Nanostructured Memristor Sensor Mimics Acetylcholinesterase
(ACHE) Active Sites In The Gorge For fM Detection Of
Acetylcholine
E. T. Chen1*, J. Thornton2 C. Ngatchou1and S-H. Duh3 1Advanced
Biomimetic Sensors, Inc., 13017 Wisteria Dr, #109, Germantown, MD
20874; 2 Bruker Nano, 19 Fortune Dr., Billerica, MA 01821;
3University of Maryland Medical System, 22 South Greene St,
Baltimore,
MD 21201; * To whom it should be contacted:
[email protected].
ABSTRACT
Many diseases including cancer, diabetes, brain injury,
epilepsy, Parkinson’s, autism and Alzheimer’s involve the
dysregulation of acetylcholinesterase (ACHE) causing inappropriate
production of the neurotransmitter acetylcholine (ACH). Providing a
nonenzymatic detection device for ACH with rapid detection time and
high sensitivity is paramount to avoid time consuming assays and
protein interferences. We report new types of nanostructured
memristor biomimetic ACHE sensors developed for detection of ACH
without using nature enzymes and are reagent-free. Memristor sensor
#1 was made for mimicking the normal active sites in the ACHE gorge
and memristor sensor #2 was made for mimicking the mutated gorge
with damaged eternal lining by knockdown of the 14 aromatic
residues. Multiple polymers were cross linked and self-assembled on
gold chips forming a “Flat-bridge/nanopore ACHE Gorge” for sensor
#1 and a “Vertical bridged-nanopore ACHE gorge for sensor #2; and
the i-V curves were observed of hysteresis loops. Results obtained
by a Chronoamperometric (CA) method for quantitation of ACH
revealed that sensor #1 has Detection of Limits (DOL) of ACH
9.7x10-16 M/cm2 over 10-15 to 10-6 M and an accuracy 101 ± 8% using
spiked NIST SRM965A human sera against calibration curve. Sensor #2
was unable to detect ACH directly. Delta Slow-Wave-Sleeping (SWS)
synapse firing from the two biomimetic neuron networks are
compared. A 5.5-fold higher intensity for Sensor 1 was observed.
Key Words: Nanobiomimetic Memristor Electrochemical Sensing;
Acetylcholinesterase (ACHE) Gorge; Acetylcholine; Reagent-less;
Slow-Wave-Sleeping (SWS); Nanobiomimetic Hyppocampal-Neocortical
Models.
INTRODUCTION
Acetylcholinesterase (ACHE) is a very important
hydrolase. It is found in nerve and muscle, central and
peripheral tissues, motor and sensory fibers, and cholinergic and
noncholinergic fibers. Its primary function in high rate hydrolysis
activity for terminating synaptic impulse transmission of
neurotransmitter acetylcholine (ACH) [1-3] and ACHE’s
non-cholinergic functions such as cellular proliferation are well
known [4-5]; ACHE dysregulation promoted cancers were reported
everywhere [6-8]. The
neurotransmitter ACH also plays an important role in spatial and
contextual learning and memory [9-12]. During slow-wave sleep,
however, declarative memory consolidation is particularly strong
[13]. Pesticides (herbicides, fungicides, insecticides) widely used
in the agriculture and industry, are neurotoxic compounds which
irreversibly inhibit ACHE, resulting in the accumulation of the
neurotransmitter ACH. Therefore, it is important to characterize
the activity of ACHE through quantifying ACH with enhanced
sensitivity and simplicity as reviewed in literature [14-15].
Improving the biosensor performance of ACH is challenged for
unavoidable biofouling and nonspecific protein bounding caused
interference by utilizing nature enzyme or coenzyme [16-19].
