English

• In host-guest chemistry, the design of new macrocycles plays a key role in the development of efficient hosts for efficient recognition of discrete guests [1-19]. The generation of the so-called "blue box" [20], the tetracationic cyclobis(paraquat-p-phenylene)₂, by Stoddart has remarkably promoted the construction of various cationic cyclophanes or cages for the binding of different guests [21-27]. More recently, several cationic cyclophanes have been successfully used to include bioactive porphyrin agents to inhibit their posttreatment phototoxicity after photodynamic therapy [28-30]. Typically, such multicationic cyclophanes have been constructed through the incorporation of pyridinium units into the macrocyclic backbones. In contrast, multianionic macrocyclic hosts have been mainly developed by introducing carboxylate or sulfate groups or side chains that bear these anionic units to the backbones to enable the required water-solubility. In this context, a variety of multianionic macrocyclic hosts [31-35], including cyclodextrin [36], calix[n]arene [37], pillar[n]arene [38-41], cucurbit[n]uril [42] derivatives as well as acyclic folded containers [43-45], have been prepared for investigating new host-guest chemistry. Given the wide clinic application of sugammadex [36], the cyclodextrin-derived multianionic reversal drug for neuromuscular blockers rocuronium and vecuronium, it is valuable to develop new anionic macrocycles to enable efficient binding of photo- and bioactive molecular guests. Here, we report the synthesis of a new tetraanionic rectangular macrocycle ta that forms 1:1 host-guest complexes through unique parallel arrangement with a variety of electron-deficient aromatic compounds, with binding constants reaching up to 10⁷ L/mol in water. We further demonstrate that the efficient encapsulation of berberine and palmatine by ta can efficiently mask their bitter taste.

  Macrocyclic tetraacid TA was prepared from the hydrolysis of tetraester TE (Fig. 1), which was produced from the coupling reaction of the corresponding diphenol and 4,4′-di(bromo-methyl) biphenyl. The crystal structures of TE and TA are shown in Figs. 2a and b. It can be found that the two biphenyl units of the ester adopted parallel arrangement with a spatial separation of ~9.4 Å. Two of the four methyl benzoates are nearly perpendicular to the two biphenyl units, with the two methyl groups seizing the cavity of the macrocycle. In contrast, tetraacid TA gave rise to an inclined rectangular cavity. The biphenyl units and two of the four benzene rings constituted the two longer planes with spatial separation of ~7.3 Å, while another two benzene rings act as the shorter planes which has a spatial separation of ~16.5 Å. Tetraanionic ta was then prepared as Na⁺ salt by treating the acid TA with aqueous NaOH solution. This multicharged macrocycle had a good solubility of 15 mmol/L (Fig. S1 in Supporting information). We envisioned that it should possess a similar rectangular cavity in water, which was expected to encapsulate large size-matching aromatic guests driven by cooperative hydrophobicity and ion-pairing electrostatic attraction.

  Figure 1

  

  Figure 1.  The structures of macrocycles TE, TA and taBOX and investigated aromatic guests. In the brackets are binding constant between the respective compound and taBOX (methods used: ^a UV–vis titration; ^b UV–vis competition titration; ^c fluorescence titration; and ^d fluorescence competition titration).

  DownLoad: Full-Size Img PowerPoint

  Figure 2

  

  Figure 2.  Crystal structure of compounds (a) TE, (b) TA, (c) complex MV⊂taBOX (left: side view, right: top view), which reveals the unique parallel arrangement of MV in the cavity of taBOX (TE CCDC No.: 2262459, TA CCDC No.: 2262460, MV⊂taBOX CCDC No.: 2262463). (d) IGM analyses (δg^inter = 0.004) for complex MV⊂taBOX (left: side view, right: top view).

