Page 1
Copyright © 2014 American Scientific PublishersAll rights reservedPrinted in the United States of America
ReviewJournal of
Nanoscience and NanotechnologyVol. 14, 433–446, 2014
www.aspbs.com/jnn
Recombinant Protein-Based Nanoscale
Biomemory Devices
A. K. Yagati1, J. Min3�∗, and J.-W. Choi1�2�∗1Research Center for Integrated Biotechnology, Sogang University, Seoul 121-742, Korea
2Department of Chemical and Biomolecular Engineering, Sogang University, Seoul 121-742, Korea3School of Integrative Engineering, Chung-Ang University, Seoul 156-756, Korea
Biomolecular computing devices that are based on the properties of biomolecular activities offera unique possibility for constructing new computing structures. A new concept of using variousbiomolecules has been proposed in order to develop a protein-based memory device that is capableof switching physical properties when electrical input signals are applied to perform memory switch-ing. To clarify the proposed concept, redox protein is immobilized on Au nanoelectrodes to catalyzereversible reactions of redox-active molecules, which is controlled electrochemically and reversiblyconverted between its ON/OFF states. In this review, we summarize recent research towards devel-oping nanoscale biomemory devices including design, synthesis, fabrication, and functionalizationbased on the proposed concept. At first we analyze the memory function properties of the proposeddevice at bulk material level and then explain the WORM (write-once-read-many times) nature ofthe device, later we extend the analysis to multi-bit and multi-level storage functions, and then wefocus the developments in nanoscale biomemory devices based on the electron transport of redoxmolecules to the underlying Au patterned surface. The developed device operates at very low volt-ages and has good stability and excellent reversibility, proving to be a promising platform for futurememory devices.
Keywords: Recombinant Protein, Cyclic Voltammetry, Nanoscale Biomemory, Nanofabrication,Nanobiochip, Biomemory Devices.
CONTENTS1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433
2. Device Structure and Fabrication of
Protein-Based Memory Devices . . . . . . . . . . . . . . . . . . . . . . . . . 435
2.1. Basic Concept of the Memory Device . . . . . . . . . . . . . . . . 435
2.2. Thin Film Fabrication Methods . . . . . . . . . . . . . . . . . . . . . 435
2.3. Characterizations on the
Proposed Biomemory Device . . . . . . . . . . . . . . . . . . . . . . . 437
3. Development of a Protein-Based Memory
Device and Its Worm Nature . . . . . . . . . . . . . . . . . . . . . . . . . . . 438
4. Enhancement in Data Storage Capacity in
Protein-Based Memory Devices . . . . . . . . . . . . . . . . . . . . . . . . . 440
4.1. Multi-Bit Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 440
4.2. Multi-Level Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . 440
5. Nanobiomemory Device Composed of Recombinant
Azurin on Nanostructured Surfaces . . . . . . . . . . . . . . . . . . . . . . 442
6. Future Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445
Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445
References and Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445
∗Authors to whom correspondence should be addressed.
1. INTRODUCTIONSince the invention of first junction transistor one of
the most important issues has been to reduce the size
of the transistor. Many engineers and scientists have
focused their research on the manipulation of structures
at micro/nano-scale.1 Relentless decrease in the size of
silicon-based microelectronics devices is posing prob-
lems. The most important among these are the limitations
imposed by quantum-size effects and instabilities intro-
duced by the effects of thermal fluctuations.2 Also, as
the size of the structures has been reduced, the complex-
ity and cost of fabrication have increased exponentially.3
The continuity of this miniaturization trend is becom-
ing difficult due to the limitations of conventional lithog-
raphy, complexity of integration, physical phenomena,
and other aspects.4 For decades, semiconductor technol-
ogy has advanced at an exponential rate as stated by
Moore’s law, in which the number of features in a given
J. Nanosci. Nanotechnol. 2014, Vol. 14, No. 1 1533-4880/2014/14/433/014 doi:10.1166/jnn.2014.9006 433
Page 2
Recombinant Protein-Based Nanoscale Biomemory Devices Yagati et al.
area of a substrate doubles every 18 to 24 months, and
the computing power and capabilities will increase with
each generation of device while the cost per function
decreases.5–7 Also, semiconductors have served as basic
materials for constructing electronic devices for the past
50 years with tremendous growth. Despite the vast growth
of semiconductors, it is uncertain whether devices that
depend on the bulk properties of materials will retain the
required characteristics to function when feature size ulti-
mately reach nanoscale dimensions.8–10 These problems
prompted scientists to focus on the use of biomolecules
as they have the advantages of functionality and speci-
ficity. Much work has been done in the bioelectronics field,
but since the 1990s, it’s been focused on creating bet-
ter bioelectronic devices by integrating biomolecules with
semiconductors.11
On the other hand, nature consists of perfectly con-
trolled and manipulated nano-scale components using
molecular recognition of various biological materials such
as deoxyribonucleic acid (DNA), ribonucleic acid (RNA),
protein, and cell. The knowledge obtained from nature
A. K. Yagati obtained his B.Sc. and M.Sc. in Electronics from Andhra University, India
(1999 and 2001), M.Tech. in Bioelectronics from Tezpur University, Assam, India (2006)
and Ph.D. in Integrated Biotechnology from Sogang University, Korea in 2011, respec-
tively. Currently he is working as a post-doctoral researcher at nanobioelectronics lab,
Sogang University with Professor Jeong Woo Choi. He also worked as a Senior Project
Assistant at the School of Medical Science and Technology (SMST), Indian Institute of
Technology, Kharagpur, India with Professor Sujoy K. Guha. His research interests include
electrochemistry; sensors based on chemically modified electrodes and working on the
development of nano-scale biomemory devices.
J. Min obtained his B.S., M.S., and Ph.D. in Chemical Engineering from Sogang Uni-
versity, Korea (1992, 1994, and 1998). Currently he is an Associate Professor at Chung-
Ang University Korea. He was a Post-doctoral Research Associate at Environmental and
Biological Engineering, Cornell University, NY, USA (2000–2001), Senior Scientist at
Innovative Biotechnologies, International, NY, USA (2002–2003), and Project Leader at
Samsung Advanced Institute of Technology, Korea (2003–2006). His research has been
mainly focused on the fields of bio-nanodevices, especially for protein-based memory,
RNA-based microsensors, and cell-based microchips.
J.-W. Choi obtained his B.S. and M.S. in Chemical Engineering of Sogang University,
Korea (1982, 1984). He received a Ph.D. in Chemical and Biochemical Engineering of
Rutgers University, USA (1990), D.Eng. In Biomolecular Engineering of Tokyo Institute
of Technology, Japan (2003), and MBA in Business School of University of Durham,
UK (2006). He is a Professor of Chemical and Biomolecular engineering, and Director
of the Institute of Integrated Biotechnology; Sogang University, and Director of Sogang-
Binggrae Research Center for Advanced Food Analysis. Professor Choi’s research has
been mainly focused on the fields of nanobioelectronics, especially for biomemory, protein
chips, and cell chips. He has published more than 300 journal papers in bioelectronics
and biotechnology field. He is also interested in doing research about management of
innovation, especially for open innovation in fused biotechnology area.
