Vesicle fusion induced by zwitterionic amphiphilic channels

Citation: Qingyu Huan, Tao Lin, Yong-Hong Fu, Jun-Li Hou. Vesicle fusion induced by zwitterionic amphiphilic channels[J]. Chinese Chemical Letters, 2024, 35(1): 108566. doi: 10.1016/j.cclet.2023.108566 shu

Vesicle fusion induced by zwitterionic amphiphilic channels

English

Vesicle fusion induced by zwitterionic amphiphilic channels

Department of Chemistry, Fudan University, Shanghai 200433, China
^☆ This paper is dedicated to the memory of Prof. Jiang Wei.
^* Corresponding author.
E-mail address: houjl@fudan.edu.cn (J.-L. Hou).
Received Date: 10 February 2023
Accepted Date: 10 May 2023
Revised Date: 07 May 2023
Available Online: 15 January 2024

Abstract: A new strategy to induce vesicle fusion has been developed by employing pillar[5]arene derivatives that were channel-like and were prepared by appending side chains onto pillar[5]arenes backbones. The channels feature with hydrophilic negatively and positively charged groups at both ends and hydrophobic Trp residues at the outer surface, which endows the channels with amphiphilicity. The zwitterionic amphiphilic channels could spontaneously incorporate into the bilayer membranes of lipid vesicles to induce vesicle fusion driven by the electrostatic interactions between negatively charged and positively charged groups.

Key words:

In biological systems, membrane fusion is one of the central processes for cargo of biological molecules [1-3], which involves the merging of the membranes of two different organelles and the mixing of the compartments encapsulated by the membranes [4-7]. The process of the fusion consists of the membrane adhesion, outer leaflets fusion, formation of the fusion pore, and merger of the membrane to achieve the final full fusion [8,9]. In biological systems, this process is induced by fusion proteins [10]. Inspired by the proteins, chemists have great interests in constructing artificial systems to mediate vesicle-vesicle fusions because of the importance for drug delivery, cancer cell apoptosis, and disease diagnosis [11-15]. Studies of membrane fusion have extensively employed lipid vesicles as artificial analogues of cells. In the past decades, a large number of strategies have been developed to induce membrane fusion, including mediated by chemical reagents, such as lantibiotic [16], calcium salt [17], modified DNA [18], and synthetic proteins and peptides [19,20], or triggered by physical methods, such as electric field pulse [21-23], laser irradiation [24], electrostatic interaction [25,26], and so forth. In the context, electrostatic interaction-induced fusion is a promising approach due to its high efficiency in promoting the membrane adhesion. In this approach, it requires premixing of the positively or negatively charged reagents with the lipid to prepare the positively or negatively charged lipid vesicles, which would affect the stability and yield of the vesicles. Herein, we report a new strategy to induce vesicle fusion by post-addition of zwitterionic amphiphilic pillar[5]arene derivatives.

Pillar[5]arene, a new kind of macrocyclic paracyclophane, was first synthesized by Ogoshi et al. and has been extensively used in supramolecular and material chemistry due to its convenient preparation, rigid backbone, and deep cavity [27-52]. It has been established that conformations of pillar[5]arene derivatives are channel-like with attached sidechains pointing to two opposite directions along the pillar axle [53-56]. We have demonstrated that the unimolecular tubular structures could be efficiently constructed by appending linear sidechains onto the pillar[5]arene backbones [57-60]. The tubular structures could be positively and negatively charged or even zwitterionic by introducing different groups at both ends of the structures [61,62]. The structures with hydrophobic surface can spontaneous incorporated into lipid bilayers because of their amphiphilic structural feature.

We envisioned that the zwitterionic tubular molecules with amphiphilic backbone would insert into the bilayers of lipid vesicles to induce the fusion of the vesicles driven by electrostatic interaction (Fig. 1a). Thus, the zwitterionic tubular molecules 1–3 were designed (Fig. 1b). The terminal amino and carboxylic acid groups and the hydrophobic peptide chains endow the molecules with amphipathy. The tryptophan (Trp) residues in the peptide chains can further enhance the bilayer-incorporation ability of the tubular molecules driven by hydrogen bonding between indole amide and lipid headgroup oxygens and dipole interaction between the hydrophobic indole ring and the lipid tails [63-65]. The compounds were synthesized by attaching Trp-contained side-chains onto pillar[5]arene backbones via click-chemistry (see Supporting information).

