Article Cite This: Langmuir 2019, 35, 8748−8757
pubs.acs.org/Langmuir
Black Lipid Membranes: Challenges in Simultaneous Quantitative Characterization by Electrophysiology and Fluorescence Microscopy Maria Tsemperouli,† Esther Amstad,‡ Naomi Sakai,† Stefan Matile,† and Kaori Sugihara*,† †
School of Chemistry and Biochemistry, University of Geneva, CH-1211 Geneva, Switzerland Institute of Materials, Ecole Polytechnique Fédérale de Lausanne, 1015 Lausanne, Switzerland
‡
Downloaded via KEAN UNIV on July 19, 2019 at 15:20:12 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
S Supporting Information *
ABSTRACT: Horizontal black lipid membranes (BLMs) enable optical microscopy to be combined with the electrophysiological measurements for studying ion channels, peptide pores, and ionophores. However, a careful literature review reveals that simultaneous fluorescence and electrical recordings in horizontal BLMs have been rarely reported for an unclear reason, whereas many works employ bright-field microscopy instead of fluorescence microscopy or perform fluorescence imaging and electrical measurements one after another separately without truly exploiting the advantage of the combined setup. In this work, the major causes related to the simultaneous electrical and fluorescence recordings in horizontal BLMs are identified, and several solutions to counteract the issue are also proposed.
■
INTRODUCTION Membrane transport plays a crucial role in the function of living organisms.1−3 The demand to investigate the properties of these ion transports and the mechanism of the transport of non-ionic compounds through transient pores led to advances in the development of electrophysiological techniques.4−11 Black lipid membranes (BLMs), first reported by Mueller and co-workers in 1962,12 are pore-spanning bilayers. They are formed over an aperture in a thin Teflon sheet by painting lipids dissolved in a nonpolar organic solvent. Over the last decades, they are most commonly used for the characterization of ion channels, peptide pores, and ionophores,13−15 besides other applications such as DNA sequencing,16 single-molecule detection,17 and characterization of molecules.18 As an evolution of the classic BLM, a variety of modern horizontal BLM chamber designs have been demonstrated for an attempt to combine electrophysiology with optical microscopy.19 However, a careful literature review reveals that simultaneous fluorescence and electrical recordings in the BLM have been rarely reported for an unclear reason, as summarized in Table 1. Many works employ bright-field microscopy instead of fluorescence microscopy or perform fluorescence imaging and electrical measurements one after another separately without truly exploiting the advantage of the combined setup. These facts are well hidden in the literature because many reports present their fluorescence and electrical data as if they had been conducted at the same time, yet only a few works analyze their time-resolved signal correlations as an evidence for the synchronized measurements.20−23 Interestingly, successful simultaneous experiments have been performed with an agarose-supported BLM,24−26 besides a few © 2019 American Chemical Society
reports that use free-standing BLM with wide-field epifluorescence microscopy27−29 instead of confocal fluorescence microscopy, including our previous work. Why does the majority of the horizontal BLM setups fail to conduct synchronized electrical and fluorescence recordings? What is the advantage of the agarose-supported BLM compared to other systems? In this work, to answer these questions, we characterized the horizontal BLM by confocal laser scanning microscopy (CLSM), wide-field epifluorescence microscopy, fluorescence spectroscopy, interfacial tension (IFT) measurements, and electrical measurements. Addressing these questions will shed light on one of the bottlenecks in the development of modern electrophysiological characterization tools.
■
EXPERIMENTAL SECTION
Reagents and Materials. The phospholipids, 1-palmitoyl-2oleoyl-sn-glycero-3-phosphocholine (POPC, #850457), 1,2-dioleoylsn-glycero-3-phosphoethanolamine-N-(lissamine rhodamine B sulfonyl) (ammonium salt) (Rhod-DOPE, #810150), 1,2-dioleoyl-snglycero-3-phosphoethanolamine-N-(7-nitro-2-1,3-benzoxadiazol-4-yl) (ammonium salt) (NBD-DOPE, #810145), and 1,1-dioleoyl-snglycero-3-phosphoethanolamine-N-(cap biotinyl) (sodium salt) (biotin-DOPE, #870273), were purchased from Avanti Polar Lipids (USA). All of the above phospholipids were dissolved in chloroform and stored in the freezer (−26 °C). Texas Red DHPE triethylammonium salt (Texas Red-DHPE, #T1395MP) and streptavidin-Alexa Fluor conjugate (streptavidin-Alexa Fluor, Received: March 9, 2019 Revised: June 3, 2019 Published: June 7, 2019 8748
DOI: 10.1021/acs.langmuir.9b00673 Langmuir 2019, 35, 8748−8757
Article
Langmuir Table 1. Literature Review of Existing Artificial Bilayer Setups for Double Optical and Electrical Recordings type of artificial bilayer setup
year
time-resolved signal correlations between electrical and fluorescence signals?