  DownLoad: Full-Size Img PowerPoint

  The crystal structure of the complex of ta and dicationic methyl viologen (MV) further confirmed this 1:1 stoichiometry (Fig. 2c). Remarkably, MV was incorporated completely in the cavity of ta. That is, MV was paralleled with the two long planes of ta, with its two methyl groups pointing to the short benzene planes of the rectangular cavity of ta. This inclusion motif is quite different from that of the complexation between Stoddart's tetracationic blue box or its extended analogues and linear electron-rich guests, where the linear guests are typically interwined with and threaded through the electron-deficient macrocycle. The 4-(phenoxymethyl)-1,1′-biphenyl units of ta may be considered as an unconventional electron-rich segment. Thus, the parallel arrangement of the 4-(phenoxymethyl)-1,1′-biphenyl units and MV in the complex indicated that the extended length of the new rectangular cavity enabled the inclusion of a linear guest to maximize the donor-acceptor interaction and minimize the exposure of hydrophobic MV to the polar aqueous medium. We further used independent gradient model (IGM) [46] to illustrate the noncovalent interactions that drove the 1:1 complexation between MV and ta (Fig. 2d).

  Totally twelve aromatic compounds (Fig. 1), including bitter drugs berberine and palmatine, were then evaluated for the encapsulating capacity of ta in water. Since tetraacid TA formed a rectangular cavity with the longer planes having a spatial separation of 7.3 Å, which is ideal for forming sandwich complexes with long aromatic guests, we expected ta formed 1:1 complexes with all these aromatic guests. To test this potential, Job's plots were obtained for its complexation with six organic dyes, i.e., acridine orange, basic blue 17, crystal violet, PIC, safranin O, neutral red and thioflavin T, by using UV–vis absorption titrations (Figs. S2–S8 in Supporting information), which supported that all the complexations occurred in the 1:1 binding motif.

  Direct or competitive UV–vis or fluorescence assays were then conducted to derive the binding constants (K_a) of the 1:1 complexes formed between ta and the aromatic guests in water (Figs. S9–S24 in Supporting information) [43], which are provided in Fig. 1. Neutral acridine and proflavin gave the lowest values. The value of acridine orange was substantially higher than that of proflavine that bears two polar amino groups, which may be ascribed to the hydrophobic dimethylamino groups of acridine orange as well as the deprotonation of proflavine. The values of basic blue 17, neutral red and safranin O were at the level of 10⁷ L/mol, indicating that ta is more efficient than well-investigated macrocycles such as cucurbit [7] uril or β-cyclodextrin [47]. Another six guests afforded values of 10⁵–10⁶ L/mol, which are still among the higher ones compared with other macrocyclic receptors [47].

  The K_as of berberine and palmatine were determined to be 2.04 × 10⁵ L/mol or 4.79 × 10⁶ L/mol. These two cationic alkaloids are clinically used for the treatment of enteritis or bacterial infection, but have an extremely bitter taste which disfavours patients during oral administration. Given the efficient binding of ta for both drugs, we further evaluate the potential of the macrocycle for masking their bitterness. For this aim, we first studied the binding motif of ta for berberine using ¹H NMR technique (Fig. S25 in Supporting information). Adding ta (1.0 equiv.) to the solution of berberine in D₂O caused all the signals of berberine to shift upfield (see Fig. 1 for signal numbering), even though several signals were not assigned due to overlapping or low resolution. Remarkably, the Ha and Hb signals underwent upfield shifting of 1.84 or 1.23 ppm, respectively. Hc-He signals of the aromatic rings could be assigned, which had a upfielding shifting of 0.78, 0.65 and 0.22 ppm, respectively. Similar upfield shiftings were also observed for palmatine in the presence of ta (Fig. S26 in Supporting information). Three signals (Ha-Hc) could be assigned (Fig. 1), which again suffered substantial upfield shifting, i.e., 1.66, 0.59 and 0.73 ppm, respectively. All these observations supported that these two guests formed 1:1 complexes with ta through the parallel stacking motif as revealed above for the complex of ta and MV. The structural difference of berberine and palmatine is only that their D rings bear a CH₂OCH₂ unit or two methoxy groups. However, the binding constant of palmatine is higher by 23.5 times than that of berberine, which might be considered as another evidence for the parallel encapsulation motif, through which the two more hydrophobic methoxy groups of palmatine could be more efficiently screened from the aqueous medium by the cavity of ta. This result is also consistent with the observation by Meng and Li et al. that a terphen [3] arene macrocycle exhibited comparable binding affinity toward berberine and palmatine through a threading binding motif [35].