Biomimetic electron-relaying system, which not only mimics the
active sites of the proteins, but also promotes direct
bio-communication between the artificial active sites and the
electrode by utilizing a nanostructured self-assembled membrane
(SAM) films offering an attractive pathway to enhance the
selectivity, sensitivity and environmental protectiveness. It was
discovered that the structures of biomimetic enzyme sensor
membranes played an important role in enabling selective detecting
of toxins for being able to distinguish isomers and different types
of cancers [16, 20-23]. However, a device with nature inspired
dynamic dipole ACHE gorge [1-3] design would be more closely
mimicking the environmental stimulation for catalyzing ACH. The
attempt of this project is to develop a device that is able to
mimic the gorge’s fast hydrolysis function without using antibody,
ACHE and any tracer.
In review of the recent advances of the ACH sensor development,
the most popular approaches are using ACHE nature enzyme or
coenzyme fabricated on the substrate surface [14-15, 24-26]. The
reported sensitivity reached 10 nM using this biosensor [14-15,
24-26]. As mentioned above, our approaches are to avoid using
nature ACHE, but set up a biomimetic ACHE gorge providing a
suitable conformational and functional “net” to attract the “fish”
of ACH for a “direct meaningful bio-communication” without any
reagent used, in order to be environmentally friendly and be able
to have an order of magnitude higher sensitivity and portability.
Further- more, a memristor type of sensor may closely mimic the
nature of the ACHE in hippocampus and thalamic, because the ACH
closely related to memory with a long, complex and
NSTI-Nanotech 2014, www.nsti.org, ISBN 978-1-4822-5827-1 Vol. 2,
2014 169
chaotic but still living relationship [27-29], hence the new
type of sensor may be revolutionizing the biosensor field.
Memristors and Memcapacitors are devices made of nanolayers that
have the capability to mimic neuronal synapse with a characteristic
of hysteresis loop in the i-V curve [30-34]. However, most of the
memristors and the memcapacitors are made of metal oxide materials
[29-34], that make mimicking the ACHE gorge’s function more
difficult. The purpose of this research is to develop a memristor
device that closely mimics the ACHE gorge with cross-linked
nanostructured polymers without using metal oxide. A “Healthy
Active Site ACHE Gorge” is defined as:
Ser200-His400-Glu327(Catalytic Site (CAS)) mimicked by Polyethylene
glycol diglycidyl ether (PEG)....imidazolyl-dimethyl-β-cyclodextrin
(M-CD)...triacetyl-β-cyclodextrin (T-CD) and W84 mimicked by
poly(4-vinylpyridine) (PVP); The 14 aromatic residues for gorge
lining were mimicked by excess amount of o-NPA (1:500-1000 of T-CD/
o-nithophenyl acetate (o-NPA)). F330, Y121 were mimicked by o-NPA,
and W279 was mimicked by PVP. By knock down all o-NPA out of the
network, we define the second device as “Mutated Active Site ACHE
Gorge” based on our hypothesis: Lacking of hydrophobic lining in
the gorge might be the key issue caused diseases, because the
nature of the ACHE gorge might be memristive.
EXPERIMENTAL
Fabrication of the Nanostructure Self-Assembling Membrane (SAM)
Gold Memristor Chip The nanostructured biomimetic “Mutated ACHE
Active Gorge” memrisor with the vertical bridged
conformational/nanopore was freshly prepared and fabricated on gold
chip. Polyethylene glycol diglycidyl ether (PEG),
triacetyl-ß-cyclodextrin (T-CD), poly(4-vinylpyridine) (PVP) were
purchased from Sigma. PVP was purified before use. The mono
imidazolyl derivative dimethyl ß-cyclodextrin (mM-ß-DMCD) was
generally synthesized according to the published procedures [35].
The appropriate amount of solutions of individual polymer and
reagents were prepared [36]. The mixture solution was made up by
mM-ß-DMCD, T-CD, PEG and PVP. The 16 channel gold electrode chips
were purchased (Fisher Scientific). The mixture solution was
injected onto the surface of the electrode, was incubated for 48
hrs at 37ºC [36] and all other clean procedures were followed by
citation 36. This memristor was used as Sensor 2. The “Healthy
Active ACHE Gorge” memrisor with the flat bridged
conformation/nanopore was freshly prepared by adding appropriate
amount of o-nitrophenyl acetate (o-NPA) into the above described
mixture solution used for fabricating the vertical bridged
memristor. All other procedures were
followed as cited in literature 36. This memristor was used as
Sensor 1. Characterization of the Membrane
The morphology of the AU/SAM was characterized using an Atomic
Force Microscope (AFM) (model Multimode 8 ScanAsyst, Bruker, PA).
