Formation and Characteristics of Lipid Blended ... - ACS Publications

Subscriber access provided by YORK UNIV

Interface Components: Nanoparticles, Colloids, Emulsions, Surfactants, Proteins, Polymers

Formation and Characteristics of Lipid Blended Block Copolymer Bilayers on Solid Support Investigated by Quartz Crystal Microbalance and Atomic Force Microscope Mudassar Mumtaz Virk, Benedikt Hofmann, and Erik Reimhult Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.8b03597 • Publication Date (Web): 23 Dec 2018 Downloaded from http://pubs.acs.org on December 26, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Formation and Characteristics of Lipid Blended Block Copolymer Bilayers on Solid Support Investigated by Quartz Crystal Microbalance and Atomic Force Microscope Mudassar Mumtaz Virk, Benedikt Hofmann, Erik Reimhult* Institute for Biologically Inspired Materials, Department of Nanobiotechnology, University of Natural Resources and Life Sciences Vienna, Muthgasse 11, 1190 Vienna, Austria

KEYWORDS. Lipid membrane, block copolymer membrane, vesicle fusion, solvent inversion, quartz crystal microbalance with dissipation monitoring (QCM-D), force spectroscopy, atomic force microscopy (AFM)

ACS Paragon Plus Environment

1

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 33

ABSTRACT. Liposomes grafted with polymer have long been used in drug delivery applications and block copolymersomes have emerged as attractive and more robust alternatives for both drug delivery and artificial organelle applications. Hybrid membranes that could combine the respective advantages of fluid lipid and robust polymer bilayers are an attractive and enticing alternative. The properties of membranes in amphiphile vesicles are challenging to study and many applications benefit from surface-based access to the membrane. We therefore explore the self-assembly and mechanical properties of supported hybrid bilayers (SHBs) composed of polybutadiene-block-poly(ethylene

oxide)

block

copolymers

and

zwitterionic

phosphatidylcholine lipids on SiO supports. Quartz crystal microbalance with dissipation 2

monitoring (QCM-D) measurements show that formation of SHB on SiO by vesicle fusion 2

depends on the mass fractions of lipids and block copolymers. AFM was used to study the microscopic mixing of lipids in the SHB to reveal that lipid phase separation is not observed in SHBs. Force spectroscopy was performed to extract information about thickness and mechanical properties of the hybrid membranes. SHBs are shown to combine the properties of lipid membranes and polymer brushes and that the tip force required to rupture the membrane decreases and the bilayer thickness increases as the block copolymer fraction is increased.


2

Page 3 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

INTRODUCTION All living cells are protected with a thin semipermeable membrane which plays a vital role in many biological processes, including in cell recognition, selective ion transfer, signaling, adhesion and fusion. These membranes are very complex in their native state comprising 1

phospholipids, glycolipids and various proteins. The high complexity of the composition and 2

structure of native cell membranes makes it difficult to extract direct physical understanding of their properties in the native state. This has inspired the development of artificial amphiphilic membranes that capture the main aspects of cell membranes and allows for the study of single components in biological membranes in a controlled environment; prominent examples are model membranes in the form of vesicles (liposomes) and solid supported lipid bilayers (SLB).

3-4

An additional motivation for research into artificial membrane systems is that due to their biocompatibility and ability to mimic advanced biological functions such as controlled permeability, they have in their own right become an important tool for biosensing, drug delivery and compartmentalization of (bio)chemical reactions. , SLBs are more stable than freestanding 5-9

membranes, large and giant vesicles, as they are supported by a solid substrate. They also allow for the use of a wide range of analytical and surface sensitive techniques to study their properties and interactions with external and embedded molecules. For example, structural studies, mobility and fluidity of lipid bilayers have been investigated extensively using AFM, ellipsometry, neutron reflectometry 13

13-15

10-13

QCM-D,

10-14

and fluorescence microscopy . SLBs have also been used 14, 16

to investigate the interaction and binding mechanism of e.g. peptides, proteins and antibacterial drugs with membranes.

