Laura S Connelly1, Brian Meckes, Joseph Larkin, Alan L Gillman, Meni Wanunu, Ratnesh Lal. 1. Materials Science and Engineering Program, ‡Department of Bioengineering, and §Department of Mechanical and Aerospace Engineering, University of California-San Diego , 9500 Gilman Drive, La Jolla, California 92093, United States.
Abstract
Accurately defining the nanoporous structure and sensing the ionic flow across nanoscale pores in thin films and membranes has a wide range of applications, including characterization of biological ion channels and receptors, DNA sequencing, molecule separation by nanoparticle films, sensing by block co-polymers films, and catalysis through metal-organic frameworks. Ionic conductance through nanopores is often regulated by their 3D structures, a relationship that can be accurately determined only by their simultaneous measurements. However, defining their structure-function relationships directly by any existing techniques is still not possible. Atomic force microscopy (AFM) can image the structures of these pores at high resolution in an aqueous environment, and electrophysiological techniques can measure ion flow through individual nanoscale pores. Combining these techniques is limited by the lack of nanoscale interfaces. We have designed a graphene-based single-nanopore support (∼5 nm thick with ∼20 nm pore diameter) and have integrated AFM imaging and ionic conductance recording using our newly designed double-chamber recording system to study an overlaid thin film. The functionality of this integrated system is demonstrated by electrical recording (<10 pS conductance) of suspended lipid bilayers spanning a nanopore and simultaneous AFM imaging of the bilayer.
Accurately defining the nanoporous structure and sensing the ionic flow across nanoscale pores in thin films and membranes has a wide range of applications, including characterization of biological ion channels and receptors, DNA sequencing, molecule separation by nanoparticle films, sensing by block co-polymers films, and catalysis through metal-organic frameworks. Ionic conductance through nanopores is often regulated by their 3D structures, a relationship that can be accurately determined only by their simultaneous measurements. However, defining their structure-function relationships directly by any existing techniques is still not possible. Atomic force microscopy (AFM) can image the structures of these pores at high resolution in an aqueous environment, and electrophysiological techniques can measure ion flow through individual nanoscale pores. Combining these techniques is limited by the lack of nanoscale interfaces. We have designed a graphene-based single-nanopore support (∼5 nm thick with ∼20 nm pore diameter) and have integrated AFM imaging and ionic conductance recording using our newly designed double-chamber recording system to study an overlaid thin film. The functionality of this integrated system is demonstrated by electrical recording (<10 pS conductance) of suspended lipid bilayers spanning a nanopore and simultaneous AFM imaging of the bilayer.
In recent years, advancement
in fabrication techniques has led to novel nanoporous structures with
an array of applications in biotechnology,[1−5] polymer science,[6] and
energy.[7,8] Small changes in the nanoscale features
of these pores determine the specific conducting properties of ions
through and around the pores.[9−11] Ion-conducting nanopores, including
biological channels and receptors, may also interact with the surrounding
environment and change over time. For example, living systems rely
on the coordinated activity of membrane ion channels and receptors
that control ionic and metabolic homeostasis and cell–cell/extracellular
communications through regulation of ions, metabolites, and RNA transport.
Dysfunction of ion channels is associated with pathophysiology and
diseases such as Alzheimer’s disease and Parkinson’s
disease, addiction, and some genetic disorders.[12−15] Improved therapeutic development,
diagnosis, and/or prevention is therefore dependent on an accurate
understanding of these channels’ structure–activity
relationship.To fully understand the structure–function
relationship of nanopore-containing thin films and membranes, structure
and function must be correlated directly through simultaneous measurements.
