Cytomimetic Large-Scale Vesicle Aggregation and Fusion Based on

Publication Date (Web): November 30, 2011 ... The intervesicular host–guest recognition interactions between β-CD units in CD-BPs and Ada units in ...
0 downloads 0 Views 3MB Size
Download PDF
ARTICLE pubs.acs.org/Langmuir

Cytomimetic Large-Scale Vesicle Aggregation and Fusion Based on Host Guest Interaction Haibao Jin, Yong Liu, Yongli Zheng, Wei Huang, Yongfeng Zhou,* and Deyue Yan School of Chemistry and Chemical Engineering, State Key Laboratory of Metal Matrix Composites, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, P. R. China

bS Supporting Information ABSTRACT: Herein, we have shown a large-scale cell-mimetic (cytomimetic) aggregation process by using cell-sized polymer vesicles as the building blocks and intervesicular host guest molecular recognition interactions as the driving force. We first prepared the hyperbranched polymer vesicles named branched polymersomes (BPs) around 5 10 μm through the aqueous self-assembly of a hyperbranched multiarm copolymer of HBPO-star-PEO [HBPO = hyperbranched poly(3-ethyl-3-oxetanemethanol); PEO = poly(ethylene oxide)]. Subsequently, adamantane-functionalized BPs (Ada-BPs) or β-cyclodextrin-functionalized BPs (CD-BPs) were prepared through the coassembly of HBPO-star-PEO and Ada-modified HBPO-star-PEO (HBPO-star-PEO-Ada), or of HBPOstar-PEO and CD-modified HBPO-star-PEO (HBPO-star-PEO-CD), respectively. Macroscopic vesicle aggregates were obtained by mixing CD-BPs and Ada-BPs. The intervesicular host guest recognition interactions between β-CD units in CD-BPs and Ada units in Ada-BPs, which were proved by 1H nuclear Overhauser effect spectroscopy (NOESY) spectrum and the fluorescence probe method, are responsible for the vesicle aggregation. Additionally, the vesicle fusion events happened frequently in the process of vesicle aggregation, which were certified by double-labeling fluorescent assay, real-time observation, content mixing assay, and component mixing assay.

’ INTRODUCTION Cell aggregation through specific intercellular molecular recognitions is one of the most important biologic processes in living systems, which plays critical roles in hemostasis, immune response, inflammation, embryogenesis, and development of neuronal tissue.1 As an artificial membrane, vesicle has been widely used to mimick the biological cellular processes in vivo, due to the similarity with cell membrane in bilayer structure and the properties of fluidity, semipermeability, and deformability,2 and the advantages of higher stability and easier functional versatility.3 Nowadays, a nomenclature of “cytomimetic chemistry”, as reviewed by Menger et al.,4 was generated to describe the cell-mimetic behaviors by using the vesicles as the model membranes. As a part of cyctomimetic chemistry, vesicle aggregation to mimick cell aggregation has also attracted much attention. It may extend from the simple adhesion behavior of several vesicles to the complicated behavior of multivesicular arrays aiming to build the artificial tissue-like entities.5 Up to now, all the reported vesicles including liposomes (lipid vesicles), surfactant vesicles and polymersomes (polymer vesicles) have been used to mimic cell aggregation. Among them, liposomes are most widely studied, and a large amount of papers have been reported by Zasadzinski,6 Menger,7 Lehn,8 Boxer,9 Meier,10 Paleos,11 Bong,12 and Webb13 et al. to realize the liposome aggregation through intervesicular interactions including hydrogen bonding, electrostatic interaction, iron coordination and so r 2011 American Chemical Society

forth. Ravoo14 and Kim15 have reported the aggregation of surfactant vesicles self-assembled from the cyclodextrin (CD)or cucurbituril (CB)-based amphiphiles driven by the intervesicular host guest interactions. Polymersomes that self-assembled from linear block copolymers have the advantages of higher stability and easier customization when compared to the liposomes or surfactant vesicles, and are very promising in cytomimetic aggregation.16,17 Discher,18 Hammer,19 Zhou20 and Ryan21 et al. have realized the aggregation of polymersomes through intervesicular hydrophobic interactions or avidin biotin recognitions. In short, rapid progress has been made in the studies on cytomimetic aggregation by using vesicles as the model systems. Nevertheless, most of the work is limited to the aggregation of vesicles with submicroscopic sizes, and the scale of the reported vesicle aggregates is smaller than 100 μm. However, in nature, the cells are generally micrometer-sized, and their aggregates are in macroscopic scale. For example, the epidermis of skin is a typical large-scale cell aggregate. Thus, to better mimick cell aggregation, it is important to prepare large-scale aggregates from cell-sized Special Issue: Bioinspired Assemblies and Interfaces Received: October 1, 2011 Revised: November 20, 2011 Published: November 30, 2011 2066