Figure 1

Figure 1. (a) Schematic for the vesicle fusion induced by the zwitterionic tubular molecules. (b) The chemical structure of compounds 1–3.

DownLoad: Full-Size Img PowerPoint

The possibility of 1–3 to induce the fusion of the lipid vesicles was firstly investigated using Förster resonance energy transfer (FRET) experiments. For the experiments, large unilamellar vesicles (LUVs) consisting of 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DOPC), and N-(7-nitro-2,1,3-benzoxadiazol-4-yl) (NBD) labeled and Rhodamine (Rho) labeled 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE) were prepared, where the NBD and Rho moieties act as FRET energy donor and acceptor, respectively. Under a certain DOPE/DOPC ratio, NBD and Rho moieties have chance to move to close with each other and generate FRET that was monitored by exciting NBD moiety at 450 nm and detecting emission at 530 nm. The fusion of NBD -and Rho-labeled vesicles with vesicles without labeling can decrease the DOPE/DOPC ratio. As a result, the distance between NBD and Rho moieties would increase, which leads to lower FRET (Fig. 2a).

Figure 2

Figure 2. (a) Schematic for the experimental setup. The vesicle fusion will cause the increasement of the distance between NBD and Rho moieties, leading to the decrease of the FRET between the two moieties. (b) Changes in the normalized fluorescent intensity of NBD (λ_ex = 450 nm, λ_em = 530 nm) with time after mixing the NBD- and Rho-labeled vesicles with the unlabeled vesicles in the presence of 1–3 (mRL = 7.8%).

DownLoad: Full-Size Img PowerPoint

By mixing the labeled vesicle assay with the vesicle assay without labeling in the absence of the zwitterionic pillar[5]arene derivatives, minimal NBD fluorescence intensity change was observed (Fig. 2b), indicating the absence of the vesicle fusion. However, mixing both vesicle assays in the presence of 1–3 (molar ratio relative to lipid (mRL, %)) led to increasement of the fluorescence intensity with time, which clearly indicates the decreasing of FRET. This result provides the first evidence for the fusion of the vesicles mediated by 1–3. Compound 2 exhibited higher ability to mediate vesicle fusion in comparison with 1 and 3 as indicated by its faster fluorescence intensity increasing speed. The different ability of these three compounds to mediate vesicle fusion was ascribed to their different lipid membrane-incorporation ability. Compound 2 possessed the highest membrane-incorporation ability due to its length (3.3 nm) matched well with the thickness of the lipid bilayers (3.5 nm).

The fusion of the vesicles was further verified by inter-vesicle salt transport experiments. For the experiments, LUVs loading MgCl₂ (LUV^MgCl2) and Cl⁻ sensitive fluorescent probe 3-(6-methoxy-1-quinolinio)propanesulfonate (SPQ, LUV^SPQ) were prepared respectively. Mixing of both vesicle assays in the presence of compound 2 caused the significant decrease of SPQ fluorescence intensity (Fig. 3a), indicating the fluxing of MgCl₂ from LUV^MgCl2 into LUV^SPQ. Vesicle-based MgCl₂ transport experiments demonstrated that this compound could not mediate MgCl₂ transmembrane transport (Fig. S7 in Supporting information), due to the high dehydration energy of divalent Mg²⁺ that prevents this cation from entering the cavity of the channel molecules and blocks the cotransport of Mg²⁺ and Cl⁻. The blockage excluded the possibility of MgCl₂ transport in the pathway of LUV^MgCl2-bulk-LUV^SPQ. Thus, above SPQ fluorescence intensity decreasing under mixing both vesicle assays in the presence of 2 should be a result of the blending of the vesicle inner solution that was caused by the vesicle fusion. The extent of the fluorescence intensity decreasing was dependent on the dose of compound 2, supporting that the vesicle fusion was mediate by this compound.