can be used to solve the miniaturization issues. Moreover,
engineers and scientists have applied this knowledge to
implement artificial nano/microstructures for developing
bioelectronic devices.12–14 Bioelectronics is a rapidly pro-
gressing interdisciplinary research field that aims to inte-
grate biomaterials and electronic elements into functional
devices.15–17 Genetic engineering and chemical methods
can optimize the binding, catalytic, and electron transport
properties in the nanometer-sized biomaterials. As a con-
sequence, there has been an intense interest in developing
molecular-based electronic materials for use in both mem-
ory architectures and circuit elements.18–20
In storage technology, the increase in density of stor-
age devices based on silicon and inorganic materials
is becoming increasingly difficult. Hence biomaterials
such as protein and DNA with their inherent size and
electronic properties could be alternatives to succeed
the silicon-based technologies in developing bioelectronic
devices.21 A bioluminescent molecular switch for glu-
cose was developed22 by inserting glucose-binding pro-
tein (GBP) into the structure of the photoprotein aequorin
434 J. Nanosci. Nanotechnol. 14, 433–446, 2014
Page 3
Yagati et al. Recombinant Protein-Based Nanoscale Biomemory Devices
(AEQ). In the presence of glucose, GBP undergoes a con-
formational change, bringing the two segments of AEQ
together, “turning on” bioluminescence and detecting glu-
cose. A photonic DNA memory was developed23 based
on the positional information of DNA for scaling up the
address space of the DNA memory. Use of the optical
techniques is useful in controlling positional addresses in
parallel. Multiple logic gates based on electrically wired
surface-reconstituted enzymes was developed,24 and a mul-
tifunctional logic gate25 based on folding/unfolding transi-
tions of a protein. Further, a lithography technique was pro-
posed based on a protein pattern on a solid surface that can
perform write, read, and erase operations.26 Also, viruses
show their good side in the field of bionanoelectronics.
Being viruses, they carry a bad reputation regardless of
wherever we find them, but they exhibit memory phenom-
ena when they are coupled with nanoparticles.27 Further,
inversion switch using the fim and hin inversion recom-
bination systems was used to create a sequential memory
switch.28 This switch is capable of transitioning from one
state to another in a manner analogous to a finite state
machine, while encoding the state information into DNA.
In this review, we analyze and focus on using geneti-
cally engineered recombinant protein to store information
as a charge storage element in the design and construction
of a nanoscale biomemory device. Redox reactions of pro-
teins are utilized for information transfer and energy stor-
age in major natural mechanisms. This is because redox
reactions in proteins are of great interest in various fields
including biology and chemistry29–31 particularly proteins
with the redox property have been introduced in molecular
electronic devices.32–34
2. DEVICE STRUCTURE AND FABRICATIONOF PROTEIN-BASED MEMORY DEVICES
2.1. Basic Concept of the Memory DeviceElectron transfer through proteins, a fundamental ele-
ment of many biochemical reactions, has been studied
extensively in solution and is one of the most fundamental
processes in biological systems,35 crucial for different bio-
logical energy conversion processes from respiration to
photosynthesis, and prominent in diverse metabolic cycles.
Electron transfer reactions are also performed by a range
of proteins in which electron tunneling occurs over long
distances.36 While these reactions proceed in the proteins,
when present in their natural, usually aqueous environ-
ment, there have been attempts to explore the electronic
conductance of various proteins in both wet electrochem-
ical and solid state conditions. Hence, redox reactions of
proteins or enzymes are utilized for information transfer
and energy storage in major natural mechanisms. There-
fore, redox reactions are of great interest to develop bio-
based devices by mimicking the biomechanisms.
Recently, Yang’s group developed a digital mem-
ory device composed of tobacco mosaic virus (TMV)
conjugated with platinum nanoparticles (TMV-Pt).37 The
TMV-Pt nanocomposite has properties that show electrical
bistability depending on the voltage controlled for con-
ductance states. The mechanism of this memory device,
due to charge transfer from the RNA to Pt NPs under the
high electric field and the charge trapped in the nanoparti-
cles, subsequently changes the conductivity of the material
system. This conjugate system was one of the innovative
candidates for future electronic devices. The Rinaldi group
also developed a transistor based on the properties of met-
alloprotein, in which they explain the influence of the gate
voltage on the current flow when source to drain is inter-
connected with protein molecules. It is observed that the
device behaves both like p-channel and n-channel MOS-
FET, which is an innovative investigation for future elec-
tronic devices.38 Ghadiri group demonstrated molecular
logic circuits based on DNA molecules and showed a com-
plete set of modular DNA-based Boolean logic gates. This
approach is based on solid-supported DNA logic gates
that are designed to operate with single-stranded DNA
inputs and outputs. As the solution-phase serves as the
communication medium between gates, circuit wiring can
be achieved by designating the DNA output of one gate
as the input to another thus facilitating the construction
of multi-level circuits.39 These works pave the path for
developing enzyme-based bioelectronic computers. Choi’s
group also demonstrated a new concept of biomemory
device using metalloprotein as a charge storage element.
The fundamental principle of the proposed memory device
is shown in Figure 1(a). The two basic stages (charge writeand erase) are required for achieving the memory con-
cept. Metalloprotein possesses two distinct electrochemical
states (oxidation and reduction) that can be controlled by
applied potential.
The application of oxidation potential to the protein
molecule on the Au substrate causes the transfer of an
electron from the protein molecule to the Au substrate
resulting in storage of positive charge (write) in the
molecule as shown in Figure 1(b). On the other hand,
the application of a reduction potential causes the transfer
of an electron back to the protein molecule thereby eras-
ing the stored charge (erase). The charge confined in the
two states can be determined (read), when an open circuit
equilibrium potential is applied to the system. The contin-
uous application of oxidation, OCP, and reduction poten-
tial provide a parameter for constructing protein-based
biomemory devices in which write, read, and erase charges
can manipulate information like a conventional inorganic
memory device. It is a key mechanism to retain and han-
dle the electrochemical properties of a metalloprotein for
mimicking and developing a biomemory system.40–42
2.2. Thin Film Fabrication MethodsProtein adsorption on surfaces in a biological environ-
ment is the first step in biofilm formation. Different
J. Nanosci. Nanotechnol. 14, 433–446, 2014 435
Page 4
Recombinant Protein-Based Nanoscale Biomemory Devices Yagati et al.
Figure 1. (a) A cyclic voltammogram showing the potentials used
for developing protein-based memory devices. (b) The electron transfer
mechanism of ferredoxin on Au electrode: (1) application of oxidation
voltage causes an electron to move from the protein to Au electrode
(leads to positive charge); (2) application of reduction voltage leads an
electron back to the protein (leads to charge neutral).
physical surface parameters are influencing the pro-
tein adsorption on surfaces such as topography, wetting
behavior, surface charge, and mechanical stiffness. To
form a well-ordered, self-assembled monolayer (SAM)
consisting of biomolecules on solid surfaces, three
main issues have to be considered: (i) orientation,
(ii) address, and (iii) surface coverage. Many researchers
are working to fabricate monolayers of biomolecules
with high orientation. A successful immobilization scheme
requires:
(1) reproducibility and stability of the protein layer,
(2) uniformity of the surface structures
(3) the ability to control the immobilization density of the
immobilized molecules,
(4) if possible, higher coverage/density is preferred, and
(5) orientation of molecules so that maximum binding
density can be achieved.