However, current techniques cannot provide real-time, direct, and
simultaneous observation of the 3D structure and activity of these
pores.Atomic force microscopy (AFM) allows dynamic high-resolution
imaging of biological samples in physiological environments,[16−22] including 3D structures of individual ion channels in native hydrated
environments.[12,13,16,23−30] The open interface of the AFM allows its integration with other
techniques, including bilayer electrical recording and light/fluorescence
microscopy. The missing link for integrating AFM imaging and electrical
recording is a lack of an appropriate nanoscale support system.Nanopore samples have emerged as an exciting class of nanosensors
that have gained attention for their sensitivity to conductance changes,
especially in relation to the translocation of biomolecules, and numerous
nanoporous devices have been made from natural, artificial, and hybrid
materials.[1,3−5,31−38] Graphene is a promising and reliable material because of its unique
mechanical, electronic, thermal, and optical properties.[31,34,35,39−41] Graphene is thin enough to be precisely drilled using
a transmission electron microscope (TEM) and is strong enough to be
freely suspended over microscale pores.[10,31,35,42,43] Although much of the recent research on single-nanopore conductance
has been directed toward the application of DNA sequencing or biomolecule
translocation, micro- and nanopore devices have also been used to
study the activity of ion channels.[15,44]Here, we describe a novel graphene nanopore support
system (Figure 1) for simultaneous localized
high-resolution AFM and ionic conductance recording of nanoporous
thin films. Solid-state single-nanopore support substrates were fabricated
to fit into the open interface of the multimode AFM with our recently
developed two-chamber system (Figure 1). We
demonstrate the applicability of the integrated nanopore support system
combining AFM imaging and electrical recording using suspended lipid
bilayers. We show that a lipid bilayer deposited over the graphene
nanopore seals the pore. These bilayers can be imaged repeatedly with
AFM and retain their electrical properties. Electrical conductance
measurements reveal a dramatic reduction in the conductance, >1
μS for the open pore to <10 pS, for the bilayer-covered pores,
indicating complete coverage and sealing of the nanopore. The device
and setup that we present here demonstrates the imaging resolution,
nanopore size, and conductance sensitivity on scales compatible with
what is needed for the structure–activity study of ion channels.
The use of this technology would have major implications for, but
not limited to, the study of neurological disorders, pathological
studies, therapeutic screening, and drug addiction.
Figure 1
Schematic of the integrated
AFM system for imaging and conductance measurements. (A) The nanopore
support is glued into the top chamber of the double-chamber cup such
that the only liquid path connecting the chambers is through the deposited
lipid bilayer (green) over the nanopore. The double-chamber cup is
placed on the scanner head, and the liquid cell with mounted cantilever
is placed on top of the sample to allow for AFM imaging. Electrodes
are connected to the bottom chamber though the double-chamber cup
and to the top solution through an open port in the liquid cell for
measuring conductance activity. The electrodes are fed to an amplifier
and computer for analysis. (B) Schematic of the nanopore support (black)
with a deposited lipid bilayer (green) suspended over the pore to
seal the ionic conductance. (C) Schematic image of the double-chamber
cup design. The nanopore support from panel B fits into the top removable
piece and is set into the bottom piece.
Schematic of the integrated
AFM system for imaging and conductance measurements. (A) The nanopore
support is glued into the top chamber of the double-chamber cup such
that the only liquid path connecting the chambers is through the deposited
lipid bilayer (green) over the nanopore. The double-chamber cup is
placed on the scanner head, and the liquid cell with mounted cantilever
is placed on top of the sample to allow for AFM imaging. Electrodes
are connected to the bottom chamber though the double-chamber cup
and to the top solution through an open port in the liquid cell for
measuring conductance activity. The electrodes are fed to an amplifier
and computer for analysis. (B) Schematic of the nanopore support (black)
with a deposited lipid bilayer (green) suspended over the pore to
seal the ionic conductance. (C) Schematic image of the double-chamber
cup design. The nanopore support from panel B fits into the top removable
piece and is set into the bottom piece.