dx.doi.org/10.1021/la203857s | Langmuir 2012, 28, 2066–2072

Langmuir vesicles. Unfortunately, little work has been performed in this area. In recent years, we have reported a new type of polymer vesicle named “branched polymersomes” (BPs) via the self-assembly of amphiphilic hyperbranched multiarm copolymers.22 BPs have a good membrane fluidity, strong stability, cell-like size, and a large amount of functional groups favorable to form multivalent interactions.23 Thus, BPs are promising in mimicking the cell aggregation process. As the proof-of-principle experiments, very recently we have prepared large-scale vesicle aggregates between β-CD-modified BPs (CD-BPs) and azobenzene (Azo)-modified BPs (Azo-BPs) based on intervesicular β-CD/Azo molecular recognition interactions.24 Scheme 1. Aggregation of BPs Triggered by the Host Guest Recognition Interaction between β-CD Groups and Ada Groups

ARTICLE

Herein, we have shown another example of large-scale cytomimetic aggregation from BPs. An intervesicular interaction based on β-CD/adamantane (Ada) molecular recognition, which is stronger than β-CD/Azo interaction,25 was introduced to induce a more efficient vesicle aggregation. For this purpose, two populations of micrometer-sized BPs, one functionalized with β-CD groups (CD-BPs) and the other with Ada groups (Ada-BPs), were prepared through a coassembly method (Scheme 1). Subsequently, CD-BPs and Ada-BPs spontaneously aggregated together through the multivalent CD/Ada host guest interactions into macroscopic aggregates. Two advances were made in such a vesicle aggregation process due to the strong intervesicular CD/Ada interactions. First, we obtained a very large vesicle aggregate, and to our knowledge, it may represent the largest vesicle aggregate reported up to now. Second, strong and frequent fusion events were observed, which indicates that the aggregation-induced fusion is highly dependent on the intervesicular interaction.

’ EXPERIMENTAL SECTION Materials. Hyperbranched multiarm copolymers of HBPO-starPEO, where HBPO and PEO represent the hydrophobic hyperbranched poly(3-ethyl-3-oxetanemethanol) core and hydrophilic poly(ethylene oxide) arms, respectively, were synthesized through cationic ring-opening polymerization (CROP) according to our previous method.22 Two polymer samples, having the same HBPO cores with a number-average molecular weight (Mn) of about 4300 according to gel permeation

Scheme 2. Synthesis Schemes of HBPO-star-PEO-CD (1) and HBPO-star-PEO-Ada (2)a

a Reagents and conditions: (i) butanedioic anhydride, TEA, dried THF, reflux, 24 h; (ii) β-CD, dicyclohexylcarbodiimide (DCC), dried dimethylformamide (DMF), room temperature (r.t.), 24 h; (iii) 1-adamantanecarbonyl chloride, TEA, dired CHCl3, r.t., 24 h.