17-20

Physical and chemical stability, such as fusion with other membranes and enzymatic degradation is an issue that requires improvement for pure lipid systems in all formats and


3

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 33

applications. Grafting of polymers such as poly(ethylene glycol) and inclusion of lipids with long fatty acid chains and cholesterol to densify the membrane are often applied approaches to reduce or slow down degradation. Although, this is reported to increase stability of e.g. liposomal systems, it also sacrifices some aspects such as fluidity and tunable permeability of the membrane. A common goal of supported lipid membrane research is the inclusion of membrane proteins for detailed study or use in biosensors. However, in case of SLBs comprising pure phospholipids, there is only a very thin layer of water between the lower leaflet of the lipid bilayer and the substrate (1-2nm),

5,

21

which frequently leads to strong interaction of the

incorporated membrane proteins with the substrate. This problem typically makes proteins embedded in SLBs unsuitable for functional studies. Various approaches to remedy this have been suggested, such as nanopatterned supports, polymer-tethered and polymer-grafted supported lipid bilayers.

5, 9

Amphiphilic block copolymer mimics of cell membranes is a highly interesting alternative for all research areas described above. They exhibit superior mechanical stability to lipid membranes

22-25

and polymersomes have lower water permeability than liposomes. This is mainly 26

a result of the hydrophobic interior of block copolymer membranes being thicker than for lipid membranes. Thus, this improvement in mechanical properties comes at the expense of a thickness mismatch with biological species that can hinder successful incorporation of membrane proteins. However, the Palivan and Meier groups have shown that robust membrane channels can be reconstituted in pure block copolymer membranes if the block copolymer is chosen such that it allows strong deformation and relaxation of the line tension around the incorporated protein.

27-29

Additional examples of reconstitution of proteins in polymersomes were


4

Page 5 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

recently reviewed by Habel and Beales.

30-31

However, pure polymer-based membranes are likely

not always compatible with conserved function of many membrane proteins as they require lipid native environments to retain their functional conformation and specific functions.

32-33

By blending lipids and block copolymers, one can construct lipid environments for integration of proteins or making use of other lipid-specific functions that are embedded in a more robust polymer membrane to improve overall stability and structure. The physical state and properties of these hybrid systems can be tuned by varying the fractions of polymers and lipids. The mixing behavior of polymers and lipids in hybrid systems depends on several factors including mass fractions, length and stiffness of the polymer chains, and the melting temperature of the lipids.

34-35

Due to their combination of stability and biocompatibility, hybrid membranes are receiving increasing attention for protein reconstitution

36-37

to PEGylated liposomes in drug delivery.

35, 38-40

as well as for their potential use as an alternative

In most of the studies, the mixing behavior of lipids

and block copolymer was investigated in vesicles and to date only few studies have explored physiochemical properties of hybrid bilayers on solid supports (SHB), where other techniques can be used for the investigation to, e.g. directly measure mechanical properties or kinetics of interaction with and molecular diffusion in hybrid membranes. Exemplifying this, in a rare study Parikh and coworkers studied the lateral mobility of 1-palmitoyl-2-oleoyl-sn-glycero-3phosphatidylcholine (POPC) in SHBs comprising polybutadiene-block-poly(ethylene oxide) (MW 1.8 kDa) and POPC at different mass fractions of POPC.

41

It is a prerequisite to be able to form SHBs on suitable surfaces such as silicon oxide (microscopy slides, silicon wafers, biosensor chips, etc.) to investigate and to make use of them. The preferred method to form SLBs is through self-assembly by adsorption, rupture and fusion of liposomes, and a similar straightforward approach to assemble SHBs from mixed composition


5

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 33

lipopolymersomes is desired. In this work, we aim to investigate the kinetics of formation of SHBs composed of polybutadiene(1200 Da)-block-poly(ethylene oxide)(1000 Da) (PBD-b-PEO) blended with different weight fractions (30 % w/w, 50 % w/w, 70 % w/w) of 1,2-dipalmitoyl-snglycero-3-phosphocholine (DPPC) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphatidylcholine (POPC) lipids by vesicle fusion on silicon oxide surfaces using Quartz Crystal Microbalance with Dissipation monitoring. The homogeneity and mechanical properties of the formed SHBs were further investigated by atomic force microscopy and force spectroscopy. A strong dependence on composition is observed for both self-assembly and resulting mechanical properties. EXPERIMENTAL SECTION Materials: 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) and 1,2-dipalmitoyl-snglycero-3-phosphocholine (DPPC) were purchased from Avanti Polar Lipids and used as received without further purification. Chloroform, tetrahydrofuran (THF), Phosphate buffered saline (PBS) tablets, HEPES, H SO and Hellmanex III were purchased from Sigma-Aldrich. 2