Materials and Methods
Materials
Silicon
oxide membranes (SiO2) were purchased from AppNano (Mountain
View, CA). Silicon oxide membranes are 200 nm thick and 20 ×
20 μm2 wide freestanding windows supported by a 300
μm thick silicon substrate. The windows were formed by KOH anisotropic
etching of a 450 × 450 μm2 opening on the backside
of the silicon support (Figures 2A and 3A). Single-layer CVD graphene deposited on 20 μm
thick Cu foil (2 × 2″) was obtained from Graphene Supermarket
(Calverton, NY). A Quanta 3D FEG focused ion beam (FIB) was used to
drill through the SiO2 suspended membrane. Either an iron(III)
chloride hexahydrate (FeCl3·6H2O; ≥
98%) solution (Sigma Aldrich) or Copper Etch APS-100 (Transene Co.)
was used to dissolve the Cu substrate of the graphene. Atomic layer
deposition (ALD) of 5 nm of Al2O3 was performed
using a GEMSTAR benchtop atomic layer deposition (ALD) process system
or the Beneq TFS200 atomic layer deposition system. A transmission
electron microscope (TEM) (JEOL 2010FEG, Japan) operating in bright-field
imaging mode was used for drilling through the graphene/Al2O3 membrane layer. AFM imaging was completed using a multimode
Nanoscope IV system and liquid cell (both from Bruker, Santa Barbara,
CA) with silicon nitride cantilevers (k = 0.08 N/m,
Asylum Research, Santa Barbara, CA). Conductance measurements were
completed using a custom-designed Lexan polycarbonate double-chamber
cup (Figure 1C) and Ag/AgCl wire electrodes.
Ecoflex Supersoft 5 silicone-cured rubber was used as an insulating
sealant of the nanopore sample in the double-chamber cup. A patch-clamp
amplifier (Dagan, Minneapolis, MN) was used for amplifying currents.
Electrolyte solutions at pH 8.5 containing 1 M KCl buffered with 10
mM Tris, similar to Venkatesan et al., was used for AFM imaging in
liquid and conductance measurements.[45] The
phospholipid1,2-diphytanoyl-sn-glycero-3-phosphocholine
(DiPhyPC) was purchased from Avanti Polar Lipids (Alabaster, AL).
Figure 2
Cross-sectional
schematic sequencing of the processing of a solid-state substrate
containing a single nanopore. (A) The starting substrate is a 6 mm
× 6 mm × 300 μm silicon substrate (black) with a 200
nm SiO2 layer (gray). A 20 × 20 μm2 window of suspended SiO2 is in the center of the substrate.
(B) A focused ion beam (red) is used to find the center of the 20
× 20 μm2 SiO2 window and to drill
a 1 μm hole. (C) A graphene flake (blue) coated with PMMA (green)
floating on the surface of H2O is placed over the 1 μm
FIB hole and allowed to dry. (D) Acetone is used to dissolve the PMMA,
leaving a graphene sheet suspended over the 1 μm hole. (E) Five
nanometers of Al2O3 (red) is deposited by atomic
layer deposition (ALD) over the graphene. (F) TEM (purple) is used
to drill a single nanopore in the center of the 1 μm hole.
Figure 3
(A) Top-view SEM image of the drilled FIB 1
μm hole. Inset is a zoomed-out SEM image of the same hole. The
20 × 20 μm2 SiO2 window is visible
in the SEM image. Scale bar = 1 μm. (B) TEM image of the drilled
FIB 1 μm hole covered with a layer of graphene showing no defects.
Thicker regions appear darker. Scale bar = 500 nm. (C) TEM image of
a single drilled 20 nm pore in the graphene/Al2O3 membrane. Scale bar = 20 nm.
Cross-sectional
schematic sequencing of the processing of a solid-state substrate
containing a single nanopore. (A) The starting substrate is a 6 mm
× 6 mm × 300 μm silicon substrate (black) with a 200
nm SiO2 layer (gray). A 20 × 20 μm2 window of suspended SiO2 is in the center of the substrate.