Figure 3

Figure 3. (a) Changes in the fluorescent intensity of SPQ (λ_ex = 344 nm, λ_em = 443 nm) with time after mixing LUV^MgCl2 with LUV^SPQ in the presence of 2. (b) Plot of logV₀ vs. log[2], indicating the linear relationship of logV₀ with log[2]. The slop of the line was determined to be 1.2 ± 0.1.

DownLoad: Full-Size Img PowerPoint

The molecularity of compound 2 in the lipid bilayers was further evaluated by calculating the initial transport rate (V₀) from the slope of the stop-flow curves at t = 0 s. The log V₀ values were found to have a linear relationship with log[2] (Fig. 3b), and the slope were determined to be 1.2 ± 0.1. The fact that the slope was close to unity should be the evidence for the formation of unimolecular vesicle fusion trigger. This observation also suggested that, under the experimental conditions, the zwitterionic compound did not aggregate in the lipid bilayers.

Giant unilamellar vesicles (GUVs) usually act as a model for assessing membrane fusion because of their large dimensions for direct visualizing the fusion process. Therefore, GUVs were also employed in this work to investigate the membrane fusion. For the experiments, GUVs were prepared by swelling DOPC lipid on top of dried polyvinyl alcohol (PVA) surface. The vesicle fusion was triggered after the addition of the zwitterionic compound 2 onto the top solution of GUVs (Supplementary Video). Three vesicles that were suspended in the solution moved to close and contact with each other (Fig. 4a). Subsequently, the bilayer membranes of the contacting area fused with each other to form a hemifusion intermediate in a short term (Fig. 4b). The hemifusion membrane further expanded to result in a larger vesicle with sphere shape (Fig. 4c). The fusion process could be completed in around 2 min, which indicates a quick fusion process. The quick fusion should be a result of the strong electrostatic interaction between the zwitterionic pillar[5]arene derivatives locating in the adjacent bilayers. The zwitterionic compound 2 features with high charge density at both ends, which contributes to the strong electrostatic interaction.

Figure 4

Figure 4. Microscopy images of the GUVs recorded at (a) 30 s, (b) 45 s, (c) 118 s after addition of compound 2, highlighting the fusion of GUVs induced by 2.

DownLoad: Full-Size Img PowerPoint

In conclusion, we have developed a new strategy to induce vesicle fusion by employing zwitterionic pillar[5]arene derivatives. The derivatives were channel-like and were prepared by appending side chains onto pillar[5]arenes backbones. The pillar[5]arene channels contains hydrophilic negatively and positively charged groups at both ends and hydrophobic Trp residues at the outer surface. This structural feature endows the channels with amphiphilicity. The zwitterionic amphiphilic channels can efficiently incorporate into the bilayer membranes of lipid vesicles and induce vesicle fusion driven by the electrostatic interactions between negatively charged and positively charged groups.

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.

Acknowledgments

This work was support by the National Natural Science Foundation of China (NSFC, Nos. 21921003 and 21971046) and the Science and Technology Commission of Shanghai Municipality (STCSM, No. 22JC1403700).

