates that the exchange of complexed and uncomplexed guest is slow on the NMR time scale. There is a significant upfield shift in the chemical shifts of all dodecyl protons. The signal from the triazole proton is shifted 0.1 ppm downfield. This clearly indicates that CB8 encapsu-lates dodecyl protons but not the triazole which is in con-trast to CB6. Addition of 1 equiv of CB8 causes the appear-ance of only one set of protons and further addition of CB8 produced similar spectra. Accordingly, a [2]pseudorotaxane of CB8 with Axle A results from a 1:1 inclusion complex of CB8 with the guest; it was isolated and characterized by 1D, 2D NMR experiments and MALDI-MS. The1H NMR
spec-trum of the [2]pseudorotaxane of CB8 is shown in Figure 1 c. The signals assigned to the dodecyl protons are shifted up-field, whereas the signal from the triazole proton appears at d = 8.2 ppm. 2D NMR experiments (COSY and ROESY) (see the Supporting Information) provide structural infor-mation and based on the analysis these we can suggest that the dodecyl spacer is folded to fit inside the cavity of CB8 to maximize the hydrophobic interactions in an efficient manner as shown in the Scheme 1. The driving forces for the formation of the suggested structure are hydrophobic ef-fects and ion-dipole interactions between the ammonium
Scheme 1. Suggested structures after the formation of complexes of CB homologues with Axle A.
Scheme 2. Synthesis of Axle A (counterion chlorides are omitted for clarity).
Figure 1. The 1H NMR spectra (400 MHz, 25 8C, D
2O) of a) Axle A;
b) Axle A + 2 CB6; c) Axle A + CB7; d) Axle A + 1 CB8 (the peak at d = 2.2 ppm is assigned to acetone).
Chem. Eur. J. 2009, 15, 10360 – 10363 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemeurj.org 10361
ions of triazole units and possibly triazole proton-carbonyl interactions. Related conformational changes of alkanes in-duced by hosts have previously been reported by Rebek and co-workers; they have shown that n-alkanes exhibit helical conformations inside self-assembled capsules to maximize noncovalent interactions between host and guest.[10]
Recent-ly Kim and co-workers have also reported the encapsulation of a long aminoalkane by CB8 showing that the aminodode-cane was folded into a U-shape in order to fit inside the CB8 cavity.[11] The MALDI-tof-MS spectrum of Axle A +
CB8 complex also shows a molecular ion in agreement with the suggested structure in Scheme 1 (see the Supporting In-formation for its MALDI-tof-MS).
1H NMR titration of Axle A with an increasing
concentra-tion of CB7 yields similar results obtained from the CB8 ti-tration experiments. Again CB7 forms a 1:1 complex with Axle A and its1H NMR spectrum (Figure 1 c) indicates that
triazoles are not encapsulated by CB7 but that the central methylene protons of the dodecyl spacer were encapsulated. Apparently, the conformation of the dodecyl spacer inside the cavity of CB7 is different than in the case of CB8 be-cause the cavity of CB7 has a smaller volume. The result in-dicates that due to space restrictions, not all protons are deeply buried in the cavity of CB7 and therefore they are not as shielded as the central methylene protons. Its MALDI-tof mass spectrum shows molecular ions from both a 1:1 complex as well as a 2:1 complex (CB7: Axle A).
Although both CB7 and CB8 selectively bind to the dode-cyl recognition site of the guest, the conformation of the hy-drophobic alkyl chain is different. In CB8, the alkyl chain is folded in such a way that all of it is deeply buried in the cavity and shielded, whereas in the case of CB7, the entire dodecyl chain can not be accommodated by the cavity. On the other hand, CB6 behaves totally differently than CB7 and CB8 toward the guest; it prefers to bind the triazole site rather than the dodecyl chain.