(B) A focused ion beam (red) is used to find the center of the 20
× 20 μm2 SiO2 window and to drill
a 1 μm hole. (C) A graphene flake (blue) coated with PMMA (green)
floating on the surface of H2O is placed over the 1 μm
FIB hole and allowed to dry. (D) Acetone is used to dissolve the PMMA,
leaving a graphene sheet suspended over the 1 μm hole. (E) Five
nanometers of Al2O3 (red) is deposited by atomic
layer deposition (ALD) over the graphene. (F) TEM (purple) is used
to drill a single nanopore in the center of the 1 μm hole.(A) Top-view SEM image of the drilled FIB 1
μm hole. Inset is a zoomed-out SEM image of the same hole. The
20 × 20 μm2 SiO2 window is visible
in the SEM image. Scale bar = 1 μm. (B) TEM image of the drilled
FIB 1 μm hole covered with a layer of graphene showing no defects.
Thicker regions appear darker. Scale bar = 500 nm. (C) TEM image of
a single drilled 20 nm pore in the graphene/Al2O3 membrane. Scale bar = 20 nm.
Nanopore Fabrication Process
To fabricate a single nanopore
support, the Si/SiO2 substrates were used as a base for
the processing (Figure 2A). A single hole with
a diameter of 1 μm was drilled by FIB through the center of
the SiO2 20 × 20 μm2 suspended membrane
area (Figures 2B and 3A). A sample of graphene on Cu was spin-coated with PMMA for 50 s
and baked at 180 °C for 10 min. The Cu foil was completely dissolved
in a FeCl3·6H2O solution or copper etchant
APS-100 (∼24 h). The remaining PMMA/graphene flake was deposited
over the center of the cleaned SiO2 membranes to ensure
coverage of the entire 1 μm FIB hole area and was allowed to
dry (Figure 2C). Dried samples were soaked
in acetone to dissolve the top layer of PMMA (Figure 2D). Five nanometers of Al2O3 was deposited
on the sample by ALD (Figures 2E and 3B). A nanopore was then drilled through the center
of the graphene/Al2O3-suspended membrane by
TEM (Figures 2F and 3C).[1,46] The nanopore sample was cleaned with acetone,
isopropanol, and UV/ozone cleaner for 15 min before use in conductance
measurements.
Experimental Setup for Imaging and Conductance
The double-chamber cup was used to hold the nanopore support as
previously described in Meckes et al. (Figure 1C).[47] The nanopore sample (Figure 1B) sits on the square inset of the top chamber piece
and is sealed into the top chamber using a continuous layer of fast
curing Ecoflex Supersoft 5 (Figure 1C).AFM imaging in liquid was performed with deflection feedback on the
nanopore sample in the double-chamber cup. An Ag/AgCl electrode was
placed through a port of the liquid cell, and another similar electrode
was embedded in the opposite chamber of the double-chamber cup (Figure 1A). The entire AFM base was placed in a Faraday
cage on in-house bungee cord-suspended platform for noise isolation.
A complete schematic of the experimental setup is shown in Figure 1A.
Lipid Bilayer Preparation
DiPhyPC
liposomes were formed by drying lipids dissolved in chloroform in
a rotovap. The dried lipids were hydrated with molecular grade H2O and vortexed. The solutions were then sonicated for 10 min.Nanopore surfaces were pretreated with a droplet of lipid–hexane
solution containing 70 μL of lipid (5 mg/mL) mixed with 100
μL of hexane. Liposomes were deposited over the nanopore and
incubated for 2 h at room temperature. Several drops of 1 M KCl, 10
mM Tris, 5 mM CaCl2, pH 8.0, buffer were added to the incubated
liposome droplet and incubated for an additional 10 min to transition
single-vesicle layers to suspended single planar bilayers across the
nanopore.[10] Excess Ca2+ and
liposomes were rinsed with 1 M KCl electrolyte solution buffered with
10 mM Tris to pH 8.5. The double-chamber cup was set on the scanner
head of the AFM. The area containing the nanopore, a 20 × 20
μm2 area of 200 nm thick SiO2, was aligned
under the cantilever tip in an optical system.
