Cationic cyclophanes-in-cucurbit[10]uril: host-in-host complexes showing cooperative recognition towards neutral phenol guests

ABSTRACT Cucurbit[10]uril (CB[10]) is found to form stable host-guest complexes with tetracationic cyclophanes in H2O and DMSO. Upon encapsulation in CB[10], the affinity of cyclophanes towards anions is weakened, due to electronegative carbonyl oxygens on CB[10] providing charge repulsion environment, and encapsulation fixing the conformation of cyclophanes. Interestingly, these complexes show positively cooperative effects towards neutral phenol guests to form heteroternary complexes. The driving force for the formation of heteroternary inclusion complexes is multiple hydrogen bonding interactions confirmed by X-ray single crystal analysis, as well as hydrophobic effects. GRAPHICAL ABSTRACT

Cucurbit[n]urils (CB[n]s, n = 5-8, 10) are a kind of macrocyclic hosts and have played important roles in host-guest chemistry [20,21]. CB [10] (Figure 1) has the largest cavity in the CB[n] family with a D nh symmetry [21], showing unique molecular recognition properties towards guests, especially for large size molecules [22].
Herein, we report on the host-guest complexation between CB [10] and tetracationic cyclophanes ( Figure  1) by 1 H NMR, single-crystal diffraction, and ESI-MS. Our results show that these host-in-host complexes are soluble and stable in H 2 O and DMSO. Once encapsulated in CB [10], cyclophanes show a weakened binding affinity towards anions in DMSO. Interestingly, the binary complexes can include neutral phenol molecules to form heteroternary complexes with positively cooperative effects (cyclophanes or CB [10] could not bind phenols). dissolution is often observed once it forms inclusion complexes with water-soluble guests. The host-guest complexation between CB [10] and cyclophanes in H 2 O and DMSO was investigated in detail by 1 H NMR ( Figure  2, Figure S13-S27). As shown in Figure 2b, proton signals of both free and bound 2 are observed, indicating that the binding between CB [10] and 2 exhibits slow exchange on the 1 H NMR time scale. The binding ratio between CB[10] and 2 is calculated as 1:1 according to the 1 H NMR peak integration (Figure 2c). From the 1 H NMR spectra (Figure 2), the signals for all protons of 2 undergo upfield shifts, suggesting that the whole guest molecule is encapsulated by CB [10]. The change of chemical shift of aromatic protons H 6 is the largest (Δδ H6 = -0.626 ppm, Table S1), suggesting that the benzene moieties are located inside the cavity of CB [10], while the linking imidazole units remain at the CB [10] portal areas. Due to the exchange between active hydrogen and deuterium in D 2 O (H/D exchange), the signal for protons H 5 is invisible [30,31]. In addition, we performed the 1 H  NMR dilution experiments (1.0 mM to 0.01 mM, Figure  S14) and did not observe any signals of free 2, suggesting that the binding constants K a for CB [10]⋅2 are larger than 10 7 M −1 [27]. CB[n]s alone are poorly soluble in anhydrous organic solvents [32], while their complexes sometimes are soluble in DMSO or CH 3 CN [33,34]. With the addition of 2⋅4PF 6 into CB [10]/DMSO suspension, the dissolution of CB [10] happened. From the 1 H NMR spectra ( Figure S15), almost all proton signals of 2 undergo upfield shifts, except for H 3 on imidazole. Moreover, both free and bound proton signals can be seen ( Figure S15b), indicating that the binding between CB [10] and 2 in DMSO is also slow exchange on the 1 H NMR time scale. The change of the chemical shift of aromatic protons H 6 is the largest (Δδ H6 = −0.82 ppm), and the change of the chemical shift of imidazole H 3 is the least (Δδ H3 = 0.01 ppm, Table S1). In DMSO, the signal for proton H 5 is visible, and Δδ H5 is −0.52 ppm. The 1 H NMR peak integration ( Figure S15c) suggests the formation of a 1:1 inclusion complex CB [10]⋅2 in DMSO. Moreover, 1 H NMR dilution experiment (1.0 mM to 0.01 mM, Figure  S17) shows that signals of free 2 are not observed in 0.1 mM but can be seen in 0.01 mM. The main reason is that the lack of hydrophobic effect in DMSO leads to the relatively weaker binding between CB[10] and 2. It can be concluded that the binding constants K a of CB [10]⋅2 should be larger in H 2 O than that in DMSO.
Like 2, the encapsulation of 1 and 3 in CB [10] in D 2 O also exhibits slow exchange kinetics on the 1 H NMR time scale ( Figure S18 and Figure S24). By the 1 H NMR peak integration, CB [10]⋅1 and CB [10]⋅3 binary complexes are formed with minimum K a as 10 7 M −1 ( Figure S20 and S26). Similarly, in DMSO, the binding of 1 with CB [10] also shows slow exchange kinetics ( Figure S21), with the formation of a 1:1 inclusion complex CB[10]⋅1. Unlike 1 and 2, the binding of 3 with CB [10] shows fast exchange kinetics ( Figure S27).