4

Polybutadiene(1200 Da)-block-poly(ethylene oxide)(1000 Da) (M = 2200g/mole, Ð = 1.09) w

(PBD-b-PEO) was obtained from Polymer Source Inc. and used as received. Silicon wafers (5 mm × 5mm) were purchased from TED PELLA, INC. Preparation of hybrid vesicles: Hybrid vesicles were produced by the solvent inversion and sonication method as described before. In brief, PBD-b-PEO was dissolved in chloroform at a 35

concentration of 25 mg/ml. The block copolymer was mixed with the respective weight percentage of lipid (2.5 mg total mass of lipid and block copolymer). The chloroform was then removed using rotavapor for 30 min at 60 °C and dried under high vacuum (0.05 mbar) overnight. 200 µL of THF were added to the lipid and block copolymers cake. The flask


6

Page 7 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

containing lipid and block copolymers in THF was connected to a rotary evaporator and submerged in a water bath (𝑇"#$% > 𝑇' , where 𝑇' is the transition temperature between the gel and liquid crystal phases of the lipid membrane) at 1 atm and stirred for 30 seconds to mix the lipid and block copolymers thoroughly. The suspension of lipid and block copolymer in THF added drop-wise to 1 ml of (preheated 60 °C for DPPC to be above 𝑇' ) filtered 10 mM PBS (for AFM measurements) or 10 mM HEPES (for QCM-D measurements) with 150mM NaCl at pH 7.4 under magnetic stirring. This produces a turbid dispersion, which was kept at 60 °C under magnetic stirring for 15-20 min. The turbid dispersion was then immersed in an ultrasonic bath (Elmasonic P30H) and sonicated for 30 minutes (𝑇"#$% > 𝑇' ) at 37 Hz and 320 W to form a translucent dispersion. After sonication the dispersion was further diluted to the final concentration of 0.5 mg/ml. Dynamic Light Scattering: Characterization of the hydrodynamic size of the vesicles was performed on a Malvern Instruments Zetasizer Nano-ZS instrument with backscattered detection at an angle of 173°. The built-in software using the CONTIN algorithm was used for extracting size distributions from the correlation curves. Quartz Crystal Microbalance with Dissipation monitoring (QCM-D): QCM-D

42

measurement

were performed using a Q-sense E4 flow module (Q-Sense) with silicon oxide coated sensors (QSX 303, Q-Sense). All the QCM-D measurements were performed in 10 mM HEPES with 150mM NaCl at pH 7.4 (HEPES buffer). The HEPES buffer was first degassed and filtered through 0.2μm PVDF filters. The HEPES buffer was flowed at 50 μl/min in the Q-Sense E4 to record a baseline. The measurements were performed at room temperature (25 °C) for hybrid vesicles containing POPC and at 37 °C for hybrid vesicles containing DPPC. After recording a stable baseline, the dispersion of vesicles was injected at a final concentration of 0.05 mg/ml.


7

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 33

The formation of supported membranes was monitored by recording the response of the quartz crystal resonance frequency and energy dissipation in shear oscillation mode as function of time using odd overtones (3-13). The data presented in the figures was the lowest overtone providing measurements without instrument noise normalized to the fundamental resonance frequency of 4.95 MHz; this was mostly the 3rd overtone. Atomic force microscopy (AFM): AFM was performed using a Nanowizard AFM (JPK). All the imaging and force spectroscopy curves were performed at room temperature using sharp silicon nitride tips (DNP-S10) from Bruker. The nominal radius of the tips was 10 nm and the nominal spring constant was 0.24 N/m. All the images were recorded in intermittent contact (tapping) mode at a resolution of 512 × 512 pixels and with a scan rate of 1 Hz. The images were processed and analysed using the Nanowizard JPK process software. Prior to performing force spectroscopy on the bilayers, the spring constant of cantilevers was calibrated on bare silicon oxide substrate using the thermal noise method. Force-distance curves was recorded on different 43

randomly selected positions on substrates at tip speeds ranging from 0.6 μm/s to 1 μm/s. All the force-distance curves were converted into text files and plotted and analysed in self-written Matlab code. Formation of supported hybrid bilayers (SHB) for AFM: All the SHBs for AFM were done in 10 mM PBS with 150 mM NaCl at pH 7.4. Polished silicon wafer substrates (5 mm × 5mm) were treated with UV/ozone for 30 min. The treatment yielded oxidized surfaces free of organic contaminants. The substrates were then attached to the petri dishes with a metal disc underneath. They were further cleaned by soaking in 2 vol% Hellmanex solution for 30 min. The substrates were then thoroughly rinsed with Milli-Q water before soaking them in 4M H SO for 5 min and 2