Simultaneous AFM Imaging
of Bilayer and Electrical Recording
Conductance levels and
capacitance of the bilayers were recorded using a National Instruments
DAC with a custom LabView 8.0 program and the patch-clamp amplifier
under applied voltages of ±100 mV. When sufficient sealing of
the nanopore was established, the AFM was engaged and, in contact
mode, the area of the nanopore covered with bilayers was imaged while
simultaneously recording conductance levels. Capacitance measurements
were performed by feeding ramp function signals (10 mV amplitude,
10 Hz) across the bilayer. The capacitance of the membrane is proportional
to the amplitude of the resulting square wave. All electrical measurements
were analyzed with Clampfit 10.2. A digital lowpass Gaussian filter
with a 50 Hz cutoff was applied to all data represented. Extraneous
60 Hz noise was eliminated with a digital notch filter centered at
60 Hz with a 9 Hz bandwidth.
Results
AFM Analysis
and Conductance Characterization of Nanopore
Images of the
nanopore support were taken periodically throughout the fabrication
process by electron microscopy (EM) and AFM. A 20 × 20 μm2 suspended SiO2 area was visible in SEM (Figure 3A inset) and in AFM (Figure 4A), enabling for the eventual centered drilling of the nanopore.
The AFM height images show a large deformation pattern of the suspended
SiO2 square that is not seen in SEM (Figure 4A). This deformation is due to stress relief following the
etching of the underlying Si layer. The center area of this square
where the graphene membrane resides appears to have very little deformation
in comparison to the edges, sufficient for imaging bilayers. The focused
ion beam (FIB)-drilled hole placed in the center of this square is
also visible in SEM (Figure 3A), TEM (Figure 3B), and AFM even after deposition of the graphene/Al2O3 layer (Figure 4A,B).
Complete coverage of the FIB hole with graphene was confirmed in the
TEM (Figure 3B) before nanopore drilling (Figure 3C). AFM imaging reveals the nanopore (Figure 4C), which is found by sequentially zooming in on
the center of the FIB hole, such as that seen in Figure 4. With a very sharp AFM tip, the nanopore size can be approximated
from the AFM image and compared to the size observed in TEM (Figures 3D, 4C,D, and 6). The square shape of the nanopore shown in Figure 4C is likely due to geometry effects of the AFM tip
(radius ∼30 nm) and the depth of the pore.
Figure 4
Progressive AFM height
images in tapping mode of a complete nanopore support. (A) Image showing
the complete 20 × 20 μm2 SiO2 area
with an X-shaped deformation resulting from stress relief following
the etching of the underlying Si layer. The center of the X, where
the graphene membrane resides and AFM imaging will occur, shows very
little localized deformation. Scale bar = 5 μm, height color
scale = 294 nm. (B) Image of the FIB hole showing uniform coverage
by graphene/Al2O3. Scale bar = 1 μm, height
color scale = 208 nm. (C) Image of a single nanopore drilled in graphene/Al2O3. The rectangular shape of the nanopore is an
effect of the shape of the tip and imaging into the pore area. Scale
bar = 20 nm, height color scale = 17.8 nm. (D) Three-dimensional view
of the nanopore shown in panel C. Image size is 125 × 122 nm2.
Figure 6
(A) AFM deflection image
of 20 × 20 μm2 area containing the nanopore
without a deposited bilayer (outlined in blue). The 1 μm area
containing the graphene/Al2O3 membrane and a
single nanopore are observed in the center of the square. The black
arrow shows the zoomed view of this area in panel B. Scale bar = 5
μm. (B) High-resolution AFM image of the 1 μm area containing
the graphene/Al2O3 membrane and a single nanopore
(outlined in red) with no bilayer present. The white arrow indicates
the location of the nanopore. Scale bar = 250 nm.