Single crystals of CB [10]⋅2 in H 2 O were obtained by the evaporation at room temperature ( Figure 3, Table  S2) [35]. The crystal structure directly shows that 2 is encapsulated by the benzene rings located inside the cavity of CB [10] (Figure 3). The dihedral angle between the included phenyl plane of 2 and the equatorial plane of CB [10] is 90° ( Figure S29), indicating that 2 is inserted vertically inside the cavity of CB [10]. The multiple C-H . . . O = C hydrogen bonding interactions are observed ( Figure S29). The shortest distance between the portal oxygen atom on CB [10] and the hydrogen atom on 2 is 2.505 Å. These data indicate that the driving forces for the formation of the CB [10]⋅2 inclusion complex are multiple-hydrogen bonding interactions and hydrophobic effects. Although the linking imidazole units in the crystal are near outside the cavity of CB [10], slightly in disagreement with the conclusion based on the 1 H NMR data, it is reasonable in view of the dynamic rotation of imidazoles around the portals of CB [10].
UV-Vis and fluorescence spectroscopy were also employed to investigate the complexation of CB [10] with the cationic cyclophanes. As demonstrated in Figure S30 and S31, the absorption of all cyclophanes decreases and the fluorescence intensity of 2 or 3 changes a lot with the addition of CB [10] to aqueous solutions of guests. Generally speaking, both UV-Vis and fluorescence titration experiments clearly demonstrate the complexation of 1, 2, and 3 with CB [10].
The molecular recognition properties of 1 as an anion receptor towards anions in DMSO and CH 3 CN were reported [29]. The (C-H) + ···A − hydrogen-bonding interactions are the main driving force for the effective binding of cyclophane with anionic species. So we investigated the molecular recognition properties of cyclophanes in the absence/presence of CB [10] towards anions by 1 H NMR spectroscopy ( Figure S32-S47, Table S3). With the gradual addition of Br − (tetrabutylammonium salt as an anion source) to CB [10]⋅1 in DMSO, the signal for the proton H 5 of (C-H) + in the imidazolium moiety undergoes an upfield shift ( Figure S46a). The binding constants are measured to be K 1 = (215 ± 5) M −1 , K 2 = (10 ± 1) M −1 by nonlinear least-squares curve fitting ( Figure S46b), with the interaction parameter χ as 0.02. We believe that both electronegative carbonyl oxygens providing charge repulsion environment and encapsulation fixing the conformation of cyclophanes should be responsible for the weakened affinity of cyclophanes towards anions.
The cyclophanes themselves could not bind neutral phenol guests in H 2 O ( Figure S48-S51), but interestingly, we found CB [10]•cyclophane complexes could ( Figure  S52-S56). The schematic illustration shows the formation of ternary complexes in the presence of CB [10] with positive cooperative effects (Figure 4a, Figure S52), owing to the cooperativity factor α ≫1. For example, with gradual addition of 6 to the CB [10]•2 solution, the signals for proton H 6 on aromatic rings and H 4 on methylene of 2 undergo upfield shifts. In contrast, the signals for proton H 2 /H 3 on imidazole rings and H 1 on methylene undergo downfield shifts (Figure 4b). The binding between 6 and CB [10]•2 exhibits fast exchange on the 1 H NMR time scale. At the beginning of titration, the signals for protons of 6 are too broad to be observed; with more 6, the signals for protons of 6 appear and undergo downfield shifts. These experimental phenomena indicate that 6 is encapsulated into the cavity of CB [10]•2. Applying a non-linear equation to fit the experimental data, the binding constant K a value is measured to be (1.70 ± 0.11) × 10 4 M −1 (Figure 4c, Table 1). The ion peak at m/z 785.57 is detected, corresponding to 1:1:1 complex CB [10]•2•6 ([CB[10] + 2 + 6 + Br − ] 3+ = 785.57) ( Figure S57). Additional neutral phenol guests (4, 5, 7) are selected to test the binding affinity of CB [10] ⋅Cyclophane. The binding constants K a (M −1 ) are shown in Table 1. Among the complexes, CB[10]⋅1 exhibits higher binding affinity than other complexes, suggesting that the cavity size of CB [10]⋅1 is more suitable for those guest molecules. It could be also seen that CB[10]⋅1 prefers to bind 6 rather than other neutral phenol guests, implying a better size match between CB [10]⋅1 and 6. Meanwhile, the 1 H NMR results show that CB[10]•2 could not bind to 4 in DMSO ( Figure S58), suggesting that the main driving force is the hydrophobic effect for the formation of ternary complexes in water.
The crystal structure of CB [10]⋅2⋅6 directly shows the formation of the ternary complex CB[10]⋅2⋅6 ( Figure 5, Table S4) [36]. Compared to the structure of CB [10] Figure S59-S60). The driving force for the formation of CB [10]⋅2⋅6 ternary complex is multiple-hydrogen bonding interactions and hydrophobic effects. This is the first reported single crystal structure of the heteroternary complex in CB [10] chemistry.