4

then again thoroughly rinsing them with Milli-Q water. 200-400 μl of vesicle solution (0.5


8

Page 9 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

mg/ml) was added to the substrates for 40 min. After 40 min, the petri dishes were filled with PBS and rinsed further with 20 ml of PBS. In case of vesicles blended with DPPC, the petri dishes were filled with water and heated at 55 °C for 20 min to make sure the substrate was above the melting temperature of DPPC (𝑇' = 41 °C). The hot water was then removed from the petri dishes and the vesicle solution preheated to 55 °C was added to the substrate and incubated for 40 min at 55°C. The samples were also rinsed with PBS preheated to 55 °C. The samples were cooled down at room temperature before measurements. All measurements were performed at room temperature. RESULTS AND DISCUSSION First, we used Quartz Crystal Microbalance with Dissipation monitoring (QCM-D) to investigate the self-assembly kinetics and hydration of supported hybrid membranes formed by vesicle fusion on silicon oxide surfaces. QCM-D is a standard technique to study vesicle adsorption and rupture kinetics due to its sensitivity to changes in hydration and viscous losses; these are strongly influenced by conformation and structure of macromolecules and selfassembled aggregates adsorbed to the surface.

44-45

The hydration of adsorbed layers is included in

the mass response. The viscous and hydrodynamic losses of water measured by the dissipation are low for water coupled in dense and rigid films such as supported lipid bilayers, while they are high for extended water-rich nanostructures such as liposomes, polymer brushes and hydrogels. This makes it relatively easy to distinguish formation of supported planar bilayers of lipids from adsorption of intact liposomes.

45-46

Before the investigation of formation of bilayers by QCM-D,

dynamic light scattering (DLS) was performed to ensure the formation of vesicles of different compositions and to measure their hydrodynamic size distribution as shown in Figure 1. The size distributions are mainly monomodal and, most importantly, there is no indication of small


9

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 33

micelle formation for samples containing a fraction of PBD-b-PEO block copolymer. However, the size distributions for the hybrid vesicles of POPC and PBD-b-PEO are rather broad and in the case of 30% and 70% POPC content they suggest that the size distribution could be bimodal. It can therefore not be ruled out that worm-like micelles are formed in addition to vesicles. Wormlike micelles were observed for large hydrophobic mismatch between triblock copolymers and DPPC by Le Meins and coworkers but were not observed for diblock copolymer/DPPC blends that were more similar to our system. However, recently Lonetti and coworkers found worm47

like micelles for extruded blends of diblock PBD-b-PEO/DPPC blends that formed giant unilamellar vesicles by electroformation, which introduces the presence of worm-like micelles 48

also for bilayer-forming mixtures of diblock copolymer and lipid a possibility. We emphasize, that in our previous work on hybrid vesicles, including encapsulation and release studies, no presence of such worm-like vesicles were observed; they were not observed in a study of 38

formation and permeability of large unilamellar PBD-b-PEO/POPC hybrid vesicles by Lim et al. Broad and bimodal size distributions of diblock copolymersomes were, however, observed 48

by us for formation of hybrid vesicles by bath sonication and they are often observed also for 49

single-component amphiphiles when bath sonication is used. This can be observed also in Fig. 1b for the 100% DPPC liposomes (cf. Fig. 1b). In summary, while mainly monomodal vesicles are formed, the bath sonicated blends could also contain bimodal vesicle size distributions or even contain some worm-like micelles.


10

Page 11 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 1. Intensity weighted size distributions of hybrid vesicles of lipid mixed with PBD-b-PEO used in QCM-D measurements measured by DLS and shown for different weight fractions of (a) POPC and (b) DPPC.

Figure 2(a-d) shows the QCM-D frequency and dissipation response of the 3 overtone during rd

the formation of SHB for different weight fractions of POPC and DPPC. Higher overtones show a corresponding behavior. Upon injection of the vesicle solution, initially a large decrease in frequency (increase in mass) and increase in dissipation is observed due to adsorption of intact vesicles on the surface. It is followed by an increase in frequency and decrease in dissipation. This is the typical signature of surface-induced vesicle rupture and formation of a supported membrane. The loss of mass is due to the release of trapped solvent from inside and between the 45

vesicles. The release of water and densification of the layer also leads to a more rigid (bi)layer with lower dissipation than that of adsorbed vesicles.