Supplementary materials

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2023.108566.
1. [1]
  R. Blumenthal, M.J. Clague, S.R. Durell, Chem. Rev. 103 (2003) 53–70. doi: 10.1021/cr000036+
2. [2]
  L.V. Chernomordik, J. Zimmerberg, Curr. Opin. Struct. Biol. 5 (1995) 541–547. doi: 10.1016/0959-440X(95)80041-7
3. [3]
  L.V. Chernomordik, M.M. Kozlov, Cell 123 (2005) 375–382. doi: 10.1016/j.cell.2005.10.015
4. [4]
  D. Fasshauer, Biochim. Biophys. Acta Mol. Cell Res. 1641 (2003) 87–97. doi: 10.1016/S0167-4889(03)00090-9
5. [5]
  W. Hong, Biochim. Biophys. Acta Mol. Cell Res. 1744 (2005) 120–144. doi: 10.1016/j.bbamcr.2005.03.014
6. [6]
  V. Rossi, D.K. Banfield, M. Vacca, et al., Trends Biochem. Sci. 29 (2004) 682–688. doi: 10.1016/j.tibs.2004.10.002
7. [7]
  A.T. Brunger, Q. Rev. Biophys. 38 (2005) 1–47.
8. [8]
  J. Rizo, T.C. Südhof, Nat. Struct. Biol. 5 (1998) 839–842. doi: 10.1038/2280
9. [9]
  B. Gong, B.K. Choi, J.Y. Kim, et al., J. Am. Chem. Soc. 137 (2015) 8908–8911. doi: 10.1021/jacs.5b05385
10. [10]
  L.V. Chernomordik, M.M. Kozlov, Annu. Rev. Biochem. 72 (2003) 175–207. doi: 10.1146/annurev.biochem.72.121801.161504
11. [11]
  D.L. Perrier, L. Rems, P.E. Boukany, Adv. Colloid Interface Sci. 249 (2017) 248–271. doi: 10.1016/j.cis.2017.04.016
12. [12]
  J. Xu, W. Cao, P. Wang, H. Liu, Pharmaceuticals 15 (2022) 876. doi: 10.3390/ph15070876
13. [13]
  B.P. Jena, J. Cell. Mol. Med. 8 (2004) 1–21. doi: 10.1111/j.1582-4934.2004.tb00255.x
14. [14]
  X. Shi, Q. Cheng, Y. Zhang, Methods 177 (2020) 95–102. doi: 10.1016/j.ymeth.2019.09.017
15. [15]
  M.N. Jones, Curr. Opin. Colloid Interface Sci. 1 (1996) 91–100. doi: 10.2307/1163382
16. [16]
  A.J.M. Driessen, H.W. van den Hooven, W. Kuiper, et al., Biochemistry 34 (1995) 1606–1614. doi: 10.1021/bi00005a017
17. [17]
  J. Wilschut, N. Duzgunes, D. Papahadjopoulos, et al., Biochemistry 19 (1980) 6011–6021. doi: 10.1021/bi00567a011
18. [18]
  J.A. Peruzzi, M.L. Jacobs, T.Q. Vu, et al., Angew. Chem. Int. Ed. 58 (2019) 18683–18690. doi: 10.1002/anie.201911544
19. [19]
  S.C. Hsu, C.D. Hazuka, R.H. Scheller, et al., Trends Cell Biol. 9 (1999) 150–153. doi: 10.1016/S0962-8924(99)01516-0
20. [20]
  G.R. Prescott, O.A. Gorleku, L.H. Chamberlain, et al., J. Neurochem. 110 (2009) 1135–1149. doi: 10.1111/j.1471-4159.2009.06205.x
21. [21]
  J. Khadake, B. Meldal, S. Panni, et al., Nat. Commun. 10 (2019) 10. doi: 10.1038/s41467-018-07709-6
22. [22]
  C.K. Haluska, K.A. Riske, V. Marchi-Artzner, et al., Proc. Natl. Acad. Sci. U.S.A. 103 (2006) 15841–15846. doi: 10.1073/pnas.0602766103
23. [23]
  M.R. Cheetham, J.P. Bramble, D.G.G. McMillan, et al., J. Am. Chem. Soc. 133 (2011) 6521–6524. doi: 10.1021/ja2007615
24. [24]
  Y. Suzuki, K.H. Nagai, T. Hamada, et al., Langmuir 33 (2017) 2671–2676. doi: 10.1021/acs.langmuir.7b00448