Next, we wondered what would happen when both CB6 and CB8 are available to bind with Axle A. First, 2 equiv of CB6 were added to the aqueous solution of [2]pseudoro-taxane-CB8 and the progress of the reaction was followed by recording1H NMR spectra at regular intervals. Figure 2
shows the spectra of mixture recorded after 10 min., 2 h, 24 h and 96 h. These spectra reveal that the triazole proton signal at d = 8.1 ppm disappeared and new peaks appeared at d = 6.47 and 6.49 ppm for encapsulated triazole protons. Additionally, new peaks in between 1.1–1.8 ppm for dodecyl protons are observed. These results imply that CB8 remains associated with the dodecyl chain, but it has adopted a dif-ferent conformation whereas CB6 prefers to bind the di-ACHTUNGTRENNUNGammonium triazoles. After carefully analyzing the spectra taken over time, we came to a conclusion that when three components (Axle A + 1 CB8 + 2 CB6) are mixed together, initially hetero[4]pseudorotaxane forms as a major complex. However, after a certain time the reaction reaches an equi-librium resulting in hetero[4]pseudorotaxane minor and [3]pseudorotaxane major complexes. As can be seen in the
1H NMR spectra (Figure 2b–d) there are two distinct sets of
signals for the protons of hetero[4]pseudorotaxane and [3]pseudorotaxane.
This was also supported by 2D COSY NMR spectrum (Supporting Information, Figure S11).The signals have been assigned on the spectra accordingly (Figure 2b–d) and the ratio of hetero[4]pseudorotaxane to [3]pseudorotaxane has been calculated from the1H NMR spectra by comparing the
integration of dodecyl protons of hetero[4]pseudorotaxane and [3]pseudorotaxane in between d = 0.4–1.1 ppm and d = 1.1–1.8 ppm, respectively. Accordingly, the ratios of het-ACHTUNGTRENNUNGero[4]pseudorotaxane to [3]pseudorotaxane have been esti-mated as 85:15, 75:25, 45:55 and 30:70 for after 10 min, 2 h, 24 h, and 96 h, respectively. These results were also con-firmed by comparing the integration of triazole peaks at d = 6.47 and 6.49 ppm for hetero[4]pseudorotaxane and [3]pseu-dorotaxane respectively. We observed that the ratio of het-ACHTUNGTRENNUNGero[4]pseudorotaxane to [3]pseudorotaxane in the mixture remains as 30:70 even after 2 weeks. These results indicate that the entrapment of CB8 in the formation of hetero[4]-pseudorotaxane is kinetically driven and over time the ther-modynamically stable [3]pseudorotaxane forms as a major complex.
Second, we mixed together Axle A, 2 equivalents of CB6 and 1 equiv of CB8 in D2O at the same time. The
1H NMR
spectrum recorded after ten min. mixing reveals almost the same result as the previous experiment. Last, excess CB8 was added to the aqueous solution of [3]pseudorotaxane-CB6. Its 1H NMR spectrum was recorded at different time
intervals and even after two weeks the spectrum is the same as the spectrum of [3]seudorotaxane-CB6. This suggests that
Figure 2. The 1H NMR spectra (400 MHz, 25 8C, D
2O) of a) Axle A +
CB8; b) 10 min later after addition of 2 equiv CB6 to (a); c) 2 h later after addition of 2 equiv CB6 to (a); d) 24 h later after addition of 2 equiv CB6 to (a); e) 96 h later after addition of 2 equiv CB6 to (a). Rectangle and sphere denote protons from hetero[4]pseudorotaxane and [3]pseudorotaxane respectively; the peak at d = 3.25 ppm is assigned to MeOH residue.
www.chemeurj.org 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Chem. Eur. J. 2009, 15, 10360 – 10363
10362
CB6 behaves as a blocking group and prevents CB8 from being threaded on to the dodecyl spacer.
After 24 h stirring the three components (Axle A + 1 CB8 + 2 CB6) at RT, the solvent was removed under re-duced pressure and the isolated solid was characterized by MALDI-tof-MS. The spectrum shows peaks for molecular ions of [3]pseudorotaxane (Axle A + 2 CB6) and hetero[4]-pseudorotaxane at m/z = 2443 and 3772, respectively. Other peaks for Axle A + 1 CB8 + 1 CB6, Axle A + 2 CB8, and Axle A + 3 CB6 are at m/z = 2775, 3108, and 3439, respec-tively (Supporting Information, Figure S10).