44-45

The final absolute values can be used to

calculate the adsorbed mass on the surface, including the coupled water. For sufficiently rigid layers, the frequency is multiplied with a numerical constant related to the resonance frequency of the oscillator to obtain the adsorbed mass, following the so-called Sauerbrey relation.

50-51

A pure

POPC or DPPC supported lipid bilayer has a |∆𝑓| of 26-27 Hz and ∆𝐷~1001 . This includes a


11

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 33

thin hydration layer between the membrane and the substrate, as well as contributions from nonruptured liposomes. We can observe this behavior and quantitative response for the 100% lipid SLBs formed in Figure 2, although a second phase of liposome adsorption seemed to occur on top of the SLB for DPPC; the latter was mostly stable after rinsing. The dissipation is very sensitive to defects of liposomes adsorbed onto or within the SLB, also when the coverage of such adsorbed vesicles is low.

52


12

Page 13 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 2. Examples of QCM-D frequency (a, b) and dissipation (c, d) shifts as function of time for vesicle adsorption and supported hybrid bilayer formation from vesicles composed of PBD-b-PEO blended with different weight fractions (30%, 50%, 70% and 100% w/w) of POPC and DPPC. (e) and (f) show the mean values of the frequency and dissipation shifts, respectively. The error bars show the sample size-corrected standard error of the mean of three to four different measurements. P-values calculated from one-way ANOVA test showed the statistical significance of differences between frequency (P values less than 0.00015) and dissipation shifts (P values less than 0.025) for all compositions. 53


13

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 33

When increasing amounts of PBD-b-PEO are incorporated in the vesicles (decreasing lipid weight fractions), we observe an increase in the final frequency shift (Figure 2e) for mixtures with both POPC and DPPC lipids. All hybrid vesicles display a maximum in the dissipation. Combined with a minimum in frequency this is a signature of vesicle rupture and supported membrane formation. A distinct frequency minimum is not observed for 30 % w/w lipids, for which also the dissipation maximum is very weak. The larger frequency response of the final layer observed as the polymer fraction is increased combined with the signature of vesicle rupture and supported membrane formation can be explained by more water coupled by hybrid bilayers compared to lipid bilayers. This is expected because of the additional PEO brush attached to the membrane top and bottom. The predominant rupture of adsorbed vesicles to form a supported hybrid bilayer was confirmed by tapping mode AFM imaging (Figure 3) and is further discussed below. Note that the minimum of frequency ( Δfmin ) and the maximum of dissipation ( ΔDmax ) are lower and shifted to shorter exposure times compared to for pure liposomes as PBD-b-PEO is added to the vesicles (Figure 2(a to d)). This indicates that lower surface coverage of intact vesicles is required to induce rupture and form a planar bilayer when the vesicles are blended with polymers. A difference in diameter between vesicles of different composition (cf. Figure 1) makes a clear trend with composition obscured. However, a difference in vesicle size cannot explain that higher vesicle surface coverage apparently is needed for SLB formation from pure liposomes compared to for formation of SHB from hybrid vesicles. Since the PEO block of the polymer only has weak attraction to the surface, the faster SHB formation implies that hybrid vesicles adsorbed to the surface can withstand a lower increase in surface tension than pure liposomes before surface-induced rupture occurs. Possibly, this is due to lipid/polymer phase


14

Page 15 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

separation in the adsorbed vesicles and the formation of domain boundaries. Alternatively, it can be due to the higher stiffness of block copolymer membranes, which renders the vesicles brittle and facilitate rupture when they are subject to the surface-induced deformation. The surface morphology of the supported hybrid bilayers was visualized by AFM using tapping mode at room temperature. Figure 3 shows the images of SHB formed by vesicle fusion composed of PBD(1200 Da)-b-PEO(1000 Da) blended with different weight fractions of POPC and DPPC. Supported hybrid bilayers composed of 50% w/w and 70% w/w POPC are very smooth with defect-free surfaces, as shown in Figure 3 (a-b), and with a close similarity to pure POPC and DPPC supported membranes shown in Fig. S1. However, defects were observed for 30% w/w POPC (Figure 3c) and the depth of defects measured from height profile is around 7-8 nm. The QCM-D data for 30% w/w POPC showed final values that compared to the other results are compatible with formation of a SHB. However, the weak minima and maxima in the frequency and dissipation, respectively, could be interpreted as the presence of intact vesicles adsorbed to the surface as well. The atomic force micrographs in Figure 3 show that predominantly a membrane with the thickness of a block copolymer membrane is formed, but the defects could be due to non-ruptured vesicles or worm-like micelles that were removed during scanning. The difficulty of imaging adsorbed vesicles by AFM without removing or rupturing the vesicles was reported previously,