Data Collected in PeakForce Tapping Mode. Probes used were
ScanAsyst-air probes (Bruker, PA). The silicon tips on silicon
nitride cantilevers have 2-5 nm radius. The nominal spring constant
0.4N/m was used. Figure 1 illustrates the 3D vertical
conformational bridge structure with “breathing nanopore” of the
AFM images as sensor 2. Figure 2 illustrates the 3D flat
conformational bridge structure with “breathing nanopore” of the
AFM images of the Biomimetic Memristor 1.
Fig 1(L). 3D vertical conformational PDC bridge structure of
the AFM images of the Biomimetic memristor 2. Fig 2 (R) 3D
horizontal conformational bridge structure of
the AFM images of the Biomimetic memristor 1.
Frequency Affects on Memristor’s Performance
Evaluations of frequency affect on memristors’ performance were
conducted by Cyclic Voltammetric method (CV) in pH 7.40 saline
solution at room temperature. For Sensor 1, 20 Hz to 500 Hz and 1
pM ACH was used with o-NPA. For Sensor 2, it had to be first used
in the presence of 2 mM o-NPA and then 10 μM ACH can be measured,
not requiring pM concentrations. Ratios of relative signal strength
were used for comparison, i.e., j/cm2/unit ACH concentration, for
Sensor 1 and 2, respectively, at the same frequency.
Quantitation of ACH Chronoamperometric method was used for
quantitation of ACH (Acetylcholine chloride, Sigma) in pH 7.4 PBS
at initial applied potential of -50 mV, then -200 mV for detection
of ACH, the final concentration range over 10-15 to 10-6 M for
sensor 1 at room temperature. Because Sensor 2 had difficulty
detecting ACH, hence, we choose Sensor 2 to detect o-NPA instead,
in the concentration range over 10-10 to 10-4M in the presence of
10 μM ACH using an electrochemical work station (Epsilon, BASi, IN)
with the companied software package. Origin 9.0 was used for all
statistic data analysis and figure plotting.
NSTI-Nanotech 2014, www.nsti.org, ISBN 978-1-4822-5827-1 Vol. 2,
2014170
Assessing Quality of Slow-Wave Sleeping (SWS)
The Double Step Chronopotentiometry (DSCPO) method was used for
assessing quality of slow-wave-sleeping of the newly developed
nanobiomimetic hyppocampal-neocortical neuronal model for
evaluation of the SWS. It is well accepted fact reported by
research groups that humans who lack quality SWS (Delta wave of
0.1-4 Hz) most likely will suffer with many illnesses, such as type
1 diabetes mellitus, seizures (Epilepsy), Alzheimer’s, Parkinson’s,
brain injury and autism [13, 37-39]. Normal SWS waves have the
highest amplitude compared with all other brain waves, because
during the deep stage 3 or 4 sleeping, brain conducts consolidation
of declarative memory through hippocampus to neocortical networking
[13, 37-40]. We set up the frequency at 0.25 Hz with a current of
±10 μA and compared the signal strength between the “healthy active
ACHE gorge neuron” and the “mutated ACHE gorge neuron”. The
specimens used were NIST SRM965A human sera with blood glucose
level 2 (70 mg/dL) with a data rate of 1 KHz without spiking
ACH.
RESULTS AND DISCUSSIONS Characterization of the Memristors We
have come to believe that the brain is a special memristor or
memcapacitor type of device that is able to conduct complex
learning and memory through chemical synapse and electric synapse
of networking. The key structural difference between the two is the
synapse gap. For electric synapse is one tenth of that of chemical
synapse [41]. For Sensor 1, the 3D lattice between the flat bridge
and the top rim of the surface of the pores are gaps of 40-56 nm;
yet the Sensor 2 has gaps between 6-121 nm. Sensor 2 can become a
hybridized memristor with bridges having 115 nm apart in height,
and the Sensor 1 related to Sensor 2 having more characteristics of
electric synapse than Sensor 2 as shown in Fig 3A (Sensor 1) and 3B
(Sensor 2).