Conclusion
In summary, CB [10] has been found to form stable hostguest inclusion complexes with cationic cyclophanes by 1 H NMR spectroscopy and single-crystal diffraction. The recognition properties of cyclophanes are deeply affected in the presence of CB [10]: namely, the ability to bind anions is weakened in DMSO. In addition, the host-in-host complexes show positive cooperative effects towards neutral phenol guests in H 2 O to form heteroternary inclusion complexes driven by multiple hydrogen bonding interactions and hydrophobic effects. This study demonstrates that the host-in-host complex shows intrinsic cooperative effects, which may have potential applications in constructing supramolecular materials and designing molecular machines.

General information
All tetrabutylammonium anions, neutral phenol compounds, and other compounds used in this study were purchased from commercial suppliers and were used without further purification. DMSO-d 6 was purchased from Cambridge Isotope Laboratories, Inc., and not dried over before use. CB [10] was prepared by the literature procedure [37]. NMR spectra were measured with spectrometers operating on Bruker Avance 500 MHz and Agilent 600 MHz DD2. Absorption measurement was performed in quartz cuvettes (1 cm x 1 cm x 4.5 cm). The UV-Vis absorption spectra were recorded on a Shimadzu UV-3600 spectrophotometer. Fluorescence spectra were obtained on a PerkinElmer LS 55 Fluorescence spectrophotometer. Organic elemental analysis was recorded on the elementary vario el iii spectrometer. Electrospray-ionisation high-resolution mass spectrometry (ESI-HRMS) experiments were  conducted on an applied Q EXACTIVE mass spectrometry system. ESI-MS were performed on a Solarix 9.4 T. The X-ray data collections were measured on a Bruker SMART APEX-II CCD system equipped with a graphite monochromator and a MoKα sealed tube (λ = 0.71073 Å) (Data collection temperature was 173 K). All measurements were performed at room temperature.