SAMs, by using chemical linker materials, have been used
to covalently attach proteins on solid supports. The applica-
tion has been used in developing bioelectronic devices, as
well as studying protein/surface and ligand/receptor inter-
actions. Since the emphasis of this review is on memory
functioning, the focus will be mainly on protein immo-
bilization. Here three different types of adsorption meth-
ods are discussed for protein immobilization; however,
the techniques described here apply to other biomolecules
as well.43
2.2.1. Physical Adsorption (van derWaals Adsorption)
Proteins adsorb onto a variety of surfaces. It is gen-
erally believed that the adsorption takes place due to
hydrophobic, ionic, and Van der Waals interactions. The
phenomenon of adsorption is essentially an attraction of
adsorbate molecules to an adsorbent surface. The prefer-
ential concentration of molecules in the proximity of a
surface arises because the surface forces of an adsorbent
solid are unsaturated. Both repulsive and attractive forces
become balanced when adsorption occurs, and the process
is nearly always an exothermic process.44
Physical adsorption is the simplest way to immobilize
proteins on surfaces. However, the binding to a surface is
usually not stable, and the activity of the protein is usually
lost. For this reason, adsorption has to be controlled care-
fully. Concentrated protein solutions should be avoided
since additional layering will occur. Additional adsorp-
tion usually involves protein-protein adsorption instead of
surface–protein adsorption. Such adsorption is inherently
unstable, and the protein may gradually desorb from the
surface.
2.2.2. Chemisorption �Activated Adsorption�In chemisorption, there is a transfer or sharing of electrons,
or breakage of the adsorbate into atoms or radicals that are
bound separately. Here, the formation of chemical bonds
between the adsorbate and adsorbent is a monolayer, often
releasing more heat than the heat of condensation. Cova-
lent immobilization of proteins on surfaces is based on the
linkage between surface and protein functional groups.45
Generally there are three different approaches to attach
proteins to a surface other than physical adsorption. The
first approach is to modify the substrate, then attach pro-
teins to the substrate. The second approach is to link
the protein of interest to sulfur-containing molecules. The
modified protein is then self-assembled onto gold substrate
through gold-thiol linkage. The third approach is to immo-
bilize the protein through a lock and key interaction, or
ligand/receptor mechanism.46
2.2.3. Site-Directed ImmobilizationSite-specific covalent immobilization, via a unique natural
or non-natural residue, affords a uniformly oriented pro-
tein array. One such method is protein recombination to
generate cysteine residues on specific parts of the protein
to bind at the gold surface. When protein has cysteine
residues in the right regions, protein could be immobi-
lized and addressed on the gold surface with good orienta-
tion and coverage. The regenerated recombinant ferredoxin
with cysteine residues by site-directed mutagenesis47�48 is
demonstrated in Figure 2.
436 J. Nanosci. Nanotechnol. 14, 433–446, 2014
Page 5
Yagati et al. Recombinant Protein-Based Nanoscale Biomemory Devices
Figure 2. Construction of recombinant plasmids and DNA sequence analysis. (a) Schematic representations of fdxA gene variants for expression.
(b) Analyses of PCR products of the fdxA gene variants. (line 1) size marker, (line 2) PCR product of fdxA gene, (line 3) size marker, (line 4) PCR
product of mfdxA gene. (c) Verification by DNA sequencing. (d) Design of expression vector for fdxA gene variants.
2.3. Characterizations on theProposed Biomemory Device
The formation of biofilm on electrode surfaces requires
morphological and physical measurements to confirm the
layer characteristics. To achieve this, scanning probe
microscopy (AFM and STM), surface plasmon resonance
(SPR), and other optical detection methods are normally
used. Surface plasmon resonance spectroscopy has been
demonstrated as a technique to monitor biomolecular
interactions between ligands immobilized on a thin metal
substrate with other biomolecules adsorbed on it. One
advantage to this technique is the lack of sample solution
interference with the spectroscopic measurement. The light
in SPR is not passed through the sample solution, eliminat-
ing many problematic matrix effects that take place with
other spectroscopic methods. Also, many researchers ana-
lyzing biological problems need to develop recombinant
proteins, and it is important to show that the recombinant
protein has the same structure as its native component.
Here, the intensity of the reflected light is measured as
a function of the angle of incidence. At a critical angle
of excitation of surface plasmon resonance, a minimum
in the intensity of the reflected light is observed. Also to
determine the optimum concentration of the protein to be
immobilized on the Au surface, SPR spectrums at vari-
ous concentrations were observed. As the concentration
of the protein is increased, the value of the SPR angle is
shifted to higher values and saturated at a concentration of
0.1 mg/ml shown in Figure 3.
Figure 3. (a) Change of SPR angle shift for the increasing concentra-
tion inset shows SPR characterization of the immobilization of 0.1 mg/ml
ferredoxin molecules on gold substrate comparative to Au surface.
J. Nanosci. Nanotechnol. 14, 433–446, 2014 437
Page 6
Recombinant Protein-Based Nanoscale Biomemory Devices Yagati et al.
Figure 4. STM topology of (a) bare Au surface and (b) ferredoxin-immobilized Au surface respectively obtained at a scan rate of 1 Hz.
To construct the molecular electronic device with func-
tional protein film, the formation of aggregated pro-
tein molecules has been considered as one of the most
important factors that dominantly affects the device per-
formance. The film formation was achieved on the Au
substrate by self-assembly technique where the protein is
immobilized and the samples were treated with Tween 20
a non-ionic surfactant. The protein treated with non-ionic
surfactant allowed well-distributed protein molecules with-
out loss of its activity. The surface morphology of bare Au
surface and protein film formed on Au surface are shown
in Figure 4.
3. DEVELOPMENT OF A PROTEIN-BASEDMEMORY DEVICE AND ITSWORM NATURE
Electron transport in nature, such as photoelectric con-
version and long-range electron transfer in photosynthetic
conversion, is one of the most efficient mechanisms
to control information and energy in natural and arti-
ficial systems because of its unidirectional transport
and high conversion yields.49–53 Particularly, sequential
redox reactions by proteins or enzymes are utilized for
information transfer and energy storage in major natu-
ral mechanisms. Therefore, redox reactions are of great
interest in various fields including biology and chem-
istry. Currently, these redox reactions by proteins or
enzymes are divided mainly into two fields, alternative
energy and next-generation electronic devices to enhance
their energy conversion efficiency and information con-
trol efficiency by mimicking biomechanisms using redox
reactions.54�55 Bioelectronic devices, mimicking biomech-
anisms, have recently emerged to combine biomaterial
functions into conventional electronic format.56�57 Several
methods have been proposed for developing new concept-
memory devices to overcome the physical and techni-
cal limitations of the conventional silicon-based memory
devices.58–61 A memory device using conducting polymers
also has been reported,62 but this device can be writ-
ten only once and cannot be used as programmable
memory. Recently, a flash memory with protein-mediated
assembly has been proposed,63 and an electrochemi-
cal memory device using conducting polymer has been
fabricated.64
Even though programmable electrical bi-stability was
observed in a memory device made from polymer
thin film65 and logical circuits have been developed
using biomolecules and organic molecules as active
components,66 electronic memory using these organic
materials and biomolecules is still in the early stages
of demonstrating the function of new concept-memory
device. The concept of biomemory device consisting of
metalloprotein by their electrochemical reaction67 has been
reported, but here we review the multiple reading of the
stored charge and temperature dependency of the stored
charge.
Ferredoxin is one of the promising biomolecules
applicable to biomemory systems because of its redox
properties and electron transport mechanisms in nature.
Ferredoxins are ubiquitous in nature and play a vital
role in electron transport mechanisms. These proteins use
oxidation-reduction chemistry to transfer an electron from
a donor site to an acceptor site. The spinach ferredoxin
belongs to 2Fe–2S class of ferredoxin (fdx) that is sim-
ple iron-sulfur proteins containing a 2Fe–2S cluster. Ferre-
doxin is a small, soluble, generally very acidic protein
involved mainly as an electron carrier of low oxidation-
reduction potentials in fundamental metabolic processes.