25. [25]
  R.B. Lira, T. Robinson, K.A. Riske, et al., Biophys. J. 116 (2019) 79–91. doi: 10.1016/j.bpj.2018.11.3128
26. [26]
  J. Voskuhl, B.J. Ravoo, Chem. Soc. Rev. 38 (2009) 495–505. doi: 10.1039/B803782P
27. [27]
  T. Ogoshi, T.A. Yamagishi, Y. Nakamoto, Chem. Rev. 116 (2016) 7937–8002. doi: 10.1021/acs.chemrev.5b00765
28. [28]
  B. Hua, L. Shao, F. Huang, et al., Acc. Chem. Res. 55 (2022) 1025–1034. doi: 10.1021/acs.accounts.2c00011
29. [29]
  D. Xia, P. Wang, F. Huang, et al., Chem. Rev. 120 (2020) 6070–6123. doi: 10.1021/acs.chemrev.9b00839
30. [30]
  Z. Li, Y.W. Yang, Acc. Chem. Res. 2 (2021) 292–305. doi: 10.1021/accountsmr.1c00042
31. [31]
  H. Zheng, L. Fu, L. Wang, et al., Nat. Commun. 14 (2023) 590. doi: 10.54254/2754-1169/3/2022839
32. [32]
  X. Liang, Y. Shen, C. Yang, et al., Chem. Commun. 58 (2022) 13584–13587. doi: 10.1039/d2cc05690a
33. [33]
  X.Q. Wang, W.J. Li, H.B. Yang, et al., J. Am. Chem. Soc. 141 (2019) 13923–13930. doi: 10.1021/jacs.9b06739
34. [34]
  J.R. Wu, G. Wu, Y.W. Yang, et al., Angew. Chem. Int. Ed. 62 (2023) e202218142. doi: 10.1002/anie.202218142
35. [35]
  B. Jiang, W. Wang, H.B. Yang, et al., Angew. Chem. Int. Ed. 56 (2017) 14438–14442. doi: 10.1002/anie.201707209
36. [36]
  T. Ogoshi, S. Kanai, S. Fujinami, et al., J. Am. Chem. Soc. 130 (2008) 5022–5023. doi: 10.1021/ja711260m
37. [37]
  X.Y. Yu, Y. Chen, Chin. Chem. Lett. 34 (2023) 107739. doi: 10.1016/j.cclet.2022.08.019
38. [38]
  X.N. Han, Q.S. Zong, C.F. Chen, et al., CCS Chem. 4 (2021) 318–330. doi: 10.1038/s42255-021-00356-0
39. [39]
  X. Shu, W. Chen, C. Li, et al., Chem. Commun. 50 (2014) 4820–4823. doi: 10.1039/C4CC00800F
40. [40]
  Z. Liu, F. Demontrond, A. Imberty, et al., Chin. Chem. Lett. 34 (2023) 107872. doi: 10.1016/j.cclet.2022.107872
41. [41]
  Q. Wang, J. Fan, X. Bian, et al., Chin. Chem. Lett. 33 (2022) 1979–1982. doi: 10.1016/j.cclet.2021.10.040
42. [42]
  S. Liu, T. Yan, Q. Wu, et al., Chin. Chem. Lett. 33 (2022) 239–242. doi: 10.3390/land11020239
43. [43]
  Y. Chen, B. Sun, L. Wang, et al., Org. Lett. 23 (2021) 7423–7427. doi: 10.1021/acs.orglett.1c02620
44. [44]
  Y. Zhou, H. Tang, D. Cao, et al., Chem. Commun. 57 (2021) 13114–13117. doi: 10.1039/d1cc05153a
45. [45]
  X. Ji, F. Wang, F. Huang, et al., Chin. J. Chem. 38 (2020) 1473–1479. doi: 10.1002/cjoc.202000314
46. [46]
  S. Fa, T. Kakuta, T.A. Yamagishi, T. Ogoshi, CCS Chem. 1 (2019) 50–63. doi: 10.31635/ccschem.019.20180014
47. [47]
  S. Lan, L. Ling, D. Ma, ACS Appl. Mater. Interfaces 14 (2022) 4197–4203. doi: 10.1021/acsami.1c21575
48. [48]
  M. Cheng, W. Gong, H. Li, et al., Chin. J. Chem. 40 (2022) 925–930. doi: 10.1002/cjoc.202100714
49. [49]
  J. Liu, X.W. Sun, Q. Lin, et al., Chin. J. Chem. 39 (2021) 3421–3428. doi: 10.1002/cjoc.202100583
50. [50]