2067

dx.doi.org/10.1021/la203857s |Langmuir 2012, 28, 2066–2072

Langmuir chromatography (GPC) and a degree of branching (DB) of 38%, but different PEO arms, were prepared. The number-average degrees of polymerization (DParm) of the PEO arms for these two polymers were 2 and 4, respectively. Their Mn values measured by GPC were about 8500 and 12 000, respectively. Triethylamine (TEA) and tetrahydrofuran (THF) (AR grade, Shanghai Chemical Reagent Co.) were refluxed with CaH2 and then distilled prior to use. Rhodamine B (>95% pure), dansyl chloride (>99% pure), 1-adamantanecarbonyl chloride (>99% pure), and 1-pyrenebutyric acid (>97% pure) were purchased from Aldrich and used as received. Terbium(III) chloride hexahydrate (>99.9% pure) and dipicolinic acid (DPA) (>99% pure) were purchased from J & K Chemical (China). All the other chemical reagents were purchased from Shanghai Chemical Reagent Co. and used as received. Synthesis of HBPO-star-PEO-CD (Molecule 1). The synthetic process includes two steps (Scheme 2, steps i and ii). In step i, HBPOstar-PEO with DParm = 2 was reacted with butanedioic anhydride to obtain the carboxyl-terminated polymers (HBPO-star-PEO-COOH). In step ii, the carboxyl groups in HBPO-star-PEO-COOH were further reacted with the hydroxyl groups in β-CD through the esterification reaction to obtain the final CD-functionalized polymer of HBPO-starPEO-CD (1). The detailed syntheses and characterizations of 1 were shown in our previously reported literature.24 The percent grafting of CD groups is approximately 11.3% according to the NMR characterization. Synthesis of HBPO-star-PEO-Ada (Molecule 2). The synthetic process is shown in Scheme 2 (step iii). Typically, HBPO-starPEOs with DParm = 4 (3.200 g, ca.11.0 mmol OH group) were added into a 150 mL single-neck round-bottom flask equipped with a dropping funnel. 60 mL CHCl3 and triethylamine (0.62 mL, 4.5 mmol) were sequentially added under the protection of nitrogen. After the flask was immersed into the ice water bath, 1-adamantanecarbonyl chloride (0.600 g, 3.0 mmol) dissolving in 20 mL CHCl3 was added into a funnel via a syringe and dropped into the flask at a rate of one drop per second. Then, the solution was kept stirring for 24 h at room temperature. The reaction mixture was filtered and precipitated in hexane three times. The products were dried in a vacuum drying oven under room temperature for 24 h. The purified polymers were characterized by 1H NMR, and the Ada percent grafting was about 14.5%. The detailed characterizations are shown in the Supporting Information (Figure S1). Syntheses of Fluorophore-Labeled HBPO-star-PEO. HBPO-starPEO-Rb was synthesized by the esterification reaction between rhodamine B and HBPO-star-PEO (DParm = 2), and the percent grafting of rhodamine B was about 4.9%. HBPO-star-PEO-DNS was synthesized through the end-groups modification of HBPO-star-PEO (DParm = 2) with dansyl chloride, and the percent grafting of dansyl groups was about 3.09%. HBPO-star-PEO-Py was prepared through esterification between 1-pyrenebutyric acid and HBPO-star-PEO (DParm = 2), and the percent grafting of pyrene groups was about 8.59%. The details of the synthesis processes and the characterizations can be obtained in the previous report.24 Coassembly of Ada-BPs or CD-BPs. HBPO-star-PEO and HBPO-star-PEO-Ada (2) were dissolved in 5 mL of chloroform solvent for 30 min, respectively. Then, these two solutions were mixed together (the weight ratio between HBPO-star-PEO and HBPO-star-PEO-Ada was 1:1) and dried by rotary evaporation to obtain a uniformly mixed polymer solid (mixture 2). The coassembly was performed by adding mixture 2 to 5 mL of distilled water at about 25 °C with strong stirring. The appearance of turbidity in the solution indicated the formation of Ada-BPs. The final polymer concentration is 5 mg/mL, and the Ada group concentration is approximately 1.19 mM. The same method was used to realize the coassembly of HBPO-star-PEO and HBPO-starPEO-CD (1) to obtain CD-BPs. The dried solid from HBPO-star-PEO and HBPO-star-PEO-CD was called mixture 1. The final polymer

ARTICLE

concentration was 5 mg/mL, and the concentration of β-CD groups was about 1.25 mM. Encapsulation of TbCl3 or DPA in BPs. CD-BPs and Ada-BPs were prepared in either (a) 2.5 mM TbCl3 and 20 mM sodium citrate solution or (b) 30 mM DPA (sodium salt) solution. After strong stirring at about 25 °C for 24 h, the solution was dialyzed against pure water for 3 days to remove the unencapsulated components. Thus, Tb3+ and DPA were encapsulated into CD-BPs or Ada-BPs to obtain Tb-CD-BPs, DPA-CD-BPs, Tb-Ada-BPs, and DPA-Ada-BPs. For fusion measurements, the Tb fluorescence scale was calibrated as follows: 20 μM DPA was added to Tb3+ encapsulated BPs (for example, Tb-CD-BPs). Subsequently, acetone (0.5 mL) was added to release the contents (Tb3+) from the vesicles. Then evaporation was used to remove the acetone. The fluorescence observed after acetone addition was taken as the maximal (100%) value (Fmax). The fluorescence of the solution at a certain time (Ft) was normalized with Fmax to get the Tb fluorescence percentage. To determine the leakage of the vesicle contents during the fusion process, 1 mM ethylenediaminetetraacetic acid (EDTA) was added to the sample in fusion, which possesses a strong coordination interaction with Tb3+ in comparison with DPA. Thus, the highly fluorescent Tb(DPA)33 complexes leaked from the vesicles would be destroyed by the added EDTA, which leads to fluorescence quenching. Host Guest Interactions. To characterize the host guest interactions between CD and Ada units, the dansyl fluorescence probe method and two-dimensional nuclear Overhauser effect spectroscopy (2D NOESY) method were used. For the fluorescence experiments, 0.18 mM dansyl chloride aqueous solution was prepared and divided into three aliquots of 1 mL each. Nothing was added in the first aliquot. Five milligrams of HBPO-star-PEO-CD was added into the second aliquot with stirring for 2 days. In the third aliquot, 5 mg of HBPO-starPEO-CD was added with stirring for 1 day, and then 1 mg of HBPO-starPEO-Ada was added with stirring for another day. The fluorescent spectra of each sample were measured. For the 2D NOESY experiment, the solid of mixture 1 from HBPOstar-PEO-CD and HBPO-star-PEO was put into D2O to induce the coassembly of CD-BPs, and the solid of mixture 2 from HBPO-starPEO-Ada and HBPO-star-PEO was put into D2O to induce the coassembly of Ada-BPs. Then the vesicle solutions were mixed together for 24 h before the 2D NOESY spectra measurement. Characterizations. 1H NMR was performed on a Varian Mercury Plus 400-MHz spectrometer using deuterated dimethyl sulfoxide (DMSO-d6) or CDCl3 as solvents at 20 °C, and tetramethylsilane (TMS) was used as the internal reference. The 2D NOESY experiment was performed under Avance III 400, in which the 3D spatial correlations of neighboring protons were probed. Cross peaks in the 2D 1H NMR spectrum indicate close proximity, that is, distances less than 0.5 nm. The morphologies of the vesicles were observed by optical microscopy and fluorescent microscope on a Leica DM4500 B. For the microscope observation, the aqueous solution was dropped onto a glass slide and then observed directly under the microscope. Fluorescent spectra were recorded on a QC-4-CW spectrometer, made by Photon Technology International, Int. USA/CAN. The excitation wavelength of pyrene was set at 335 nm. Step increment was set as 1 nm, and scan speed was set at 480 nm/min. To the dansyl unit, the excitation wavelength was set at 336 nm. To the complex of Tb(DPA)33 , the excitation wavelength was set at 276 nm.