Progressive AFM height
images in tapping mode of a complete nanopore support. (A) Image showing
the complete 20 × 20 μm2 SiO2 area
with an X-shaped deformation resulting from stress relief following
the etching of the underlying Si layer. The center of the X, where
the graphene membrane resides and AFM imaging will occur, shows very
little localized deformation. Scale bar = 5 μm, height color
scale = 294 nm. (B) Image of the FIB hole showing uniform coverage
by graphene/Al2O3. Scale bar = 1 μm, height
color scale = 208 nm. (C) Image of a single nanopore drilled in graphene/Al2O3. The rectangular shape of the nanopore is an
effect of the shape of the tip and imaging into the pore area. Scale
bar = 20 nm, height color scale = 17.8 nm. (D) Three-dimensional view
of the nanopore shown in panel C. Image size is 125 × 122 nm2.Ion conductance measurements are
a good way to probe the pore geometry.[10,11,31,37,40] Neglecting access resistance for our large pores, pore conductance
relates to geometry via the following equation:[11,37]Where G is conductance, dpore is the pore diameter, Lpore is the pore cylindrical length, nKCl is the concentration of the buffer, e is elementary charge, σ is the surface charge density in the
nanopore, and μK and μCl are the
electrophoretic mobilities of the two solution ions potassium and
chloride. A 1 M KCl buffer was used in the work reported here. The
electrophoretic mobilities of potassium and chloride are μK = 7.616 × 10–8 m2/(V s)
and μCl = 7.909 × 10–8 m2/(V s) at room temperature.[10,37]The
predicted conductance from the given equation is dependent on pore
morphology and surface charge density of the sample. High surface
charge density for graphene/Al2O3 layers is
considered 200 mC/m2, and minimum surface charge is 0 mC/m2.[10,37] A range of solid-state nanopores was fabricated
with dpore values of 20–50 nm and Lpore ≈ 5 nm. The expected conductance
values in this nanopore size range for high and low surface charge
density samples would be approximately 1100–6100 and 940–5900
nS, respectively. Open conductance values of individual solid-state
nanopores were measured by ramping at 0.4 mV/s over ±10 mV. The
conductance of the ∼25 nm diameter nanopore sample, free of
bilayer (Figures 5A and 6), was measured to be
2765 nS, which falls in the reasonable range of conductance values
for 20–50 nm nanopores.
Figure 5
(A) I–V curve of an open nanopore shown in the inset image measuring
∼25 nm with a conductance of 2765 nS. This nanopore was used
in the simultaneous AFM imaging and electrophysiology measurements
in panels B and C and Figures 6 and 7. Scale bar = 20 nm. (B) Ionic current recording
of DiPhyPC lipid bilayer deposited over the nanopore in 1 M KCl electrolyte
solution. The observed conductance was <10 pS. (C) I–V curve of the nanopore from panel A after
deposition of a DiPhyPC lipid bilayer. The observed conductance was
reduced to <10 pS, indicating a full seal of the nanopore by the
bilayer. The inset compares the I–V curves of the open nanopore (black) and DiPhyPC bilayer-sealed
conductance (red).
(A) I–V curve of an open nanopore shown in the inset image measuring
∼25 nm with a conductance of 2765 nS. This nanopore was used
in the simultaneous AFM imaging and electrophysiology measurements
in panels B and C and Figures 6 and 7. Scale bar = 20 nm. (B) Ionic current recording
of DiPhyPC lipid bilayer deposited over the nanopore in 1 M KCl electrolyte
solution. The observed conductance was <10 pS. (C) I–V curve of the nanopore from panel A after
deposition of a DiPhyPC lipid bilayer. The observed conductance was
reduced to <10 pS, indicating a full seal of the nanopore by the
bilayer. The inset compares the I–V curves of the open nanopore (black) and DiPhyPC bilayer-sealed
conductance (red).
Figure 7
(A) AFM deflection image of square area
containing the nanopore after the deposition of the DiPhyPC bilayer
(corners highlighted blue) obtained while simultaneously recording
ionic conductance. AFM image corresponds to the same area shown in
Figure 4A. The 1 μm area containing the
graphene/Al2O3 membrane and a single nanopore
is observed in the center of the square. The black arrow points to
the zoomed view of this area in panel B. Scale bar = 5 μm. (B)
High-resolution AFM image of the 1 μm area containing the graphene/Al2O3 membrane and a single nanopore. Scale bar =
2 μm. (C) Ionic current recording trace obtained while AFM imaging.