Ferredoxin is the most abundant form of photosynthetic-
type ferredoxin present in spinach chloroplasts. The 2Fe–
2S cluster of fdx contains two atoms of iron, bridged by
two acid-labile, inorganic sulfide atoms, and coordinated
to four cysteinyl sulfurs in the polypeptide chain.68 Since
ferredoxin is a constituent in the acting site as an electron
transfer protein, it could be used as an electron accep-
tor in the development of molecular electronic devices by
mimicking biological mechanisms. Therefore, ferredoxin
438 J. Nanosci. Nanotechnol. 14, 433–446, 2014
Page 7
Yagati et al. Recombinant Protein-Based Nanoscale Biomemory Devices
Figure 5. Cyclic voltammogram obtained for ferredoxin immobilized
Au electrode showing the oxidation and reduction potentials for charge
write and erase functions of a memory device. Inset shows the open
circuit potential (OCP) for ferredoxin immobilized Au electrode, which
is used to read the stored charge.
proteins can be applied directly to verify this new concept
of memory function with biomaterial combined electronic
devices, due to their charge transfer and trap function as
well as their thermal69 and chemical stability.70
Cyclic Voltammetry (CV) for Spinach ferredoxin on
gold electrode in 10 mM Tris-HCl buffer (pH 7.4) in
potential range of 0.1 to −0�45 V versus Ag/AgCl at
a scan rate 50 mV/s was performed. Well-defined redox
peaks with an anodic redox wave at Epa =−0�12 V versus
Ag/AgCl and cathodic wave at Epc = −0�2 V Ag/AgCl
resulted as shown in Figure 5. This corresponds to the
redox process of [2Fe–2s]2+/1+ cluster in fdx. These redoxstates were used to charge storage and erase in the protein
molecule to perform memory functions. The open circuit
potential observed was −0�05 V, which is used to read the
charged states.
Figure 6. (a) Arrhenius plot obtained for conductivity with increase in
temperature clearly shows two distinctly different regions; (b) �E� versusT plot of ferredoxin on Au electrode showing the variation in standard
potential with increase in temperature.
Temperature-dependent conductance measurements
were carried out to investigate the influence of temperature
on the switching mechanism of the device. The conduc-
tivity (�) was determined at a given temperature from the
plot of current (I) versus voltage (V ). Figure 6(a) indicatesthat the conductance greatly depends on temperature.
By increasing the temperature from 289 to 338 K, progres-
sive changes in structural and functional properties have
been monitored. Up to 320 K, the conductance showed a
linear, positive shift with increasing temperature but inter-
estingly enough, from 320–338 K the conductance begins
to decrease. Conductance of the protein increases with
the rise in temperature because it acquires extra thermal
energy, but after 320 K the conductance decreases due to
the thermal unfolding of the 2Fe–2S cluster in the protein.
Thermally induced unfolding and denaturation of ferre-
doxin were monitored by differential scanning calorime-
try (DSC) with 5 �C/min scan rate going from 20 to
80 �C. The complete denaturation of this protein was esti-
mated by differential scanning calorimetry at 342 K. The
thermally induced unfolding reaction of the ferredoxin
is irreversible71 due to the degradation of the cluster in
the unfolding state.72 Also, E� potentials were decreas-
ing with increasing temperature but up to 320 K, this
change in E� potential could be the structural fluctuations
that occur within the protein73 as shown in Figure 6(b).
At 320 K, slope change in entropy (�S� = dE�/dT ) dueto the non-spontaneous disruption of the Iron-sulfur clus-
ter is followed by the spontaneous complete unfolding of
the protein. This behavior is because of loss of iron-sulfur
cluster and denaturation of protein. Therefore, the mem-
ory function of this device is stable is up to 320 K above
which the device’s working performance will change, with
no information being stored.
For the analysis of the memory function, electrical volt-
age pulses for write-read-erase cycles were applied as
shown in Figure 7(a). In each memory cycle, an oxida-
tion voltage of −120 mV was applied to store the charge
in the protein layer; reduction voltage of −200 mV was
applied to erase all the stored charge in the device. Two
pulses of −50 mV (open circuit voltage) were applied
after oxidation and reduction voltages, to read the charged
states. To read the stored charge multiple times an OCP
of −50 mV was applied, and the current reveals that
the stored charge can be read multiple times. Figure 7(b)
shows the oxidation voltages of −120 mV can repro-
ducibly charges on the device. A set of three voltage
pulses of −50 mV (OCP) with small disconnecting times
showed the current switching, thereby maintaining the oxi-
dized state of the memory device. Finally, a voltage pulse
of −200 mV erases all the stored charge. The detailed
change of current (I) is a function of time for two mem-
ory cycles. Also, the device switching speed was tested
with fast write pulse in the write-read-erase-reset sequence
in 10−6∼10−9 sec duration without degradation in perfor-
mance. Notably, it has less than 20% decay in its charged
J. Nanosci. Nanotechnol. 14, 433–446, 2014 439
Page 8
Recombinant Protein-Based Nanoscale Biomemory Devices Yagati et al.
Figure 7. (a) Current response obtained upon application of sequential
redox potentials for operation of memory function (b) multiple times
reading test shows the current response upon multiple times of OCP
where Iwrite and Ierase correspond to write and erase currents and Ireadrepresenting the multiple times reading of the stored charge.
state after 2 weeks, but almost no change in 2 weeks con-
firms the nonvolatile nature of biomemory device.
4. ENHANCEMENT IN DATA STORAGECAPACITY IN PROTEIN-BASEDMEMORY DEVICES
4.1. Multi-Bit ApproachLogic and memory are the essential functions in normal
electronic products. All the developments of these memory
cells are driven by the market requirement for low cost. For
the memory cells, the traditional way of achieving low cost
per bit is by scaling size or multilevel storage, and great
developments have been achieved in this areas.74�75 Multi-
bit storage is attracting more and more research attention
due to the scaling method limited by photolithography.76
The strategy to increase memory density imposes a multi-
bit approach wherein the charge storage element contains
molecules with multiple redox states.77�78 Here, an efficient
way of azurin thin film formation on four Au electrodes
is introduced where each will act as a memory structure
to achieve the basic memory function with multiple bit
storage capacity.
The schematics for the experimental system for multi-bit
and multi-level systems are shown in Figure 8. The redox
properties of 4 different azurin variants (Ni type; Cu type;
Mn type; Fe type) were examined as each modified azurin
having different metals would affect the electrochemical
properties of the azurin variants. Cyclic voltammetry (CV)
of each self-assembled azurin variant layer showed differ-
ent redox potentials for each metal substitute, which can
be utilized for memory function.
Direct immobilized azurin variant layers have three dis-
tinct conducting states and each variant has different redox
potentials. Briefly, in the case of the Cu-type azurin, apply-
ing of an OP resulted in the transfer of electrons from the
immobilized azurin layer into the Au substrate, and pos-
itive charges were stored in the azurin layer. The reverse
process occurred during the reduction step. When the RP
was applied, the electrons were transferred back into the
azurin layer and the stored charge was erased. The open
circuit potential (OCP) was then used to read these charged
states. Using these charged states, different logic perfor-
mance can be realized by the application of these poten-
tials in a sequential manner. In the case of Fe-type azurin,
simple “write” and “erase” functions are studied. Also,
in the case of Ni type Azurin, apart from these two charged
states, an open circuit voltage will confirm (read) both
charged states making one kind logical sequence of write,
read, and erase functions. Apart from these, the Mn-type
azurin also displayed a WORM-type current response with
three defined parameters. In this case, when an OCP was
applied at three very small disconnecting times to read
the stored information, the observed current responses of
the stored charge could be read three times. Finally, in the
case of Fe-type azurin, potentials were applied to per-
form WORM-type memory with an OCP to confirm the
erase state of the biomemory. These results are shown in
Figure 9.