In conclusion, a guest containing two distinct binding sites which differ chemically and geometrically from each other was prepared through the click reaction; the selectivity and recognition behavior of cucurbit[n]uril homologues (n = 6,7,8) towards each binding site were studied. The formation of 1:2, 1:1 and 1:1 complexes between the guest and CB6, CB7, and CB8, respectively, have been confirmed by
1H NMR spectroscopy and mass spectrometry. Moreover, it
was demonstrated that the CB6, CB7 and CB8 had the abili-ty to recognize and self-sort the recognition sites according to their chemical nature, size and shape. The formation of a hetero[4]pseudorotaxane containing CB6 and CB8 through sequence-specific self-sorting was also observed; its forma-tion could be controlled kinetically and by the order of com-ponent addition.
Experimental Section
Synthesis of N,N’-bis-[1-(2-aminoethyl)-1H-[1,2,3]triazol-4-ylmethyl]do-decane-1,12-diamine: N,N’-Bis-(2-azidoethyl)dodecane-1,12-diamine· 2 HCl (150 mg, 0.36 mmol), propargylamine·HCl (83 mg, 0.91 mmol) were dissolved in an ethanol-water mixture (5 mL, 2:3, v/v). Subsequently CuSO4M. > 5 H20 (4.5 mg, 0.015 mmol) and sodium ascorbate (7.5 mg,
0.036 mmol) were added. After stirring at RT overnight, the mixture was poured into water (5 mL) and 1 m NaOH solution (2 mL) was added. It was stirred for 30 min and extracted with chloroform. Organic phase was collected and the solvent was removed under reduced pressure. The re-maining solid residue was suspended in water and stirred further 30 min. White precipitates were collected by filtration and dried under high vacuum 3 h. (170 mg, 92 %). M.p. 122–123 8C; 1H NMR (400 MHz, CDCl3, 25 8C): d = 1.18 (m, 8 H; F,G,H), 1.38 (m, 2 H; E), 1.51 (m, 6 H; -NH), 2.54 (t,3J HH=7.2 Hz, 2 H; D), 3.03 (t,3JHH=5.9 Hz, 2 H; A), 3.92 (s, 2 H; C), 4.37 (t,3J HH=5.9 Hz, 2 H; B), 7.47 ppm (s, 1 H; T);13C NMR (100 MHz, D2O, 25 8C): d = 27.4, 29.6, 29.7, 30.2, 37.7, 49.6, 49.7, 50.4,
122.8 (triazole, =CH), 148.8 ppm (triazole, =CR); elemental analysis calcd (%) for C22H44N10C, 58,90; H, 9,89; N, 31,22; found: C, 58,63; H,
10.01; N, 31.07. Hydrochloride salt of the ditriazole Axle A compound was prepared by dissolving it in diluted HCl solution. After stirring over-night at RT, the solvent was removed under reduced pressure and the solid residue was washed with THF. It was dried under high vacuum 5 h. (180 mg, 95 %). 1H NMR (400 MHz, CDCl 3, 25 8C): d = 1.30 (m, 8 H; F,G,H), 1.65 (m, 2 H; E), 3.05 (t, 3J HH=7.2 Hz, 2 H; D), 3.56 (t,3JHH= 5.9 Hz, 2 H; A), 4.35 (s, 2 H; C), 4.85 (t,3J HH=5.9 Hz, 2 H; B), 8.12 (s, 1 H; T).
Acknowledgements
This work was generously supported by the Royal Society of Chemistry, T.W.J. Traveling Fellowship and EU-UNAM-REGPOT Grant No:203953.
Keywords: cucurbituril · rotaxanes · recognition · self-sorting · supramolecular chemistry
[1] Y. Murakami, J. Kikuchi, Y. Hisaeda, O. Hayashida, Chem. Rev. 1996, 96, 721 – 758.