The initial noise increases at the beginning of the recording and
during the 2–4 min time of recording correlate to physical
interactions with the system (AFM base and Faraday cage door) and
subside upon establishment of physical isolation of the entire system.
Applied voltages of ±100 mV were applied to confirm pore sealing.
(A) AFM deflection image
of 20 × 20 μm2 area containing the nanopore
without a deposited bilayer (outlined in blue). The 1 μm area
containing the graphene/Al2O3 membrane and a
single nanopore are observed in the center of the square. The black
arrow shows the zoomed view of this area in panel B. Scale bar = 5
μm. (B) High-resolution AFM image of the 1 μm area containing
the graphene/Al2O3 membrane and a single nanopore
(outlined in red) with no bilayer present. The white arrow indicates
the location of the nanopore. Scale bar = 250 nm.
Simultaneous AFM and Electrical Recording of a Suspended Lipid Bilayer
Following the deposition of 1,2-diphytanoyl-sn-glycero-3-phosphocholine (DiPhyPC) bilayers on the nanopore support
mounted in the double-chamber cup system, the pore conductance decreased
to <10 pS, as measured in the range ±100 mV (Figure 5C, red). The capacitance of the device with the
bilayer was measured to be ∼375 pF. Contributing factors to
this value may include current passing through the Al2O3 layer to the graphene sheet, which increases the capacitive
area and the geometry of a thin membrane.[1,40,48] However, capacitances in these ranges are
frequently utilized for planar lipid bilayer (PLB) recording of ion
channels.[49] In other experiments, partial
sealing of the same nanopore was observed by conductance value to
drop only to ∼0.83 nS (data not shown), indicating an incomplete
seal.The integrity of the fully sealed suspended lipid bilayer,
as determined by capacitance and electrical recording, remained stable
while engaging the AFM (Figure 7). The interaction
force of the AFM was minimized such that it did not interfere with
the activity or structure of membrane. The AFM images in Figure 7A,B of the suspended bilayer over the nanopore shows
complete coverage. The electrical recording and simultaneously obtained
AFM images of the bilayer over the nanopore are shown in Figure 7. Increases in noise were observed during the adjustment
of the Faraday cage during the electrical recording. Throughout the
AFM imaging (>1 h), switching of the voltage bias did not impact
the conductance value of the bilayer.(A) AFM deflection image of square area
containing the nanopore after the deposition of the DiPhyPC bilayer
(corners highlighted blue) obtained while simultaneously recording
ionic conductance. AFM image corresponds to the same area shown in
Figure 4A. The 1 μm area containing the
graphene/Al2O3 membrane and a single nanopore
is observed in the center of the square. The black arrow points to
the zoomed view of this area in panel B. Scale bar = 5 μm. (B)
High-resolution AFM image of the 1 μm area containing the graphene/Al2O3 membrane and a single nanopore. Scale bar =
2 μm. (C) Ionic current recording trace obtained while AFM imaging.
The initial noise increases at the beginning of the recording and
during the 2–4 min time of recording correlate to physical
interactions with the system (AFM base and Faraday cage door) and
subside upon establishment of physical isolation of the entire system.
Applied voltages of ±100 mV were applied to confirm pore sealing.