4.2. Multi-Level ApproachAn alternative approach for multi-bit functionality is by
mixing different redox active molecules whose potentials
are well separated. This multi-level approach is aimed
to store more than 1 bit of information. Here in this
approach, recombinant azurin containing cysteine residues
was produced by site-directed mutagenesis and immo-
bilized directly on Au surface; then, cytochrome c was
adsorbed onto the immobilized azurin layer by electrostatic
bonding. This heterolayer is composed of recombinant
azurin and cytochrome c, which is capable of storing the
two pairs of information, referred to as the multistate
memory. The electrons were flowing into the immobi-
lized azurin molecule when oxidation potential (OP) was
applied to the fabricated electrode, which was assigned
an information value of ‘1’. In contrast, when reduction
potential (RP) was applied, which was assigned an infor-
mation value of ‘0’, the stored electrons were released
from the azurin.
440 J. Nanosci. Nanotechnol. 14, 433–446, 2014
Page 9
Yagati et al. Recombinant Protein-Based Nanoscale Biomemory Devices
Figure 8. (a) Schematic diagram shows the immobilization of 4 different variants of azurin on Au electrodes for the basic operating mechanisms of
4-bit operation of the fabricated biomemory chip. (b) The electron-transfer mechanism of a cytochrome c/recombinant azurin layer on Au surface.
Cyclic voltammetry of the recombinant azurin/
cytochrome c heterolayer immobilized on the Au surface
shows two anodic waves at Epa of 0.062 V versus Ag/AgCl
and cathodic wave at Epc of 0.184 V versus Ag/AgCl,
which correspond to the redox process of recombinant
azurin center Cu2+/1+. In addition, the anodic wave at Epa
of 0.131 V versus Ag/AgCl and cathodic wave at Epc of
0.324 V versus Ag/AgCl were observed, which correspond
to the redox process of the cytochrome c center Fe3+/2+
shown in Figure 10(a).
Figure 10(b) shows the faradaic currents obtained for a
multi-level operation of a protein-based memory device.
Faradaic currents were obtained when the heterolayer was
applied with an oxidation potential of 0.324 V versus
Ag/AgCl and the cytochrome c layers became oxidized.
The electrode was left open for a small duration of time,
and again when connecting the OCP of 0.104 V versus
Ag/AgCl, large amplitude currents were monitored in the
absence of background currents. When a reduction poten-
tial 0.131 V was applied, the stored charge was erased.
This was evident by the small amplitude currents. In the
same way, faradaic currents were obtained when an oxida-
tion potential of 0.184 V versus Ag/AgCl was applied to
the heterolayer. Under these conditions, the recombinant
azurin layers became oxidized, which was then reduced
when a reduction potential of 0.062 V was applied to the
recombinant azurin layer. Then, the electrode was left open
for a small duration of time and again connecting the OCP
of 0.150 V versus Ag/AgCl resulted in the appearance
of small amplitude currents in the absence of background
currents. Here, applying 0.184 V was another writing step
whereas applying 0.150 V was another reading step with
respect to the measurement of currents. Finally, the appli-
cation of 0.062 V erased the stored charge in the recom-
binant azurin layer. Similarly, when two pairs of redox
potentials were applied to the protein film for duration of
320 ms transient currents for the two pairs of the charge
to write, read, and erase functions were clearly monitored,
which are prerequisites for any molecular memory storage
device.
J. Nanosci. Nanotechnol. 14, 433–446, 2014 441
Page 10
Recombinant Protein-Based Nanoscale Biomemory Devices Yagati et al.
Figure 9. (a) Verification of multi-functional memory performance;
(a) simple write and erase function on Cu-type azurin (b) Ni-type azurin
with read confirmation of the charged states (c) Fe-type azurin with mul-
tiple reading performance (d) Mn-type azurin of multiple read and also
confirmation of its erased states.
Figure 10. (a) Cyclic voltammogram showing the behavior of
cytochrome c and azurin hetero conjugates onto the gold surface;
(b) characteristics of multistate protein-based biomemory device; two
pairs of input pulse potentials of oxidation (charge write 1, 2), open-
circuit (charge read 1, 2), and reduction (charge erase 1, 2) potentials
were applied to gold electrodes simultaneously, having a pulse width of
20 ms. (ii) The corresponding charging currents were measured for a
total duration of 320 ms.
5. NANOBIOMEMORY DEVICE COMPOSEDOF RECOMBINANT AZURIN ONNANOSTRUCTURED SURFACES
Due to continuously increasing demand for ultimate
miniaturization of electronic systems, molecular electronic
devices are currently thriving as alternative technologies
because of their promising potential in writing, read-
ing, and processing of information at the nanoscale.79�80
Recently, colloidal or nanosphere lithography has emerged
as an effective and inexpensive method for patterning
nanostructures over a large area in which the colloidal
nanospheres act as a mask. Regular pattern of desired
material can be obtained after deposition and the removal
of mask.81–83 Electrical conduction in the self-assembled
electro-active protein on nanostructured electrode surfaces
that are well swollen with electrolyte/solvent has been
a subject of great interest for the fundamental studies
on the mechanism and kinetics of charge transport as
well as potential applications (such as optical/electronic
devices, biosensors, and bioelectronics) based on their
redox properties.84–87 Development is occurring towards
a nanoscale protein-based memory device, in which the
recombinant protein is self-assembled on Au pattern-
formed indium tin oxide (ITO) coated glass plate. By using
electrochemical scanning probe microscopy with in-situcyclic voltammetry, memory function experiments were
performed. To realize a practical biomemory device, a sin-
gle detection system is not enough. To ensure the possi-
bility of multiple detection systems to read the stored and
erased charges states, optical and magnetic detection sys-
tems were introduced.
The schematics for the experimental setup and
the nanoscale biomemory principle are shown in
Figure 11. Electrochemical experiments on Az/Au–ITO
were conducted by ECSTM with in-situ cyclic voltamme-
try. Experiments were carried out in 10 mM HEPES buffer
solution, where Az/Au–ITO substrate acts as the work-
ing electrode (vs. Ag/Ag+� as shown in Figure 11(a). The
operation mechanism of the proposed nanoscale biomem-
ory is depicted in Figure 11(b). To examine the redox
properties for analyzing the reduction and oxidation prop-
erties of azurin at nanoscale, ECSTM with in-situ cyclic
voltammetry was performed. Obviously, the cyclic voltam-
mogram of bare Au nanopattern did not reveal any redox
peaks; however, the CV for azurin on the Au nanopat-
tern clearly depicted the peaks for reduction and oxida-
tion of Az molecules at −0�04 V and 0�26 V respectively
at a scan rate of 50 mV/s, which corresponded to the
redox process of the Cu2+/1+ center in azurin, as shown in
Figure 11b(i), (ii). The formal reduction potential (E1/2�,calculated as E1/2 = �Epc +Epa�/2, was 0.11 V. A set of
images of the same sample area of protein on Au triangle
was obtained at constant bias while varying the potential.