  At pH 1.2, the typical acidity of gastrointestinal tract, ta was acidified to insoluble TA. Accordingly, the ¹H NMR spectrum of berberine and palmatine (0.3 mmol/L) in D₂O in the absence and presence (0.3 mmol/L) of ta gave rise to the same set of signals of the two drugs (Figs. S27 and S28 in Supporting information), and the signals of ta disappeared completely. These observations indicated that in the above acidic medium, ta was acidified to insoluble TA and thus lost the ability of including the two drugs, which meant that the drugs included by ta in neutral water would be released in gastrointestinal tract.

  Encouraged by the above experimental results, we then evaluate the biocompatibility of ta. The CCK-8 assay was first utilized to evaluate the cytotoxicity of ta for L02 and H9C2 cells. The results demonstrated a high survival rate (> 90%) of the cells at the high concentration of 0.25 mmol/L (Fig. 3a). To reveal the safety profile in vivo of ta, we further chose healthy BALB/c mice (5 male and 5 female) of close body weights to do the acute gavage test by using the high dose of 1000 mg/kg. The body weights and behaviours of the mice were monitored for two weeks (Fig. 3b). Compared with control group which was treated with distilled water, the body weights, physical and social behaviours of the ta-treated group had no obvious difference. Moreover, no detectable toxicity was observed histopathologically in major organ samples of the mice after 14 days (Fig. S29 in Supporting information). All the results indicated that ta had a relatively high biocompatibility.

  Figure 3

  

  Figure 3.  (a) Relative viabilities of H9C2 and L02 cells after incubation for 24 h with macrocycle taBOX at the indicated concentrations measured by CCK-8 assay (mean ± SD, n = 6). (b) Changes in body weight of mice (blue) after acute gavage test to taBOX (1.0 g/kg) as compared with the control group (distilled water, red). Data were represented as mean ± SD, n = 10, which had no statistically significant difference in body change over 14 days.

  DownLoad: Full-Size Img PowerPoint

  Mice have bitter taste receptors similar to those of human beings. We thus used mice model to evaluate if ta could mask the bitterness of berberine and palmatine by using drinking preference tests. Thus, three groups of BALB/c mice (n = 5) were firstly housed individually and accessed to two drinking bottles containing distilled water for 1 week to acclimatize to the environment. Then, they were subjected to two drinking bottles. One of the bottles contained distilled water, while another one contained ta, the mixture of ta and berberine or palmatine, and berberine or palmatine (Fig. 4), respectively. The concentration of both ta and the drugs was kept at 50 µmol/L by following the reported method [35]. The difference of the preference of the mice in choosing the bottles for drinking, which was determined by the amounts of the different liquids consumed, was used for evaluating the masking of ta for the bitterness of berberine. It can be seen that, for distilled water and ta, mice had no preference (Figs. 4e and f), which indicated that the ta had no aversive taste. In contrast, berberine and palmatine had only a 24% and 27% preference, respectively (Figs. 4a and b), which can be reasonably ascribed to their bitter taste. However, in the presence of the identical concentration of ta (Figs. 4c and d), the mice did not exhibit observable preference for distilled water and the mixing liquids. Clearly, the inclusion by macrocycle efficiently screened the bitterness of the drugs. It is noteworthy that, although the ability of the macrocycle for including berberine was considerably lower than that for palmatine, ta still completely suppressed its bitterness, which may be regarded as evidence for the unique parallel encapsulation mechanism which maximized the screening of the macrocycle for the drug to approach the taste receptors of the mice.