’ RESULTS AND DISSCUSSION Preparation of Binary BPs. The obtained HBPO-star-PEO with DParm = 2 can self-assemble into a large population of giant vesicles (BPs) about 5 10 μm (Figure S2 in the Supporting Information) on average according to the statistic result of 2068

dx.doi.org/10.1021/la203857s |Langmuir 2012, 28, 2066–2072

Langmuir

ARTICLE

Figure 2. Vesicle fusion during aggregation. Optical (A) and fluorescent overlay (B) micrographs of large-scale vesicle aggregates from (red fluorescent) CD-BPs and (green fluorescent) Ada-BPs after keeping for 3 days. (C) Real-time fusion sequences between two giant polymer vesicles. The number in the symbol labeled on each image denotes the elapsed time (in seconds), and the time of the first image is set as zero. Scale bars represent 20 μm.

Figure 1. Optical micrographs of the functional vesicles and vesicle aggregates. (A) Red-fluorescent CD-BPs labeled by rhodamine B groups. (B) Green-fluorescent Ada-BPs labeled with dansyl groups. (C) Optical micrograph of large-scale vesicle aggregates; the inset shows the magnified image of the purple rectangle frame. Scale bars represent 20 μm.

optical microscopy. In our previously work, the vesicle structure was carefully characterized by transmission electron microscopy (TEM) and small-angle X-ray scattering (SAXS) measurements, which indicate that the BPs are unilamellar vesicles with a bilayer structure around 20 nm.24 To prepare Ada-BPs, HBPO-starPEO-Ada (2) and HBPO-star-PEO were mixed together in the cosolvent chloroform, and then water was added into the dried mixtures (mixture 2) to induce the self-assembly. Due to the similar structure, molecule 2 was expected to be spontaneously doped into the membranes of BPs to form the functional Ada-BPs (1.19 mM of Ada units). A similar method was used to prepare the CD-BPs (1.25 mM of Ada unit) by the coassembly of HBPO-star-PEO-CD (1) and HBPO-star-PEO. There are no obvious changes in size and morphology after the coassembly when compared to the single-component BPs. Figure 1a,b shows the fluorescent images of CD-BPs and Ada-BPs, respectively. For the observation, CD-BPs were doped with the red-fluorescent HBPO-star-PEO-Rb (rhodamine B-labeled), while Ada-BPs were doped with the green-fluorescent HBPO-star-PEO-DNS (dansyllabeled) during the coassembly process. These fluorescent micrographs clearly illustrate the hollow interior of vesicular morphology, as evidenced by a significant decrease in fluorescence intensity toward the center of the spheres. The pyrene fluorescence probe method was used to further certify the coassembly process.26 For the experiments, pyrenelabeled HBPO-star-PEO (HBPO-star-PEO-Py) was synthesized, which can self-assemble into micrometer-sized vesicles named Py-BPs. It is known to us that static excimers form upon contact