0.03 0.02 0.01 0.00 -0.01 -0.02 -0.03-4
-3
-2
-1
0
1
2
3
4Effect of frequncy on the Memristor 2 with vertical
bridged/nanopore of Au/biomimetic mutated ACHE in the presence of
10 μM ACH and 2 mM o-NPA
J(μA
/cm
2 )
AppE (V)
5 Hz 10 Hz 50 Hz 100 Hz 200 Hz 300 Hz 500 Hz
Fig 3A illustrates frequency affects on Memristor 1 with
typical bipolar nonlinear characteristics; Fig 3B shows
Memristor 2’s typical bipolar linear behavior.
The Memristor 1 is 4580-fold more sensitive to
ACH than Memristor 2 at 1 μM ACH and 1 cm2 sensor area,
indicating that a hydrophobic lining of the inner ACHE gorge is so
critically important for learning and memory.
Quantitation of ACH Following are preliminary data demonstrating
the performance characteristics of the ACH sensor 1 using a CA
method. Sensor 1 was able to measure ACH from 10-15 to 10-6 M shown
in Fig 4A and the current vs. concentration plot in Fig 4B was
presented. Km is 0.24 nM by Lineweaver-Burk plot and Vmax is 0.61
nM/s. The DOL is 9.7x10-16 /cm2.
0.00 0.02 0.04 0.06 0.08 0.10
0
1
2
3
4Au/"Healthy ACHE gorge" sensor 1detects ACH
control 10-15M ACH 10-12M ACH 10-10M ACH 10-8M ACH 10-6M ACH
J (m
A/cm
2 )
Time (s)10-15 10-14 10-13 10-12 10-11 10-10 10-9 10-8 10-7
10-6
0
6
12
18
24
30
36
Data: Data3_BModel: Exp1P1Equation: y = exp(x-A)Weighting: y No
weighting Chi^2/DoF = 119.8256R^2 = 5.6417E-7 A -2.1021 ±0.5461
ACH Concentration (M)
Cur
rent
(μA
)
Au/"Healthy ACHE gorge" sensor 1 with 0.031cm2 detectsACH in PBS
over fM to μM
Fig 4A illustrates CA curve profiles and 4B
shows the current vs. ACH concentration plot. The imprecision
was 8.2 and 8.3% using NIST 965A human sera with and w/o spiked ACH
with triplicate runs. Accuracy was 101% (LCI 85%; UCI 117%) of
spiked NIST sera against the calibration curve at 1.2x10-7M ACH.
Because the CA curves were sine waves, hence imprecision results
were ±0.4-0.6% by averaging of each of the two groups. Mutated ACHE
Gorge Needs a Hydrophobic Lining
It was observed that Sensor 2 lacking a hydrophobic lining in
the gorge, was unable to sense ACH. Fig 5A and 5B illustrate the CA
profiles and the calibration curve with a DOL of o-NPA 5.6
x10-13M/cm2 under 10 μM ACH.
0 20 40 60 80 100 120
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0Au/"Mutated ACHE gorge" with vertical bridge/nanopre
structure
sensor 2 detects o-NPA in 0.01 mM ACH in pH 7.4 PBS,Y = -0.002 +
0.016X with Sy/x = 0.01, r=0.9998
n=15
J (A
/cm
2 )
Concentration (μM) Fig 5A illustrates CA profiles of detection
of o-NPA.
Fig 5B is the calibration plot.
Assessing Quality of Slow-Wave Sleeping (SWS) The DSCPO method
was used for assessing quality of slow-wave sleeping of the newly
developed nanobiomimetic hyppocampal-neocortical neuronal models.