  Figure 4

  

  Figure 4.  Preference ratios of mice between two drinking bottles that contained distilled water and (a) berberine, (b) palmatine, (c) taBOX and berberine, (d) taBOX and palmatine, and (e, f) taBOX. All examples had the same concentration of 50 µmol/L. The distilled water group was used as control (mean ± SD, n = 5). Significant differences were assessed in (a–c) using unpaired t – tests; ns = not significant; ****P < 0.0001.

  DownLoad: Full-Size Img PowerPoint

  In summary, we have synthesized a tetraanionic macrocycle that has a rectangular hydrophobic cavity, which can strongly include large hydrophobic aromatic molecules through a unique parallel encapsulation mechanism, which is expected to maximize the screening of the included aromatic guests from the aqueous medium. For bitter drugs berberine and palmatine, this encapsulation mechanism enables the masking of their bitterness. The macrocycle can be prepared from readily available precursors and allows for modifications for both the backbone and the side chains. Introducing alkylsufonate side chains is expected to achieve increased water-solubility. Future endeavours will point to the design of macrocycles that have increased height for including stacking, dimeric aromatic guests or increased width for larger bioactive molecules or linear supramolecular or dynamical covalent guests.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgment

  We are grateful to the National Natural Science Foundation of China (Nos. 22271059, 21921003, 21890730 and 21890732) for financial support.

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2023.109119.

  
  
• 1. [1]

     D.J. Cram, J.M. Cram, Science 183 (1974) 803–809. doi: 10.1126/science.183.4127.803

  2. [2]

     S. Dong, B. Zheng, F. Wang, F. Huang, Acc. Chem. Res. 47 (2014) 1982–1994. doi: 10.1021/ar5000456

  3. [3]

     C.L. Deng, S.L. Murkli, L.D. Isaacs, Chem. Soc. Rev. 49 (2020) 7516–7532. doi: 10.1039/D0CS00454E

  4. [4]

     K. Yang, Z. Zhang, J. Du, W. Li, Z. Pei, Chem. Commun. 56 (2020) 5865–5876. doi: 10.1039/D0CC02001J

  5. [5]

     K. Velmurugan, M. Mohan, B. Li, et al., Mater. Adv. 1 (2020) 2646–2662. doi: 10.1039/D0MA00625D

  6. [6]

     W.C. Geng, J.L. Sessler, D.S. Guo, Chem. Soc. Rev. 49 (2020) 2303–2315. doi: 10.1039/C9CS00622B

  7. [7]

     W. Peng, H. Wang, D. Zhang, Z. Li, Chin. J. Org. Chem. 42 (2022) 863–870. doi: 10.6023/cjoc202108025

  8. [8]

     M.A. Beatty, F. Hof, Chem. Soc. Rev. 50 (2021) 4812–4832. doi: 10.1039/D0CS00495B

  9. [9]

     S. Zhang, I. Boussouar, H. Li, Chin. Chem. Lett. 32 (2021) 642–648. doi: 10.1016/j.cclet.2020.06.035

  10. [10]

      K. Wang, Q. Liu, L. Zhou, et al., Chin. Chem. Lett. 34 (2023) 108559. doi: 10.1016/j.cclet.2023.108559

  11. [11]

      K. Wang, M. Zhao, T. Zhang, H. Yue, X. Hu, Chin. Chem. Lett. 34 (2023) 107848. doi: 10.1016/j.cclet.2022.107848

  12. [12]

      J. Cheng, B. Gao, H. Tang, et al., Sci. China Chem. 65 (2022) 539–545.

  13. [13]

      Q. Liu, X. Tian, Y. Shen, et al., RSC Adv. 11 (2021) 38115–38119. doi: 10.1039/D1RA07958A

  14. [14]