Figure 3. Fluorescent spectrum evidence of vesicle fusion. (A) The content mixing assay experiments by measuring the fluorescence of the Tb(DPA)33 complex. (B) The component mixing assay experiments by measuring the changes of IE/IM of pyrene probes.

between two pyrene chromophores and the relative ratio of the fluorescence intensity of the excimer emission (IE at 480 nm) to that of monomer emission (IM at 398 nm) (IE/IM) is related to 2069

dx.doi.org/10.1021/la203857s |Langmuir 2012, 28, 2066–2072

Langmuir

ARTICLE

Figure 4. The 1H NOESY spectrum of Ada-BPs and CD-BPs after mixing them together in D2O for 24 h.

the pyrene distribution in the assemblies.26 According to the result in Figure S3 (Supporting Information), strong excimer emission peaks occurred in the Py-BP solution, which indicated the tight contact of pyrene fluorophores in the vesicle membranes. With the introduction of molecule 2 through the coassembly process, IE/IM decreased accordingly, indicating the increase of the distance between the pyrene fluorophores. In other words, molecule 2 was inserted into the membranes of Py-BPs to dilute the fluorophores, through which the binary AdaBPs were obtained. We previously used the same method to prove the coassembly of molecule 1 into Py-BPs.24 We should note that the Ada groups may locate on the outer layer of the Ada-BPs as shown in Scheme 1 in spite of the hydrophobicity. The DP of the PEO arms in HBPO-star-PEOAda is about 4, which is bigger than that of HBPO-star-PEO (DP = 2). So, after the coassembly of HBPO-star-PEO-Ada and HBPO-star-PEO in water to form Ada-BPs, the Ada groups have more chance to stay in the outer layer of the vesicles. As direct evidence, we supplemented the 1H NMR result of Ada-BP solution in D2O in the Supporting Information (Figure S4), which clearly showed the protons of Ada groups. Such a result indicates that the Ada groups have a high mobility, which is generally related to the outer vesicle shells rather than the inner solid-like hydrophobic layers. Vesicle Aggregation. For the experiments, CD-BPs were mixed with Ada-BPs at a molar ratio between CD/Ada groups of 1:0.95. The solution was turbid at the beginning of the mixing. After 24 h, the solution was lucid, and there was a layer of coacervate at the bottom of the bottle. The coacervate phase was collected for analyses, and it consisted of very large vesicle

aggregates in macroscopic scale according to optical microscopy measurement. Figure 1C shows an optical micrograph (750 μm in length and 610 μm in width) of a part of the vesicle aggregate, which is composed of three-dimensional dense packing of vesicles, especially according to the magnified image as shown in the inset. To our knowledge, it is the largest vesicle aggregate reported up to now. We also noted that there were several big vesicles around 20 μm in the vesicle aggregates (Figure 1C), which were much larger than the major population of the small vesicles (the inset of Figure 1C). It was hypothesized that these big vesicles were formed by the fusion of small ones during the vesicle aggregation process. We further tracked the morphology of the vesicle aggregate after keeping in water for a longer time at the room temperature. For the experiments, red-fluorescent CD-BPs (labeled by rhodamine B) and green-fluorescent Ada-BPs (labeled by dansyl) were mixed together for three days, and many big vesicles formed in the depletion of the smaller ones (Figure 2A) in the vesicle aggregate. The fluorescence overlay micrograph (Figure 2B) shows that the big vesicles are in orange due to the colocalization of red fluorescent chronophers from CD-BPs and green fluorescent chronophers from Ada-BPs, which strongly supports the idea that the big vesicles were formed by the fusion between CD-BPs and Ada-BPs. Real-time trace, content mixing assay, and pyrene excimer assay were used to further characterize the fusion events. Figure 2C shows a real-time fusion process of two giant vesicles. Two vesicles got close first. Then, after undergoing the intermediates of an “8” shape and oblate sphere, the two separate vesicles gradually fused into one vesicle. The whole process 2070

dx.doi.org/10.1021/la203857s |Langmuir 2012, 28, 2066–2072

Langmuir

Figure 5. Fluorescent emission spectra of dansyl chloride in water (curve 1), in CD-BP solution (curve 2), and with the further introduction of the competitive HBPO-star-PEO-Ada (curve 3).