Fig 6 illustrates Sensor 1 has 5.5-fold higher peak intensity than
Senor 2 in the Delta wave at 0.25 Hz, that indicates Sensor works
perfectly at deep stage 4 sleep for memory consolidation; however
Sensor 2 has the lowest wave intensity indicating loss of
declarative memory and learning capability. It was noteworthy, that
the healthy SWS has biphases, and the unhealthy SWS only has one
phase with all peaks above zero
1.0 0.8 0.6 0.4 0.2 0.0 -0.2 -0.4 -0.6 -0.8 -1.0-100
-75
-50
-25
0
25
50
75
100Frequncy effect on peak current of an 0.031cm2Au/"Flat
bridge" nanopore ACH gorge model sensor detects1 pM ACH in PBS
Cur
rent
(μA)
APPE (V)
1 kHz
20Hz
Control ......
0.00 0.02 0.04 0.06 0.08
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.10
0.11
GCur
rent
(A)
Time (s)
A, B, CDE
F
AU/"vertical ATP bridge" naopore sensor responses to various
o-NPA in the presence of 0.1 mM ACHA: control, B: 10-10M;C:10-7M;D:
10-5M; E: 20x10-5M: F: 40x10-5MG: 10-4M
NSTI-Nanotech 2014, www.nsti.org, ISBN 978-1-4822-5827-1 Vol. 2,
2014 171
in which they are losing the membrane potential
reversibility
[42-43]. 0 20 40 60 80 100 120
-10
-5
0
5
10
15
A 0.031 cm2 Au Memristor/"Flat bridge-nanopore" "ACHE healthy
Gorge" membrane with
an insulator/Pt in NIST Human sera with level 2 glucose 70
mg/dL. Electrodes were connected at 2250 at 0.25 Hz with ± 10
μA.(A); Au memristor/"verital-bridge-nanopore" mutatedACHE gorge at
same experimental condition.(B)
Vot
age
(V)
Time (s)
A
B
Fig 6. Compares Sensor 1 and 2 in the SWS waves at 0.25
Hz with Sensor 1 (black), Sensor 2 (red).
CONCLUSION Detection of ACH with accuracy and sensitivity in fM
DOL under enzyme –free and reagent-free conditions was demonstrated
and the Km value agreed with literature using nature ACHE [44].
Using human blood samples shown a 5.5-fold higher intensity in
Delta SWS synapse by Sensor 1 compared with Sensor 2 indicate that
enhancing the ACHE gorge hydrophobicity is necessary when its
lining was damaged.
REFERENCES 1. Sussman JL, Harel M, Frolow F, Oefner C, Goldman
A,
Toker L, Silman I, Science 253, 872-879, 1991; 2. Gilson MK,
Straatsma TP, McCammon JA, Ripoll
DR, Faerman CH, Silman I, Sussman JL, Science 263, 1276-1278,
1994.
3 Silman I, Sussmana JL, Chem Biol Interact. 175, 3-10,
2008.
4. Whyte KA, Greenfield SA. Exp Neurol 184,496-509, 2003.
5. Swart C, Characterization of the interaction between
acetylcholinesterase and laminin: A template for discovering
redundancy, Dissertation, Stellenbosch University, 2012.
Stellenbosch University http://scholar.sun.ac.za
6. Zanini D, Schmatz R, Pelinson LP, Pimentel VC, Molecular and
cellular, xxx 2013.
7. Battisti V, Bagatini MD, Maders LDK et al., Biomedicine &
Pharmacotherapy, 66(4), 249-255, 2012.
8. Zhao YJ, Wang XY, Wang T et al., Hepatology, 53, 493-503,
2011.
9. Kenkichi T, Sakimoto Y, Kimura F & Mitsushima D,
Scientific Report, 4 (3738), 1-7, 2014.