[2] a) J.-M. Lehn, Science 2002, 295, 2400 – 2403; b) J.-M. Lehn, Chem. Eur. J. 2000, 6, 2097 – 2102.
[3] a) A. W. L. Isaacs, J. Am. Chem. Soc. 2003, 125, 4831 – 4835; b) S. Liu, C. Ruspic, P. Mukhopadhyay, S. Chakrabarti, P. Y. Zavalij, L. Isaacs, J. Am. Chem. Soc. 2005, 127, 15959 – 15967; c) P. Mukhopad-hyay, P. Y. Zavalij, L. Isaacs, J. Am. Chem. Soc. 2006, 128, 14093 – 14102; d) M. V. Rekharsky, H. Yamamura, Y. H. Ko, N. Selvapalam, K. Kim, Y. Inoue, Chem. Commun. 2008, 2236 – 2238; e) W.-H. Huang, P. Y. Zavalij, L. Isaacs, Org. Lett. 2008, 10, 2577 – 2580. [4] a) T. Ooya, D. Inoue, H. S. Choi, Y. Kobayashi, S. Loethen, D. H.
Thompson, Y. H. Ko, K. Kim, N. Yui, Org. Lett. 2006, 8, 3159; b) Y. Liu, X.-Y. Li, H.-Y. Zhang, C.-J. Li, F. Ding, J. Org. Chem. 2007, 72, 3640.
[5] W. Jiang, H. D. F. Winkler, C. A. Schalley, J. Am. Chem. Soc. 2008, 130, 13852 – 13853.
[6] a) W. L. Mock in Comprehensive Supramolecular Chemistry, Vol. 2 (Ed.: F. Vogtle), Pergamon Press, Oxford, 1996, pp. 477 – 493; b) I.-S. Kim, J. Jung, S.-Y. Kim, E. Lee, J.-K. Kang, S. Sakamoto, K. Yama-guchi, K. Kim, J. Am. Chem. Soc. 2000, 122, 540 – 541; c) J. Lagona, P. Mukhopadhyay, S. Chakrabarti, L. Isaacs, Angew. Chem. 2005, 117, 4922 – 4949; Angew. Chem. Int. Ed. 2005, 44, 4844 – 4870; d) K. Kim, N. Selvapalam, Y. H. Ko, K. M. Park, D. Kim, Chem. Soc. Rev. 2007, 36, 267 – 279.
[7] a) V. Sindelar, S. Silvi, S. E. Parker, D. Sobransingh, A. E. Kaifer, Adv. Funct. Mater. 2007, 17, 694 – 701; b) J. W. Lee, I. Hwang, W. S. Jeon, Y. H. Ko, S. Sakamoto, K. Yamaguchi, K. Kim, Chem. Asian J. 2008, 3, 1277 – 1283; c) D. Tuncel, N. Cindir, U. Koldemir, J. Inclu-sion Phenom. Macrocyclic Chem. 2006, 55, 373 – 380; d) D. Tuncel, H. B. Tiftik, B. Salih, J. Mater. Chem. 2006, 16, 3291 – 3296; e) D. Tuncel, O. Ozsar, H. B. Tiftik, B. Salih, Chem. Commun. 2007, 1369 – 1371; f) U. Rauwald, O. A. Scherman, Angew. Chem. 2008, 120, 4014 – 4017; Angew. Chem. Int. Ed. 2008, 47, 3950 – 3953. [8] D. Tuncel, M. Katterle, Chem. Eur. J. 2008, 14, 4110 – 4116. [9] H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001, 113,
1198; Angew. Chem. Int. Ed. 2001, 40, 1162. [10] J. Rebek, Jr., Chem. Commun. 2007, 2777 – 2789.
[11] Y. H. Ko, H. Kim, Y. Kim, K. Kim, Angew. Chem. 2008, 120, 4174 – 4177; Angew. Chem. Int. Ed. 2008, 47, 4106 – 4109.
Received: June 3, 2009 Published online: September 8, 2009
Chem. Eur. J. 2009, 15, 10360 – 10363 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemeurj.org 10363