finished within 20 s. Such a vesicle fusion sequence is very similar to that of BPs driven by ultrasoinication.27 The vesicle fusion events happened frequently during the aggregation process. We observed many fusing vesicles with different sizes (Figure S5 in the Supporting Information), which indicated that the vesicle size was not a limit to fusion. The content mixing assay method was based on the formation of the fluorescence-enhanced chelate complex of Tb(DPA)33 .28 For the experiments, CD-BPs were encapsulated with Tb3+ (Tb-CD-BPs), and Ada-BPs were encapsulated with DPA (DPA-Ada-BPs). Both Tb-CD-BPs and DPA-Ada-BPs had very low fluorescence intensity. However, when they were mixed together, the fluorescence intensity increased greatly (Figure 3A) due to the formation of the Tb(DPA)33‑ complex generated through mixing the aqueous vesicle contents, which supported fusion events well. As the control experiments, we mixed Tb-CD-BPs with DPA-CD-BPs or Tb-Ada-BPs with DPA-Ada-BPs, and no obvious changes in the fluorescent intensity were observed, which indicates that the vesicle fusion is triggered by the intervesicular CD/Ada host guest interactions. Subsequently, EDTA was added to the vesicle solution to determine whether there was vesicle content leakage during fusion. EDTA can destroy Tb(DPA)33 and form the much stronger Tb-EDTA complex with low fluorescence. As shown in Figure 3A, a rapid decrease of the fluorescent intensity was observed with the addition of EDTA. Since EDTA can not penetrate into the vesicle lumens of BPs (Figure S6 in the Supporting Information), such a fluorescent decrease must indicate the partial leakage of Tb(DPA)33 from the vesicle lumens during the fusion process. Pyrene excimer experiments (Figure 3B) were further performed to prove the mixing of vesicle components due to the vesicle fusion.29 When CD-BPs labeled by the pyrene molecules (Py-CD-BPs) were mixed with Ada-BPs, IE/IM decreased with time. This phenomenon indicated the insertion of the components from Ada-BPs into those of Py-CD-BPs to dilute the pyrene chronophers due to the vesicle fusion. As the control experiments, Py-CD-BPs were mixed with CD-BPs, or the Py-Ada-BPs were mixed with AdaBPs, and no obvious change of IE/IM occurred. Thus, both the content mixing assay and the component mixing assay experiments prove that the fusion between CD-BPs and Ada-BPs is driven by the CD/Ada molecular recognition interactions.

ARTICLE

We also noted two characteristics for the vesicle fusion. First, we found that the vesicle fusion occurred so frequently that almost all the small vesicles in the aggregates fused into larger ones (Figure 2A,B). In our previous work, we also found vesicle fusion during the aggregation of CD-BPs and Azo-BPs;24 however, the fusion was not as frequent as that in the present work. We attributed such widely occurring vesicle fusion events to the stronger intervesicular host guest interactions between CD and Ada groups. Second, we found that vesicle fusion could not proceed infinitely. The vesicle aggregates as shown in Figure 2A could be kept for at least 1 month without evident changes, and the fusion cycles seemed to be stopped totally. We still had no evidence to disclose the mechanism inside it. We propose that at some stage all the CD and Ada units within the vesicles form intravesicular complexes due to the sequential fusion cycles, and thus the vesicles can no longer recognize each other due to the failure to form intervesicular CD/Ada molecular recognition, which may lead to the cessation of vesicle fusion cycles (Figure S7 in the Supporting Information). In Scheme 1, we propose that the vesicle aggregation process is triggered by the intervesicular host guest molecular recognition interaction between β-CD units and Ada units. To directly certify this point, 1H NOESY measurement of CD-BPs and Ada-BPs was performed in D2O at 25 °C. A series of identifiable NOE cross peaks between the H-3 and H-5 protons on the inside of the β-CD torus and the key protons on the Ada units were observed (Figure 4). The result indicated that the Ada groups entered the hydrophobic cavity of the β-CD rings to form the host guest inclusion complexes. Since the Ada or CD groups were located in the vesicles, such a 1H NOESY result indicated the formation of intervesicular interactions through the CD/Ada recognitions, which triggered the vesicle aggregation from CD-BPs and AdaBPs. In addition, a competitive complexation experiment was designed to further prove the driving force hypothesis. As a typical fluorescent group, dansyl is very sensitive to the environment. It exhibits strong fluorescence in a hydrophobic environment and weak fluorescence in bulk water. When dansyl chloride was added into CD-BP aqueous solution, an enhanced fluorescence appeared due to the complexation of dansyl groups into the hydrophobic β-CD torus (Figure 5). Subsequently, HBPOstar-PEO-Ada was added into the solution system, and the fluorescence intensity was weakened since dansyl groups were replaced by Ada groups due to the stronger complex ability between Ada and β-CD groups than that between dansyl and β-CD groups. Thus, both the 1H NOESY measurement and competitive complexation experiments confirmed the host guest CD/Ada molecular recognition interactions between CD-BPs and Ada-BPs, which drove the large-scale vesicle aggregation.