      K. Wang, R. Zhang, Z. Song, et al., Adv. Sci. 10 (2023) 2206897–2206906. doi: 10.1002/advs.202206897

  15. [15]

      J. Cheng, B. Gao, H. Tang, et al., Sci. China Chem. 65 (2022) 539–546.

  16. [16]

      Q. Li, J.D. Sun, B. Yang, et al., Chin. Chem. Lett. 33 (2022) 1988–1992. doi: 10.1016/j.cclet.2021.10.017

  17. [17]

      M.S. Li, M. Quan, X.R. Yang, et al., Sci. China Chem. 65 (2022) 1733–1740. doi: 10.1007/s11426-022-1312-5

  18. [18]

      X. Jiang, H. Yu, J. Shi, et al., CCS Chem. 4 (2022) 2127–2139. doi: 10.31635/ccschem.021.202100948

  19. [19]

      T.H. Shi, M.X. Wang, CCS Chem. 3 (2021) 916–931. doi: 10.31635/ccschem.020.202000287

  20. [20]

      B. Odell, M.V. Reddington, A.M.Z. Slawin, et al., Angew. Chem. Int. Ed. 27 (1988) 1547–1550. doi: 10.1002/anie.198815471

  21. [21]

      B.M. Rambo, H.Y. Gong, M. Oh, J.L. Sessler, Acc. Chem. Res. 45 (2012) 1390–1401. doi: 10.1021/ar300076b

  22. [22]

      E.J. Dale, N.A. Vermeulen, M. Juricek, et al., Acc. Chem. Res. 49 (2016) 262–273. doi: 10.1021/acs.accounts.5b00495

  23. [23]

      Z. Liu, S.K.M. Nalluri, J.F. Stoddart, Chem. Soc. Rev. 46 (2017) 2459–2478. doi: 10.1039/C7CS00185A

  24. [24]

      I. Neira, A. Blanco-Gómez, J.M. Quintela, M.D. García, C. Peinador, Acc. Chem. Res. 53 (2020) 2336–2346. doi: 10.1021/acs.accounts.0c00445

  25. [25]

      T. Jiao, G. Wu, Y. Zhang, et al., Angew. Chem. Int. Ed. 59 (2020) 18350–18367. doi: 10.1002/anie.201910739

  26. [26]

      X. Chi, J. Tian, D. Luo, et al., Molecules 26 (2021) 2426. doi: 10.3390/molecules26092426

  27. [27]

      H. Nian, L. Cheng, L. Wang, et al., Angew. Chem. Int. Ed. 60 (2021) 15354–15358. doi: 10.1002/anie.202105593

  28. [28]

      I. Roy, S. Bobbala, R.M. Young, et al., J. Am. Chem. Soc. 141 (2019) 12296–12304. doi: 10.1021/jacs.9b03990

  29. [29]

      I. Roy, A. Garci, Y. Beldjoudi, et al., J. Am. Chem. Soc. 142 (2020) 16600–16609. doi: 10.1021/jacs.0c05445

  30. [30]

      J.D. Sun, Y. Liu, Z. Zhao, et al., RSC Med. Chem. 14 (2023) 563–572. doi: 10.1039/D2MD00463A

  31. [31]

      J. Dong, A.P. Davis, Angew. Chem. Int. Ed. 60 (2021) 8035–8048. doi: 10.1002/anie.202012315

  32. [32]

      H. Zhang, L.L. Wang, X.Y. Pang, L.P. Yang, W. Jiang, Chem. Commun. 57 (2021) 13724–13727. doi: 10.1039/D1CC05818E

  33. [33]

      X. Han, P. Li, Y. Han, C. Chen, Angew. Chem. Int. Ed. 60 (2022) e22202527.

  34. [34]

      D.V. Eker, S.K. Samanta, A.P. Davis, Chem. Commun. 56 (2020) 9268–9271. doi: 10.1039/D0CC03609A

  35. [35]