  J.R. Wu, Y.W. Yang, CCS Chem. 3 (2020) 836–843.
51. [51]
  J. Wang, M. Cen, Y. Yao, et al., Chin. Chem. Lett. 33 (2022) 1475–1478. doi: 10.3390/sym14071475
52. [52]
  Y. Zhang, J. Chen, Q. Meng, et al., Chin. Chem. Lett. 34 (2023) 107697. doi: 10.1016/j.cclet.2022.07.040
53. [53]
  T. Ogoshi, K. Kitajima, Y. Nakamoto, et al., Org. Lett. 12 (2010) 636–638. doi: 10.1021/ol902877w
54. [54]
  T. Ogoshi, M. Hashizume, Y. Nakamoto, et al., Chem. Commun. 46 (2010) 3708–3710. doi: 10.1039/c0cc00348d
55. [55]
  K. Kato, S. Fa, T. Ogoshi, et al., Chem. Soc. Rev. 51 (2022) 3648–3687. doi: 10.1039/d2cs00169a
56. [56]
  S. Ohtani, K. Kato, T. Ogoshi, et al., Coord. Chem. Rev. 462 (2022) 214503. doi: 10.1016/j.ccr.2022.214503
57. [57]
  Y.W. Li, Y.H. Fu, J.L. Hou, Chin. J. Chem. 40 (2022) 1293–1297. doi: 10.1002/cjoc.202100836
58. [58]
  W. Si, Z.T. Li, J.L. Hou, et al., Acc. Chem. Res. 48 (2015) 1612–1619. doi: 10.1021/acs.accounts.5b00143
59. [59]
  J.Y. Chen, Z.T. Li, J.L. Hou, et al., Faraday Discuss. 209 (2018) 149–159. doi: 10.1039/C8FD00009C
60. [60]
  W. Si, Z.T. Li, J.L. Hou, et al., Angew. Chem. Int. Ed. 50 (2011) 12564–12568. doi: 10.1002/anie.201106857
61. [61]
  Z.J. Yan, D. Wang, J.L. Hou, et al., J. Am. Chem. Soc. 142 (2020) 15638–15643. doi: 10.1021/jacs.0c00601
62. [62]
  Z.J. Yan, Y. Zhang, J.L. Hou, et al., J. Am. Chem. Soc. 143 (2021) 11332–11336. doi: 10.1021/jacs.1c06000
63. [63]
  V. Fonseca, P. Daumas, L.O.S. Andersen, et al., Biochemistry 31 (1992) 5340–5350. doi: 10.1021/bi00138a014
64. [64]
  S.A. Seoh, D. Busath, Biophys. J. 68 (1995) 2271–2279. doi: 10.1016/S0006-3495(95)80409-1
65. [65]
  Q. Xiao, W.W. Haoyang, J.L. Hou, et al., Chem. Commun. 57 (2021) 863–866. doi: 10.1039/d0cc06974d
Figure 1 (a) Schematic for the vesicle fusion induced by the zwitterionic tubular molecules. (b) The chemical structure of compounds 1–3.

下载: 全尺寸图片幻灯片

Figure 2 (a) Schematic for the experimental setup. The vesicle fusion will cause the increasement of the distance between NBD and Rho moieties, leading to the decrease of the FRET between the two moieties. (b) Changes in the normalized fluorescent intensity of NBD (λ_ex = 450 nm, λ_em = 530 nm) with time after mixing the NBD- and Rho-labeled vesicles with the unlabeled vesicles in the presence of 1–3 (mRL = 7.8%).

下载: 全尺寸图片幻灯片

Figure 3 (a) Changes in the fluorescent intensity of SPQ (λ_ex = 344 nm, λ_em = 443 nm) with time after mixing LUV^MgCl2 with LUV^SPQ in the presence of 2. (b) Plot of logV₀ vs. log[2], indicating the linear relationship of logV₀ with log[2]. The slop of the line was determined to be 1.2 ± 0.1.

下载: 全尺寸图片幻灯片

Figure 4 Microscopy images of the GUVs recorded at (a) 30 s, (b) 45 s, (c) 118 s after addition of compound 2, highlighting the fusion of GUVs induced by 2.

下载: 全尺寸图片幻灯片