10. http://www.biopsychiatry.com/alzheim.htm 11. Shen JN, Wang
DS, Wang R, Int J Clin Exp Pathol 5(9),
900-913, 2012. 12. Nordberg A, Rinne JO, Kadir A and Långström
B,
Nature, 6, 78-87, 2010. 13. Steffen G and Born J, PNAS, 101(7),
2140-2144, 2004. 14. Bucur MP, Bucur B and Radu GL,
15. Khan A, Khan AAP, Asiri AM, Rub MA et al., Appl Biochem
Biotechnol 169, 1927-1939, 2013.
16. Chen ET, Duh SH, Ngatchou C, Thornton JT, and Kissinger PT,
Nanotech (3),101-104, 2011.
17. Lonsdale R, Hoyle S, Grey DT, Ridder L and Mulholland AJ,
Biochemistry 51, 1774−1786, 2012.
18 Frushicheva MP, Cao J, Chu ZT, and Warshel A, PNAS 107(39),
16869–16874, 2010.
19. Liu S, Dai Z, Chen H and Ju H, Biosensors and
Bioelectronics, 19(9), 963, 2004.
20. Chen ET. US Patent 6,582.583, June 24, 2003. 21. Chen ET, US
Patent 8,083,926, December 27, 2011. 22. Chen ET, Thornton JT,
Ngatchou C, Duh SH and
Kissinger PT, NSTi-Nanotech (3), 115-118, 2013. 23. Chen ET,
Thornton JT, Ngatchou C, Duh SH and
Kissinger PT, NSTI- Nanotech (2), 107-110, 2013. 24. Goykhman I,
Korbakov N, Bartic C, Borghs G et al. , J.
AM. CHEM. SOC. 131, 4788–4794, 2009. 25. Perry M, Li Q, and
Kennedy RT, Anal Chim Acta.
653(1), 1–48, 2009. 26. Periasamy AP, Umasankar Y and Chen S-M,
Sensors 9,
4034-4055, 2009. 27. Micheau J and Marighetto A, Behav Brain
Res, 221(2),
424-429, 2011. 28. Gais S and Born J, PNAS, 101 (7), 2141-2144,
2004. 29. Ohno T, Hasegawa T, Tsuruoka T, Terabe K, Gimzewski
JK, Nature Materials, 10, 591-595, 2011 30. Martinez-Rincon J
and Pershin YV, Electron Devices,
IEEE Transactions 58 (6), 1809–1812, 2011. 31. Martinez-Rincon
J, Ventra MD, Pershin YV Physical
Review B, 81(19), 195430-1-195430-7, 2010. 32. Pickett MD,
Medeiros-Ribeiro G and Williams RS, A
Nature Materials, DOI: 10.1038/NMAT3510, 2012. 33. Kozma, Pino
RE, Pazienza GE, Advances in
neuomorphic memristor science and applications, Springer
publisher, 2012.
34. Ventra MD, Pershin YV, Nanotechnology 24, 255201, 2013.
35. Chen ET and Pardue HL, Anal Chem, 5, 2563-7, 1993. 36. Chen
ET, Thornton JT, Ngatchou C, Duh SH and
Kissinger PT, NSTi- NanoTech (2), 107-110, 2013. 37. Snogdal LS,
Folkestad L, Elsborg R et al., Diabetes
Research and Clinical Practice, page unknown, 2012. 38.
Ferini-Strambi L, Galbiati A and Marelli S, Frontier of
Neurology, 4(159), 1-3, 2013. 39. Noebels JL at al., Jasper's
basic mechanisms of the
epilepsies, 4th edition, National Center for Biotechnology
Information (US); 2012.
40. Marshall L, Binder S, Frontiers in human neuroscience,
7(614), 1-67, 2013.
41. Buzsáki G et al., Nature Reviews 13, 407-420, 2012. 42. Chen
ET, Ngatchou C, Sensors and Transducers J, 15,
special, 42-48, 2012. 43. Altamirano M, Coates CW, Grundfest H
and
Nachmansohn D, J. General Physiology, 91, 1953. 44. Wallace BG
and Gillon JW, The Journal of Neuroscience
2(8), 1108-1118, 1982.
NSTI-Nanotech 2014, www.nsti.org, ISBN 978-1-4822-5827-1 Vol. 2,
2014172