13, 45, 54

although defects of DPPC liposomes adsorbed

within or on the lipid bilayer (cf. QCM-D results above) are visible in Fig. S1(a). If worm-like micelles are present, they are also likely to be removed by interaction with the AFM tip due to their even weaker adhesion to the surface.


15

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 33

Figure 3. AFM imaging in tapping mode at room temperature of supported hybrid bilayers formed by fusion of vesicles blended with (a) 70% w/w POPC (b) 50% w/w POPC (c) 30% w/w POPC (d) 70% w/w DPPC (e) 50% w/w DPPC (f) 30% w/w DPPC. The height profiles corresponding to the white lines in the micrographs are shown underneath each image. All images were acquired at room temperature, at which POPC is in the fluid phase and DPPC in the gel phase.

Supported hybrid bilayers formed with DPPC showed defects in the atomic force micrographs for all concentrations of lipid (Figure 3d-f). The irregular shape of the imaged defects could


16

Page 17 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

reflect either the shape of worm-like micelles or adsorption of multiple vesicles that were removed by the tip during imaging; they could also simply reflect the restricted and irregular spreading of supported membranes formed by vesicle fusion on the surface around such defect sites.

However, it should be noted that the AFM measurements were performed at room

45, 55

temperature. At room temperature DPPC is in the gel phase, which leads to reduction of membrane area and increase in lipid membrane thickness compared to the condition at higher temperature at which the SLB/SHB are formed. The SHB composed of 50% (Figure 3c) and 70% w/w of DPPC (Figure 3d) show scattered defects from which a thickness of the SHB of 56nm can be determined. Lower SHB surface coverage was observed for the 30% w/w of DPPC SHB compared to the other weight fractions of DPPC (Figure 3f). Due to the reduction of membrane area, it is highly likely that the defects seen at higher lipid fraction are due to this shrinking combined with the lack of long-range mobility of amphiphiles in SHB as well as 41, 56

surface pinning of the membrane.

45, 55

The SHB with 30% DPPC fraction shows a very high defect

intensity. The latter large defect areas are again likely to have been populated by adsorbed vesicles or worm-like micelles which were removed by the AFM tip during imaging. That a higher such defect intensity is observed for lower lipid fraction indicates that the lipid fraction is important for SHB formation. In our case, vesicles with 100% block copolymer did not reproducibly self-assemble into supported membranes. This contrasts with previous works in which block copolymer bilayers were self-assembled from polymer micelles

56

and on

hydrophobic/hydrophilic patterned surfaces. It is possible that the difference in block copolymer 41

composition and morphology (cf. ) and the use of partly hydrophobic patterns and lower 56

resolution characterization techniques (cf. ) explain these respective differences. 41

Blends of polymers and lipids can either mix homogeneously or phase separate to form


17

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 33

separate domains of lipids and polymers. Their mixing behavior depends on the physical properties of polymer chains, the melting temperature of the lipids and the relative volume fractions of lipids and polymers as reviewed by Le Meins et al. PBD-b-PEO block copolymers 34

in the molecular weight range used in this work mixed with lipid are known to phase separate in giant vesicles when the lipid component is in the gel phase.

35, 40

Thus, homogeneous distributions

are expected in SHBs composed of different weight fractions of POPC as the imaging was performed at room temperature (22 °C), which is above the melting temperature of POPC (𝑇' = −2 °C), but micron-scale phase separation is expected to be observed in membranes containing DPPC (𝑇' = 41 °C). However, no phase separation was observed by AFM for any of the SHB compositions, as shown in Fig. 2. This indicates that either a homogeneous mixture of lipids and polymers was obtained across the whole hybrid bilayer or that phase separation occurs at very small length scales (

Formation and Characteristics of Lipid Blended ... - ACS Publications

Recommend Documents