Montal−Mueller bilayers30 Montal−Mueller bilayers26
1996 2003
no yes
solvent-free pore-suspending lipid bilayers31 solvent-free pore-suspending lipid bilayers27
2000
no
2015
micropipette aspirated liposomes32 BLM29
imaging technique a
research objective
FRAP/confocal wide-field epifluorescence confocal fluorescence
lateral diffusion of lipids conformational states of gramicidin channels
no
wide-field epifluorescence
platform development
2014
no
confocal fluorescence
1972
yes
BLM33 BLM25
2001 2003
no yes
BLM20
2007
no
34
2009
no
BLM35
2010
no
wide-field epifluorescence bright-field b FRET/wide-field epifluorescence FRAP/wide-field epifluorescence wide-field epifluorescence confocal fluorescence
BLM21
2016
no
bright-field
nano-BLM28
2018
yes
millimeter-sized BLM36
2016
no
BLM in microfluidics37 free-standing BLM solvent-free SLB agarose-based BLM22 agarose-based BLM23
2006 2012
no no
wide-field epifluorescence wide-field epifluorescence confocal fluorescence confocal fluorescence
1999
no
c
agarose-based BLM24
2009
yes
TIRF
BLM
TIRF
alamethicin single channel recordings
alamethicin single channel recordings characterization of mechanosensitive channels recording fluorescence changes in ANS-stained lipid bilayers under applied voltages electrical recording of single MaxiK channel C-less conformational changes of single gramicidin channels lateral diffusion of lipids characterization of formed BLM array lipid phase separation activity of gramicidin A as a function of lipid phase transition platform development for the repetitive bilayer formation α-hemolysin reconstitution calibration of voltage sensitive dye lipid phase separation BLM thinning process lipid lateral diffusion membrane integration of bacterial porin B lateral diffusion of lipids conformational changes of single ion channels characterization of open states of α-hemolysin pores
FRAP: fluorescence recovery after photobleaching. bFRET: fluorescence resonance energy transfer. cTIRF: total internal reflection fluorescence.
a
#S11227) and valinomycin (#V1644) were purchased from Thermo Fischer Scientific (Switzerland). Both Texas Red-DHPE and streptavidin-Alexa Fluor conjugate were received in powder and were dissolved in chloroform or Milli-Q water at the final concentration of 1 mg/mL. Aliquots of valinomycin (1 mg/mL) were made in ethanol and stored in the refrigerator (2 °C). Aliquots of α-hemolysin (Sigma-Aldrich, Switzerland) were made in the buffer solution to a final concentration of 1 mg/mL. Decane (reagent plus ≥99%), hexane (reagent plus ≥99%), chloroform (reagent plus ≥99%), and sucrose (≥99.5%) were purchased from Sigma-Aldrich (Switzerland), and ethanol (analytical reagent) was purchased from Fluka (Switzerland). The buffer solution was prepared with 10 mM 4(2-hydroxyethyl)piperazine-1-ethanesulfonic acid (HEPES) (Fluka, Switzerland) and 150 mM sodium chloride (Sigma, Germany) in ultrapure water filtered through Milli-Q Gradient A10 filters (Millipore AG, Switzerland). The pH was adjusted to 7.4 using 6 M NaOH (Acros Organics, Switzerland). Painting Solution. For the preparation of painting solution, POPC lipids dissolved in chloroform were first dried under the stream of nitrogen and placed in vacuum (50 mbar) for at least 2 h at room temperature. In the case where mixed lipids were used, POPC lipids in chloroform were mixed with either 1 wt % of fluorescently labeled or biotin-functionalized lipids and then were dried as described above. Note that when the spherical aberration is being taken care of, imaging BLMs generally works also at 0.25 wt % fluorescent lipids or