It can be seen that there was a variation in the conduc-
tance of the redox molecules upon varying the potential,
442 J. Nanosci. Nanotechnol. 14, 433–446, 2014
Page 11
Yagati et al. Recombinant Protein-Based Nanoscale Biomemory Devices
Figure 11. (a) Schematic diagrams for azurin on Au nanodots with ECSTM setup and (b) electron transfer mechanism in Az/Au nanodots during
oxidation and reduction potentials. In-situ CV’s for (i) bare and (ii) Az/Au/ITO at a scan rate of 50 mV s−1.
which was reflected in a difference in apparent height with
respect to the background. Figure 12 shows the images for
the applied potential of 0.26 and −0�04 V, respectively for
a bias voltage of 100 mV for oxidation and reduction states
of azurin moelcules. It was assumed that at Ebias = 0 V, the
redox state of azurin is vacant with energy higher than the
fermi energy of the substrate and tip. When a negative bias
was applied to the substrate and the tip was kept at the
positive potential, the vacant state energy and fermi energy
of the substrate were matched, allowing electrons to be
transferred from the substrate to the biomolecule, which
lowers the thermal activation allowing the biomolecule to
transfer the electron to the tip.88�89
Figure 12. ECSTM images obtained for immobilized azurin molecules
on Au substrate for different bias potentials for (a) oxidation and
(b) reduction, respectively.
The topography for the immobilized azurin molecules
on Au surface was obtained with ECSTM at these two
distinct conducting states as shown in Figure 12. How-
ever, bias potentials have induced only specific regions
between tips and substrates. The typical bright spots seen
when tuning the substrate potential in a region, appear to
be strongly potential-dependent of redox potentials. Such
kind of behavior is consistent with a resonant nature of the
current measured in STM experiments in the Au-adsorbed
azurin molecules. Figure 12(a) points at assembled azurin
at oxidation voltage (Write/Store) and Figure 12(b) shows
assembled azurin at reduction voltage (Erase). The shape
and height have been changed more and more as specific
potentials were applied. From these images, it is assumed
that there were two conductive states for charge storage.
A brighter spot emerged, which was interpreted as a small
island of oxidized azurin molecules containing a Cu cen-
ter. When a voltage of 0.26Eox was applied, an electron
tunnel formed adsorbed azurin molecules on the Au sub-
strate, and lead to a storage of positive charges in the
azurin molecules. In contrast, when a voltage of −0�04Ered
was applied, the electrons were transferred back to azurin
molecules and the stored charge was erased, which was per-
formed to erase the stored charge. In this case, the brighter
spot on the Au triangle disappeared, which confirmed that
the oxidized azurin molecules were reduced.
The possibility of an optical detection system was ana-
lyzed where an LSPR-based optical detection system90�91
J. Nanosci. Nanotechnol. 14, 433–446, 2014 443
Page 12
Recombinant Protein-Based Nanoscale Biomemory Devices Yagati et al.
Figure 13. (a) Absorption spectra of Au pattern on ITO, Az/Au–ITO clearly shows a broad peak at 732 nm and a shoulder peak at 400 nm (b) UV
spectra obtained upon sequential application of oxidized and reduced potential clearly shows the variations in the peak intensity in both oxidized and
reduced forms and (c) zoomed portion of the spectra for the shoulder peak at 400 nm and (d) zoomed portion of the image at the 732 nm peak.
is utilized. Figure 13(a) shows the absorption peak of the
Au nanopattern with azurin immobilized on Au nanopat-
tern, which clearly depicts that azurin due to Cu2+
results in the absorbance at �max = 400 nm (small peak)
and at �max = 730 nm (broad peak). The absorbance at
400 nm is attributed to His → Cu charge transfer transi-
tion, which has a low extension coefficient.92�93 The tran-
sition at 730 nm arises from the d–d transitions, which
also has a low extinction coefficient. So upon applica-
tion of the oxidation and reduction potential and simul-
taneous observation of UV signal, the variation in the
peak intensity shifts down and up at every step shown
in Figure 13(b). This observation leads to optical detec-
tion of read-out mechanism for the operation of memory
device. Figures 13(c), (d) shows the zoomed portions indi-
cating the variations of the shoulder and intense peaks
upon application of oxidation and reduction potentials. All
these spectral changes are a clear indication of the trans-
formation of Az and its redox states, and the process was
reversible.
Earlier reports reveal that reduced azurin has diamag-
netic and oxidized azurin behaves as paramagnetic.94 From
this magnetic behavior, there is a chance for another
read-out mechanism of the charged states. Hence, mag-
netic output can also be used to perform the molec-
ular switch for the read-out mechanism of the switch.
Thus, ESR on the oxidized and reduced forms of azurin
could be detected and magnetic output can also be used
for reading the state of the molecular switch. Based on
these findings, azurin on a hexagonal nanopattern can
be used as a nanoscale switch on binary logic circuits.
The applied electrochemical voltage can be considered
the input; when oxidation voltage was applied, the azurin
turned to its oxidized form (Input 1), and when reduc-
tion voltage was applied the azurin turned back to original
form (Input 0). Four output mechanisms can be detected
from these inputs: (a) an absorption peak at 400 nm, (b) an
absorption peak at 732 nm, (c) the ESR signal, and the
(d) electrochemical output signals. So, a binary logic gate
can be realized from these four outputs.
Based on the electrochemical nature of the protein,
it was reversibly converted between its redox states (oxi-
dized and reduced forms) for memory switching oper-
ations. Additionally, the optical and magnetic responses
can be used as a readout mechanism. Because the azurin
was immobilized on an Au pattern, it can be addressed
locally, operates at very low voltages with its stability,
robustness, and reversibility, making the system a very
promising memory device. The proposed approach will
directly impact the implementation of recent advances in
biotechnology such as biocomputing systems which use
biomolecules, such as DNA/proteins, to perform compu-
tational calculations involving storing, retrieving, and pro-
cessing data.
444 J. Nanosci. Nanotechnol. 14, 433–446, 2014
Page 13
Yagati et al. Recombinant Protein-Based Nanoscale Biomemory Devices
6. FUTURE PERSPECTIVEThe discussed results back the conviction that proteins
are interesting materials in developing nanoscale mem-
ory devices. Not only for developing memory devices,
but also due to their structural and functional versatility,
they are potentially suited to a wide range of applications.
Further development toward commercial applications can
be envisaged as (at least) tenfold. On the edge of
molecular electronics, protein-based circuits can be con-
ceived of as an interesting solution, once appropriate
techniques for patterning and interconnecting devices are
made available.95�96 The nanoscopic size of the comput-
ing units, down to a single molecule, and their possibly
low cost can be the strengths of new molecular machines.
Conversely, new architectures and computing paradigms
may be required to meet the constraints of this novel
biohardware.97�98 In the memory device operation, memory
characteristics should be obtained without the operation
of STM, which demands the proper external connecting
circuitry mandatory for developing a new generation of
nanoscale memory devices. Also these molecules coupled
to Au nanopattern are of great interest and could be the
possible candidate for next-generation memory devices.
Using the self-assembly properties of proteins and possi-
bly other molecules (e.g., DNA) can lead to very complex
macroscopic structures, achieving finer-than-lithographic
resolution without using traditional lithography.99–101 This
is, in fact, what is needed to build memories and com-
plex computing circuits; that is, this may be the milestone
marking the beginning of the true post-silicon era.
ABBREVIATIONSAEQ Aequorin
AFM Atomic force microscope
AZ Azurin
CV Cyclic voltammetry
DNA Deoxyribonucleic acid
DSC Differential scanning calorimetry
ECSTM Electrochemical scanning tunnelling
microscope
ESR Electron spin resonance
GBP Glucose binding protein
ITO Indium tin oxide
MOSFET Metal oxide semiconductor field effect
transistor
OCP Open circuit potential
OP Oxidation potential
RNA Ribonucleic acid
RP Reduction potential
SAM Self-assembled monolayer
SPR Surface plasmon resonance
STM Scanning tunnelling microscope
TMV Tobacco mosaic virus
UV Ultra violet
WORM Write-once-read-many times.