’ CONCLUSION In conclusion, as a cytomimetic model of the cell aggregation, we have prepared macroscopic vesicle aggregates by using cellsize hyperbranched polymer vesicles as the building blocks and intervesicular multivalent host guest molecular recognitions between β-CD units and Ada units as the driving force. Membrane fusion events occurred frequently during the vesicle aggregation process according to double-labeling fluorescent assay, real-time observation, content mixing assay, and component mixing assay. These results indicate that the cell-sized hyperbranched polymer vesicles have great potential in cytomimetic vesicle aggregation, and a strong intervesicular interaction 2071

dx.doi.org/10.1021/la203857s |Langmuir 2012, 28, 2066–2072

Langmuir such as the β-CD/Ada host guest interaction will lead to a very large-scale vesicle aggregate and widely occurring vesicle fusion events. We anticipate that such porous macroscopic vesicle aggregates may find applications in tissue engineering materials.

’ ASSOCIATED CONTENT

bS

Supporting Information. Additional details as described in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]; tel.: +86-21-54742665; fax: +8621-54741297.

’ ACKNOWLEDGMENT We thank the National Basic Research Program (2009CB930400, 2012CB821500), the National Natural Science Foundation of China (21074069, 20874060, 50873058 and 91127047), the Foundation for the Author of National Excellent Doctoral Dissertation of China, the Open Project of State Key Laboratory of Supramolecular Structure and Materials (SKLSSM201118), and the Fok Ying Tung Education Foundation (No. 114029). ’ REFERENCES (1) (a) Hynes, R. O.; Lander, A. D. Cell 1992, 68, 303–322. (b) de Saint Basile, G.; Menasche, G.; Fischer, A. Nat. Rev. Immunol. 2010, 10, 568–579. (2) (a) Menger, F. M.; Keiper, J. S. Adv. Mater. 1998, 10, 888–890. (b) D€obereiner, H. G. Curr. Opin. Colloid Interface Sci. 2000, 5, 256–263. (c) Discher, D. E.; Eisenberg, A. Science 2002, 29, 967–973. (d) Hammer, D. A.; Robbins, G. P.; Haun, J. B.; Lin, J. J.; Qi, W.; Smith, L. A.; Ghoroghchian, P. P.; Therien, M. J.; Bates, F. S. Faraday Discuss. 2008, 139, 129–141. (3) Egli, S.; Schlaad, H.; Bruns, N.; Meier, W. Polymers 2011, 3, 252–280. (4) Menger, F. M.; Angelova, M. I. Acc. Chem. Res. 1998, 31, 789–797. (5) Chiruvolu, S.; Walker, S.; Israelachvili, J.; Schmitt, F. J.; Leckband, D.; Zasadzinski, J. A. Science 1994, 264, 1753–1756. (6) Richard, A.; Marchi-Artzner, V.; Lalloz, M. N.; Brienne, M. J.; Artzner, F.; Gulik-Krzywicki, T.; Guedeau-Boudeville, M. A.; Lehn, J. M. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 15279–15284. (7) (a) Menger, F. M.; Seredyuk, V. A.; Yaroslavov, A. A. Angew. Chem., Int. Ed. 2002, 41, 1350–1352. (b) Menger, F. M.; Seredyuk, V. A. J. Am. Chem. Soc. 2003, 125, 11800–11801. (c) Menger, F. M.; Zhang, H. L. J. Am. Chem. Soc. 2006, 128, 1414–1415. (8) (a) Marchi-Artzner, V.; Jullien, L.; Gulik-Krzywicki, T.; Lehn, J.-M. Chem. Commun. 1997, 117–118. (b) Marchi-Artzner, V.; GulikKrzywicki, T.; Guedeau-Boudeville, M. A.; Gosse, C.; Sanderson, J. M.; Dedieu, J. C.; Lehn, J. M. ChemPhysChem 2001, 2, 367–376. (c) MarchiArtzner, V.; Brienne, M.-J.; Gulik-Krzywicki, T.; Dedieu, J.-C.; Lehn, J.-M. Chem.—Eur. J. 2004, 10, 2342–2350. (9) Chan, Y.-H. M.; van Lengerich, B.; Boxer, S. G. Biointerphases 2008, 3, FA17–FA21. (10) C. Constable, E.; Mundwiler, S.; Meier, W.; Nardin, C. Chem. Commun. 1999, 1483–1484. (11) (a) Sideratou, Z.; Foundis, J.; Tsiourvas, D.; Nezis, I. P.; Papadimas, G.; Paleos, C. M. Langmuir 2002, 18, 5036–5039. (b) Tsogas, I.; Tsiourvas, D.; Nounesis, G.; Paleos, C. M. Langmuir 2005, 21, 5997–6001. (c) Paleos, C. M.; Tsiourvas, D.; Sideratou, Z. ChemBioChem 2011, 12, 510–521.