General procedure for the preparation of cationic cyclophanes
Synthesis of 1 [29]: To a solution of 1-(1H-imidazol-1ylmethyl)-1H-imidazole [38] (2.7 mmol, 401.5 mg) in dry CH 3 CN (120 ml), 2,6-bis(bromomethyl)-pyridine (2.7 mmol, 717.6 mg) in dry CH 3 CN (600 ml) was added dropwise over a period of 0.5 h. The mixture was heated under reflux overnight. After cooling to room temperature, the solvent was evaporated in vacuo. The residue was purified by column chromatography (Eluent: MeOH: H 2 O:saturated NH 4 Cl = 6:3:1, R f = 0.33) to afford a solid residue 1⋅4Br as its tetrahalide after evaporation of the solvents under vacuum. This crude product was dissolved in the water, and a concentrated aqueous solution of NH 4 PF 6 was added until no further precipitation was observed under ultrasonic conditions. The precipitate was centrifuged and washed with water several times, drying at high vacuum to give 1⋅4PF 6 . The product 1⋅4PF 6 was dissolved in anhydrous CH 3 CN and a concentrated acetonitrile solution of tetraethyl ammonium bromide (TEAB) was added until no further precipitation was observed under ultrasonic conditions. Then, the precipitate was centrifuged and washed with anhydrous CH 3 CN several times, drying at high vacuum to give product 1⋅4Br 467.0 mg, the yield was 32%. 1⋅4Br: 1

Synthesis of 2:
To a solution of 1-(1H-imidazol-1ylmethyl)-1H-imidazole (1 mmol, 148 mg) in dry CH 3 CN (40 ml), α,α'-Dibromo-p-xylene (1 mmol, 264 mg) in dry CH 3 CN (20 ml) was added dropwise over a period of 0.5 h. The mixture was heated under reflux overnight. After cooling to room temperature, the solvent was evaporated in vacuo. The residue was purified by column chromatography (eluent: MeOH:saturated NH 4 AcO: MeNO 2 = 10:9:5, R f = 0.18) to afford a solid residue 2⋅4Br as its tetrahalide after evaporation of the solvents under vacuum. This crude product was dissolved in the water and a concentrated aqueous solution of NH 4 PF 6 was added until no further precipitation was observed under ultrasonic conditions. The precipitate was centrifuged and washed with water several times, drying at high vacuum to give 2⋅4PF 6 . The product 2⋅4PF 6 was dissolved in anhydrous CH 3 CN and a concentrated acetonitrile solution of tetraethyl ammonium bromide (TEAB) was added until no further precipitation was observed under ultrasonic conditions. Then, the precipitate was centrifuged and washed with anhydrous CH 3 CN several times, drying at high vacuum to give product 2⋅4Br 80. 8

Synthesis of 3:
To a solution of 1-(1H-imidazol-1ylmethyl)-1H-imidazole (2.94 mmol, 436 mg) in dry CH 3 CN (120 ml), 4,4'-Bis(bromomethyl)biphenyl (2.94 mmol, 1 g) in dry CH 3 CN (60 ml) was added dropwise over a period of 0.5 h. The mixture was heated under reflux overnight. After cooling to room temperature, the solvent was evaporated in vacuo. The residue was purified by column chromatography (Eluent: MeOH: saturated NH 4 AcO: MeNO 3 = 10: 9: 1, R f = 0.22) to afford a solid residue 3⋅4Br as its tetrahalide after evaporation of the solvents under vacuum. This crude product was dissolved in the water and a concentrated aqueous solution of NH 4 PF 6 was added until no further precipitation was observed under ultrasonic conditions. The precipitate was centrifuged and washed with water several times, drying at high vacuum to give 3⋅4PF 6 . The product 3⋅4PF 6 was dissolved in anhydrous CH 3 CN and a concentrated acetonitrile solution of tetraethyl ammonium bromide (TEAB) was added until no further precipitation was observed under ultrasonic conditions. Then, the precipitate was centrifuged and washed with anhydrous CH 3 CN several times, drying at high vacuum to give product 3⋅4Br 136. 8

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
This work was supported by the National Natural Science Foundation of China [No. 21871216]. We also thank associate Prof. Hua Ke at Pingxiang University for the X-ray crystal structure measurement.