Acknowledgment: This research was supported
by the Nano/Bio Science and Technology Program
(M10536090001-05N3609-00110) of the Ministry of Edu-
cation, Science and Technology (MEST), by the National
Research Foundation of Korea (NRF) grant funded by the
Korean government (MEST) (2009-0080860), and by the
Ministry of Knowledge Economy (MKE) and Korea Insti-
tute for Advancement in Technology (KIAT) through the
Workforce Development Program in Strategic Technology.
References and Notes1. W. Wang and L. Chi, Acc. Chem. Res. 45, 1646 (2012).2. S. Ueno and S. Fujihara, J. Alloys Compd. 541, 338 (2012).3. V. Suresh, M. S. Huang, M. P. Srinivasan, and S. Krishnamoorthy,
J. Mater. Chem. 22, 21871 (2012).4. A. Asano, M. Omichi, S. Tsukuda, K. Takano, M. Sugimoto, and
A. Saeki, J. Phys. Chem. C 116, 17274 (2012).5. S. A. Ibanez, P. I. Fierens, R. P. J. Perazzo, G. A. Patterson, and
D. F. Grosz, Euro. Phys. J. B 76, 49 (2010).6. Y. K. Chen and S. Y. Kung, J. Signal Process. Sys. 53, 217 (2008).7. C. Schneider, A. Huggenberger, T. Sunner, T. Heindel, M. Strauß,
S. Gopfert, P. Weinmann, S. Reitzenstein, L. Worschech, M. Kamp,
S. Hofling, and A. Forchel, Nanotechnology 20, 434012 (2009).8. S. Cuenot, C. Fretigny, S. Demoustier-Champagne, and B. Nysten,
Phys. Rev. B 69, 165410 (2004).9. R. Valiev, Nat. Mater. 3, 511 (2004).10. E. Roduner, Chem. Soc. Rev. 35, 583 (2006).11. C. Nicolini, Biosens. Bioelectron. 10, 105 (1995).12. S. J. Kim, B. Kim, J. Jung, D. H. Yoon, J. Lee, S. H. Park, and
H. J. Kim, Appl. Phys. Lett. 100, 103702 (2012).13. S. Son, G. Kim, K. Singha, S. Park, M. Ree, and W. J. Kim, Small
7, 2991 (2011).14. H. Furukawa, T. Endo, Y. Yanagida, and T. Hatsuzawa, Jpn. J. Appl.
Phys. 47, 4810 (2008).15. N. Mohanty and V. Berry, Nano Lett. 8, 4469 (2008).16. C. Paternolli, P. Ghisellini, and C. Nicolini, IET Nanobiotechnology
1, 22 (2007).17. M. Aizawa, Electrochemistry 69, 921 (2001).18. C. P. Collier, J. O. Jeppesen, Y. Luo, J. Perkins, E. W. Wong, J. R.
Heath, and J. F. Stoddart, J. Am. Chem. Soc. 123, 12632 (2001).19. Huang, X. Duan, Y. Cui, L. J. Lauhon, K.-H. Kim, and C. M.
Lieber, Science 294, 1313 (2001).20. X. Duan, Y. Huang, and C. M. Leiber, Nano Lett. 2, 487 (2002).21. T. Goda and Y. Miyahara, Langmuir 28, 14730 (2012).22. K. T. Hamorsky, C. M. Ensor, Y. Sei, and S. Daunert, Angew. Chem.
Int. Ed. 47, 1 (2008).23. Y. Ogura, T. Beppu, M. Takinoue, A. Suyama, and J. Tanida, Control
of DNA molecules on a microscopic bead using optical techniques
for photonic DNA memory, DNA Computing, edited by A. Carbone
and N. A. Pierce, Springer, Berlin, Heidelberg (2006), pp. 213–223.24. M. Pita and E. Katz, J. Am. Chem. Soc. 130, 36 (2008).25. A. S. Deonarine, S. M. Clark, and L. Konermann, Future Gener.
Comp. Sy. 19, 87 (2003).26. A. Tinazli, J. Piehler, M. Beuttler, R. Guckenberger, and R. Tampe,
Nat. Nanotechnol. 2, 220 (2007).27. M. Knez and U. Gosele, Nat. Nanotechnol. 1, 22 (2006).28. T. S. Ham, S. K. Lee, J. D. Keasling, and A. P. Arkin, PLOS One
3, e2815 (2008).29. K. Fujita, N. Nakamura, H. Ohno, B. S. Leigh, K. Niki, H. B. Gray,
and J. H. Richards, J. Am. Chem. Soc. 126, 13954 (2004).30. S. H. Lee, Y. Jung, and R. Agarwal, Nat. Nanotechnol. 2, 626 (2007).31. S. W. Lee, W. J. Chang, R. Bashir, and Y. M. Koo, Biotechnol.
Bioprocess. Eng. 12, 185 (2007).
J. Nanosci. Nanotechnol. 14, 433–446, 2014 445
Page 14
Recombinant Protein-Based Nanoscale Biomemory Devices Yagati et al.
32. J. Y. Jyoung, S. H. Hong, W. Lee, and J.-W. Choi, Biosens.Bioelectron. 21, 235 (2006).
33. W. Lee, K. S. Park, Y. W. Kim, W. H. Lee, and J.-W. Choi, Biosens.Bioelectron. 20, 2292 (2005).
34. D. Y. Zhang, A. J. Turberfield, B. Yurke, and E. Winfree, Science318, 1121 (2007).
35. H. B. Gray and J. R. Winkler, Annu. Rev. Biochem. 65, 537 (1996).36. G. McLendon, Acc. Chem. Res. 21, 160 (1988).37. R. J. Tseng, C. Tsai, L. Ma, J. Ouyang, C. S. Ozkan, and Y. Yang,
Nat. Nanotechnol. 1, 72 (2006).38. S. D’Amico, G. Maruccio, P. Visconti, E. D’Amone, A. Bramanti,
R. Cingolani, and R. Rinaldi, IEE Proc.-Nanobiotechnol. 151, 173(2005).
39. B. M. Frezza, S. L. Cockroft, and M. R. Ghadiri, J. Am. Chem.Soc. 129, 14875 (2007).
40. J. W. Choi, B. K. Oh, J. Min, and Y. J. Kim, Appl. Phys. Lett.91, 263902 (2007).
41. A. K. Yagati, S. U. Kim, J. Min, and J. W. Choi, J. Nanosci.Nanotechnol. 10, 3220 (2010).
42. S. U. Kim, T. Lee, J. H. Lee, A. K. Yagati, J. Min, and J. W. Choi,
Ultramicroscopy 109, 974 (2009).43. G. Fletcher, S. Mason, J. Terrett, and M. Soloviev, J. Nanobiotech-
nol. 1, 1 (2003).44. F. Caruso, E.Rodda, and D. N. Furlong, J. Colloid Interf. Sci.
178, 104 (1996).45. S. Abraham, M. S. Bahniuk, and L. D. Unswoth, Biointerfaces
7, 40 (2012).46. W. J. Chen, M. N. Li, P. Yakufu, X. M. Ling, H. Qi, J. H. Xiao,
and Y. Wang, Anal. Biochem. 423, 1 (2012).47. S. U. Kim, J. H. Lee, T. Lee, J. Min, and J. W. Choi, J. Nanosci.
Nanotechnol. 10, 3241 (2010).48. J. Min, S. U. Kim, Y. J. Kim, C. H Yea, and J. W. Choi, J. Nanosci.