ARTICLE

(12) Ma, M.; Gong, Y.; Bong, D. J. Am. Chem. Soc. 2009, 131, 16919–16926. (13) (a) Mart, R. J.; Liem, K. P.; Wang, X.; Webb, S. J. J. Am. Chem. Soc. 2006, 128, 14462–14463. (b) Kwan, P. L.; Mart, R. J.; Webb, S. J. J. Am. Chem. Soc. 2007, 129, 12080–12081. (14) (a) Lim, C. W.; Ravoo, B. J.; Reinhoudt, D. N. Chem. Commun. 2005, 5627–5629. (b) Nalluri, S. K. M.; Ravoo, B. J. Angew. Chem., Int. Ed. 2010, 49, 5371–5374. (c) Voskuhl, J.; Fenske, T.; Stuart, M. C. A.; Wibbeling, B.; Schmuck, C.; Ravoo, B. J. Chem.—Eur. J. 2010, 16, 8300–8306. (15) Lee, H.-K.; Park, K. M.; Jeon, Y. J.; Kim, D.; Oh, D. H.; Kim, H. S.; Park, C. K.; Kim, K. J. Am. Chem. Soc. 2005, 127, 5006–5007. (16) Discher, B. M.; Won, Y. Y.; Ege, D. S.; Lee, J.; Bates, F. S.; Discher, D. E.; Hammer, D. A. Science 1999, 284, 1143–1146. (17) Bermudez, H.; Brannan, A. K.; Hammer, D. A.; Bates, F. S.; Discher, D. E. Macromolecules 2002, 35, 8203–8208. (18) Photos, P. J.; Bacakova, L.; Discher, B.; Bates, F. S.; Discher, D. E. J. Controlled Release 2003, 90, 323–334. (19) (a) Lin, J. J.; Ghoroghchian, P.; Zhang, Y.; Hammer, D. A. Langmuir 2006, 22, 3975–3979. (b) Robbins, G. P.; Saunders, R. L.; Haun, J. B.; Rawson, J.; Therien, M. J.; Hammer, D. A. Langmuir 2010, 26, 14089–14096. (20) Jin, H. B.; Zhou, Y. F.; Huang, W.; Yan, D. Y. Langmuir 2010, 26, 14512–14519. (21) Smart, T. P.; Fernyhough, C.; Ryan, A. J.; Battaglia, G. Macromol. Rapid Commun. 2008, 29, 1855–1860. (22) (a) Zhou, Y. F.; Yan, D. Y. Angew. Chem., Int. Ed. 2004, 43, 4896–4899. (b) Zhou, Y. F.; Yan, D. Y.; Dong, W. Y.; Tian, Y. J. Phys. Chem. B 2007, 111, 1262–1270. (23) (a) Zhou, Y. F.; Huang, W.; Liu, J. Y.; Zhu, X. Y.; Yan, D. Y. Adv. Mater. 2010, 22, 4567–4590. (b) Paleos, C. M.; Tsiourvas, D.; Sideratou, Z. Gene Ther. Mol. Biol. 2007, 11, 117–132. (24) Jin, H. B.; Zheng, Y. L.; Liu, Y.; Cheng, H. X.; Zhou, Y. F.; Yan, D. Y. Angew. Chem., Int. Ed. 2011, 50, 10352–10356. (25) (a) Breslow, R.; Czarniecki, M. F.; Emert, J.; Hamaguchi, H. J. Am. Chem. Soc. 1980, 102, 762–770. (b) Guo, M. Y.; Jiang, M.; Zhang, G. Z. Langmuir 2008, 24, 10583–10586. (c) Chen, Y.; Zhang, Y. M.; Liu, Y. Chem. Commun. 2010, 5622–5633. (d) Liu, Y.; Yu, Z. L.; Zhang, Y. M.; Guo, D. S.; Liu, Y. P. J. Am. Chem. Soc. 2008, 130, 10431–10439. (e) Ma, X.; Tian, H. Chem. Soc. Rev. 2010, 39, 70–80. (26) Winnik, F. M. Chem. Rev. 1993, 93, 587–614. (27) Zhou, Y. F.; Yan, D. Y. J. Am. Chem. Soc. 2005, 127, 10468– 10469. (28) Wilschut, J.; Duzgunes, N.; Fraley, R.; Papahadjopoulos, D. Biochemistry 1980, 19, 6011–6021. (29) Somerharju, P. Chem. Phys. Lipids 2002, 116, 57–74.

2072

dx.doi.org/10.1021/la203857s |Langmuir 2012, 28, 2066–2072