Discussion
We
have developed a nanoscale thin-film support integrated with our newly
developed double chamber capable of simultaneous electrical recording
and AFM imaging of biological membranes and membrane proteins. The
practicality and benefit of this system was demonstrated by structural
imaging while measuring the ion-insulating properties of suspended
lipid bilayer membranes.Each step in the fabrication of the
nanopore support was chosen with consideration for the ease of fabrication
as well as function. A silicon dioxide film on silicon is an ideal
material combination to use for the base structure of the nanopore
support because of its well-characterized electrical properties, reproducibility,
and commercial availability.[32,42,50] Silicon dioxide provides an insulating coating to the large area
of the substrate, an essential property to isolate the two compartments
of the electrical recording setup. FIB offers a fast and easily controlled
approach for opening the 1 μm hole in the sample to allow for
diffusion of electrolytes through the nanopore. Graphene binds tightly
to the SiO2 upon drying, which secures it as a suspended
insulating membrane over the FIB gaping hole. In contrast to the multilayer
graphene-based structure described by Venkatesan et al., using a single
graphene-deposition step allows us to achieve a suspended membrane
thickness that is more commensurate with that of a lipid bilayer while
still providing good mechanical support.[10] Electrical insulation and structural reinforcement was provided
by deposition of 5 nm Al2O3 using ALD, which
creates a uniform coating with a minimal increase in membrane thickness.[10,31,50] The final graphene/Al2O3 membrane surrounding the nanopore is <10 nm. TEM
drilling allows positioning the bilayer support on the nanometer scale.[2,42]Nanopore supports that show a smooth surface in AFM revealed
the location and size of the nanopore itself (Figures 4C and 6), and, to our knowledge, AFM
images of such small nanopores in graphene/Al2O3 membranes have not been previously shown. Characterizing the local
environment of a solid-state nanopore by AFM could potentially be
used in single-molecule studies by functionalizing the AFM tip. Additionally,
the high-resolution AFM imaging of a 20 nm nanopore, as in Figure 4C, suggests membrane proteins in a similar size
range may be individually probed in future simultaneous structure–conductance
studies.[12,13,23,24] Individual ion channels in supported membranes have
often been resolved at larger scan sizes.[12,13,23,24,51−54] These solid-state nanopores are therefore suitable
for the intended application of ion channel studies. With this technology,
the structure of suspended bilayers or membrane proteins in suspended
bilayers may be explored to achieve a better understanding of their
function.Future work with lipid bilayers and the nanopore substrates
will investigate the conductance of membrane proteins. The aim of
these efforts will be to resolve individual open and closed channel
structures localized in the suspended bilayer and to correlate characteristic
channel conductances.
Conclusions
The use of a defined
solid-state single-nanopore support in AFM allows for localized characterization
in and around the nanopore. We have used this nanopore support to
combine both imaging with AFM and functional mapping with bilayer
electrical recording. We show that single solid-state nanopores can
be fabricated in graphene reinforced with Al2O3. The hierarchy of the sample structure allows for quick and easy
location of the single nanopore in AFM. This enables accurate identification
of a nanoporous thin film, such as lipid bilayers with embedded ion
channels, when suspended over this nanopore. The presence of suspended
bilayers across the electrical recording path was confirmed and characterized
through conductance and capacitance measurements (Figure 5). Small scan sizes, <1 μm, and repeated
stable imaging of the suspended bilayers (Figure 7) suggest high-resolution imaging of thin lipid membranes
and membrane proteins is possible. Correlated structure and activity
information on ion channels obtained using this integrated system
will open the door for the study of basic physiological and biological
systems as well as for defining the underlying mechanisms of pathophysiology
and diseases, including neurodegenerative diseases, drug addiction,
biological pathways, and protein structures. The system described
here can be applied more broadly to other thin-film-based techniques,
including molecular separation, DNA sequencing, and catalysis.
Authors: Daniel Branton; David W Deamer; Andre Marziali; Hagan Bayley; Steven A Benner; Thomas Butler; Massimiliano Di Ventra; Slaven Garaj; Andrew Hibbs; Xiaohua Huang; Stevan B Jovanovich; Predrag S Krstic; Stuart Lindsay; Xinsheng Sean Ling; Carlos H Mastrangelo; Amit Meller; John S Oliver; Yuriy V Pershin; J Michael Ramsey; Robert Riehn; Gautam V Soni; Vincent Tabard-Cossa; Meni Wanunu; Matthew Wiggin; Jeffery A Schloss Journal: Nat Biotechnol Date: 2008-10 Impact factor: 54.908
Authors: Robert Y Henley; Ana G Vazquez-Pagan; Michael Johnson; Anastassia Kanavarioti; Meni Wanunu Journal: PLoS One Date: 2015-12-11 Impact factor: 3.240