Nanotechnol. 8, 4982 (2008).49. R. I. Cukier and D. G. Nocera, Annu. Rev. Phys. Chem. 49, 337
(1998).50. K. Fujita, N. Nakamura, H. Ohno, B. S. Leigh, K. Niki, H. B. Gray,
and J. H. Richards, J. Am. Chem. Soc. 126, 13954 (2004).51. H. B. Gray and J. R. Winkler, Annu. Rev. Biochem. 65, 537 (1996).52. E. I. Solomon, R. K. Szilagyi, S. D. George, and L. Basumallick,
Chem. Rev. 104, 419 (2004).53. M. Ubbink, J. A. R. Worrall, G. W. Canters, E. J. J. Groenen, and
M. Huber, Annu. Rev. Biophys. Biomol. Struct. 31, 393 (2002).54. S. W. Lee, W. J. Chang, R. Bashir, and Y. M. Koo, Biotechnol.
Bioproc. E 12, 185 (2007).55. B. Y. Jeon, S. J. Kim, D. H. Kim, B. K. Na, D. H. Park, H. T. Tran,
R. Zhang, and D. H. Ahn, Biotechnol. Bioproc. E 12, 566 (2007).56. S. H. Paek, J. H. Cho, I. H. Cho, Y. K. Kim, and B. K. Oh, Bio.
Chip J. 1, 1 (2007).57. S. W. Park, J. H. Lee, H. C. Yoon, and S. S. Yang, Bio. Chip J.
1, 35 (2007).58. R. J. Tseng, J. Huang, J. Ouyang, R. B. Kaner, and Y. Yang, Nano
Lett. 5, 1077 (2005).59. D. Y. Zhang, A. J. Turberfield, B. Yurke, and E. Winfree, Science
318, 1121 (2007).60. C. P. Collier, J. O. Jeppesen, Y. Luo, J. Perkins, E. W. Wong, J. R.
Heath, and J. F. Stoddart, J. Am. Chem. Soc. 123, 12632 (2001).61. Y. Huang, X. Duan, Y. Cui, L. J. Lauhon, K. H. Kim, and C. M.
Lieber, Science 294, 1313 (2001).62. Z. Liu, A. A. Yasseri, J. S. Lindsay, and D. F. Bocian, Science
302, 1543 (2003).63. J. H. Sachon, H. Meng, and Z. Bao, Science 294, 2138 (2001).64. G. Y. Tseng and J. C. Ellenbogen, Science 294, 1293 (2001).65. C. H. Wal, M. D. Eisaman, A. Andre, R. L. Walsworth, D. F.
Philips, A. S. Zibrov, and M. D. Lukin, Science 301, 196 (2003).
66. S. Moller, C. Perlov, W. Jackson, C. Taussig, and S. R. Forrest,
Nature 426, 166 (2003).67. J. Sarkar, S. Tang, D. Shahrjerdi, and S. K. Baneerjee, Appl. Phys.
Lett. 90, 103512 (2007).68. A. Messerschmidt, R. Huber, T. Poulos, and K. Wieghardt, Hand-
book of Metalloproteins, John Wiley & Sons, New York (2001),p. 1472.
69. K. Tomizaki and H. Mihara, J. Am. Chem. Soc. 129, 8345 (2007).70. J. W. Choi, B. K. Oh, J. Min, and Y. J. Kim, Appl. Phys. Lett.
91, 263902 (2007).71. J. Meyer, M. D. Clay, M. K. Johnson, A. Stubna, E. Munck,
C. Higgins, and P. Wittung-Stafshede, Biochemistry 41, 3096
(2002).72. C. La-Rosa, D. Milardo, D. Grasso, R. Guzzi, and L. Sprotelli,
J. Phys. Chem. 99, 14864 (1995).73. G. Battistuzzi, M. Borsari, A. Ranieri, and M. Sola, J. Am. Chem.
Soc. 124, 26 (2002).74. H. H. Berger and S. K. Wiedmann, IEEE J. Solid-State Circuits 7,
340 (1972).75. K. Itoh, K. Sasaki, and Y. Nakagome, Proc. IEEE 83, 524 (1995).76. A. K. Yagati, S. U. Kim, J. Min, and J. W. Choi, Biosens. Bioelec-
tron. 24, 1503 (2009).77. H. Imahori, H. Norieda, H. Yamada, Y. Nishimura, I. Yamazaki,
Y. Sakata, and S. Fukuzumi, J. Am. Chem. Soc. 123, 100 (2001).78. Q. Li, G. Mathur, S. Gowda, S. Surthi, Q. Zhao, L. Yu, J. S.
Lindsey, D. F. Bocian, and V. Misra, Adv. Mater. 16, 133 (2004).79. R. Waser and A. Rudiger, Nat. Mater. 3, 81 (2004).80. Z. Yao, H. W. Ch. Postma, L. Balents, and C. Dekker, Nature
402, 273 (1999).81. C. L. Haynes and R. P. Van Duyne, J. Phys. Chem. B 105, 5599
(2001).82. A. Kosiorek, W. Kandulski, H. Glaczynska, and M. Giersig, Small
1, 439 (2005).83. A. K. Yagati, T. Lee, J. Min, and J. W. Choi, Biosens. Bioelectron.
40, 283 (2013).84. W. Lee, B. K. Oh, Y. M. Bae, S. H. Paek, W. H. Lee, and J. W.
Choi, Biosens. Bioelectron. 19, 185 (2003).85. B. K. Oh, Y. K. Kim, K. W. Park, W. H. Lee, and J. W. Choi,
Biosens. Bioelectron. 19, 1497 (2004).86. J. W. Choi, K. W. Park, D. B. Lee, W. Lee, and W. H. Lee, Biosens.
Bioelectron. 20, 2300 (2005).87. J. W. Choi, Y. S. Nam, and W. Lee, Curr. Appl. Phys. 2, 79 (2002).88. S. Corni, J. Phys. Chem. B 109, 3423 (2005).89. W. Schmickler and N. Tao, Electrochim. Acta 42, 2809 (1997).90. T. K. Sau, A. L. Rogach, F. Jackel, T. A. Klar, and J. Feldmann,
Adv. Mater. 22, 1805 (2010).91. E. Hao, R. C. Bailey, G. C. Schatz, J. T. Hupp, and S. Li, Nano
Lett. 4, 327 (2004).92. E. Prenesti and S. Berto, J. Inorg. Biochem. 88, 37 (2002).93. M. G. Savelieff and Y. Lu, Inorg. Chim. Acta 361, 1087 (2008).94. C. M. Groeneveld and G. W. Canters, Eur. J. Biochem. 153, 559
(1985).95. D. T. Pham, H. Subbaraman, M. Y. Chen, X. C. Xu, and R. T.
Chen, IEEE Trans. Nanotechnol. 11, 44 (2012).96. A. Bandyopadhyay, A. K. Ray, A. K. Sharma, and S. I. Khondaker,
Nanotechnology 17, 227 (2006).97. S. P. Han, H. T. Maune, R. D. Barish, M. Bockrath, and W. A.
Goddard, Nano Lett. 12, 1129 (2012).98. I. M. Kullic, Biosystems 45, 45 (1998).99. E. Stulz, Chem. Eur. J. 18, 4456 (2012).
100. R. M. Zadegan and M. L. Norton, Int. J. Mol. Sci. 13, 7149 (2012).101. X. Shen, C. Song, J. Wang, D. Shi, Z. Wang, N. Liu, and B. Ding,
J. Am. Chem. Soc. 134, 146 (2012).
Received: 3 August 2013. Accepted: 26 August 2013.
446 J. Nanosci. Nanotechnol. 14, 433–446, 2014