less. After the evaporation of chloroform, a mixture of decane/hexane (1:1 volume ratio) was added to give the desired concentration in each case. Giant Unilamellar Vesicle Preparation. Giant unilamellar vesicles (GUVs) were prepared by the electroformation method.38,39 Briefly, 5 μL of 10 mg/mL POPC lipid solution with 1 wt % fluorescently labeled or biotin-functionalized lipids in chloroform was deposited on the surface of an indium tin oxide (ITO)-coated glass slide (Sigma, Switzerland). Then, the ITO slide was placed under vacuum (50 mbar) for the evaporation of any residual chloroform for at least 2 h at room temperature. Next, the ITO slide with the deposited lipid film was covered with another ITO slide and sealed together with a Critoseal clay (Leica Microsystems, Switzerland). A space between these two slides (1 mm) was filled with 315 mM sucrose solution, and the ITO was connected to a function generator (TG300 Series, Aim-TTi, United Kingdom) with a conductive copper tape (Sigma, Switzerland). A sinusoidal voltage of 1 V at 10 Hz was applied for 2.5 h, which resulted in lipid swelling and the GUV formation. At the end of the electroformation, the resulted GUV solution was transferred to a plastic vial and used immediately. Supported Lipid Bilayers. A small amount of the GUV suspension with the desired lipid composition was used for the formation of the supported lipid bilayer (SLB). A droplet of aqueous buffer solution (100 μL) was placed on the glass coverslip activated by O2 plasma for 2 min in prior. Next, 10 μL of the GUV suspension was 8749
DOI: 10.1021/acs.langmuir.9b00673 Langmuir 2019, 35, 8748−8757
Article
Langmuir
Figure 1. Electrophysiological characterization of BLMs and comparison of their fluorescence intensity with that of solvent-free SLBs. (a) Schematic illustration and bright-field images of the horizontal BLM formation process. (b) Electrochemical impedance spectra were recorded during the BLM formation. (c) Transmembrane currents as a function of time were measured after the addition of valinomycin and α-hemolysin into the cis chamber at an applied voltage of V = +50 mV with 2 M KCl both in cis and trans sides. Red arrows indicate the moment when the peptides were added. Note that the current range with an open pore in the Teflon membrane (without BLMs) is around 10 μA. (d) 3Dreconstituted CLSM images of a BLM and SLB patches containing POPC lipids with 1 wt % Rhod-DOPE. Cross-sectional images in z−x and y−x are also presented (scale bars correspond to 30 μm). The inset in the SLB image has the same enhancement as the one for BLM is to visualize how bright SLB is. The fluorescence intensity of a selected area (yellow rectangle) as a function of the number of slices was plotted for both BLM and SLB. Because the BLM is not perfectly horizontal, x−y slice typically captures only a part of it as it can been seen as a line in the BLM x−y image. Note that the number of slice 0 corresponds to the first slice of each 3D imaging and does not correspond to the same z position in real space. These plots were used to identify the maximum fluorescence intensity. (e) Maximum fluorescence intensity observed in the CLSM images for SLB, BLM, and their annulus containing different fluorescently labeled lipids. (f) Fluorescence spectra of POPC painting solution and vesicle suspension containing 1 wt % Rhod-DOPE, NBD-DOPE, and Texas Red-DHPE. Solid lines correspond to emission spectra and dashed lines to excitation spectra. added, where GUVs sediment, adhere, and rupture on the coverslip for forming SLB. Large Unilamellar Vesicle Preparation. For the preparation of large unilamellar vesicles (LUVs), the POPC lipid solution with 1 wt % fluorescently labeled or biotin-functionalized lipids in chloroform was added in a glass vial and dried under the stream of N2. The dried lipid samples were then placed under vacuum (50 mbar) overnight. Subsequently, the buffer solution was added for the hydration of the lipid films, and the samples were placed under a high-power probe sonication40 for 15 min at 40% power (Omni Sonic Ruptor 400 ultrasonic homogenizer, Omni International). The LUV suspension was used immediately. Horizontal BLMs. All of the optical and electrical recordings in BLMs were carried out in a home-made chamber, which has been described in our previous work.41 As a partition, a Teflon sheet (Eastern Scientific, USA) with a thickness of 25 μm and a single pore (Ø = 50−100 μm) was used. The Teflon partition separates the two compartments, where each of them contained aqueous buffer solution. For the preparation of BLMs, a small amount of painting solution (