      J.Y. Chen, L.M. Chen, Y.H. Zhang, et al., Chem. Commun. 58 (2022) 3370–3373. doi: 10.1039/D2CC00040G

  36. [36]

      J. Hawkins, S. Khanna, M. Argalious, Curr. Pharm. Des. 25 (2019) 2140–2148. doi: 10.2174/1381612825666190704101145

  37. [37]

      D.S. Guo, Y. Liu, Acc. Chem. Res. 47 (2014) 1925–1934. doi: 10.1021/ar500009g

  38. [38]

      X. Shu, K. Xu, D. Hou, C. Li, Isr. J. Chem. 58 (2018) 1194–1204. doi: 10.1002/ijch.201800013

  39. [39]

      Y. Liu, S. Jiang, W. Mao, et al., Chin. Chem. Lett. 33 (2022) 209–212. doi: 10.1016/j.cclet.2021.07.024

  40. [40]

      Y. Chai, L. Chen, Y. Zhang, et al., Chin. Chem. Lett. 33 (2022) 3003–3006. doi: 10.1016/j.cclet.2021.11.087

  41. [41]

      H. Zhu, Q. Li, L.E. Khalil-Cruz, et al., Sci. China Chem. 64 (2021) 688–700. doi: 10.1007/s11426-020-9932-9

  42. [42]

      H.K. Liu, F. Lin, S.B. Yu, et al., J. Med. Chem. 65 (2022) 16893–16901. doi: 10.1021/acs.jmedchem.2c01677

  43. [43]

      D. Ma, B. Zhang, U. Hoffmann, et al., Angew. Chem. Int. Ed. 51 (2012) 11358–11362. doi: 10.1002/anie.201206031

  44. [44]

      M.H. Le, K. Taghuo, S. Estelle, T. Schrader, Chem. Commun. 58 (2022) 2954–2966. doi: 10.1039/D1CC06737K

  45. [45]

      Z. Wang, Z. Xu, J. Li, et al., Chin. J. Org. Chem. 42 (2022) 2236–2242.

  46. [46]

      T. Lu, F. Chen, J. Comput. Chem. 33 (2012) 580–592. doi: 10.1002/jcc.22885

  47. [47]

      R.N. Dsouza, U. Pischel, W.M. Nau, Chem. Rev. 111 (2011) 7941–7980. doi: 10.1021/cr200213s

• 

  Figure 1  The structures of macrocycles TE, TA and taBOX and investigated aromatic guests. In the brackets are binding constant between the respective compound and taBOX (methods used: ^a UV–vis titration; ^b UV–vis competition titration; ^c fluorescence titration; and ^d fluorescence competition titration).

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Crystal structure of compounds (a) TE, (b) TA, (c) complex MV⊂taBOX (left: side view, right: top view), which reveals the unique parallel arrangement of MV in the cavity of taBOX (TE CCDC No.: 2262459, TA CCDC No.: 2262460, MV⊂taBOX CCDC No.: 2262463). (d) IGM analyses (δg^inter = 0.004) for complex MV⊂taBOX (left: side view, right: top view).

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) Relative viabilities of H9C2 and L02 cells after incubation for 24 h with macrocycle taBOX at the indicated concentrations measured by CCK-8 assay (mean ± SD, n = 6). (b) Changes in body weight of mice (blue) after acute gavage test to taBOX (1.0 g/kg) as compared with the control group (distilled water, red). Data were represented as mean ± SD, n = 10, which had no statistically significant difference in body change over 14 days.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Preference ratios of mice between two drinking bottles that contained distilled water and (a) berberine, (b) palmatine, (c) taBOX and berberine, (d) taBOX and palmatine, and (e, f) taBOX. All examples had the same concentration of 50 µmol/L. The distilled water group was used as control (mean ± SD, n = 5). Significant differences were assessed in (a–c) using unpaired t – tests; ns = not significant; ****P < 0.0001.

  下载: 全尺寸图片 幻灯片
•