- Email: [email protected]
Enzymeless glucose sensor integrated with chronically implantable nerve cuff electrode for in-situ inflammation monitoring
Sensors and Actuators B 222 (2016) 425–432
Contents lists available at ScienceDirect
Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb
Enzymeless glucose sensor integrated with chronically implantable nerve cuff electrode for in-situ inflammation monitoring Yi Jae Lee a , Sung Jin Park a,b , Kwang-Seok Yun b , Ji Yoon Kang a , Soo Hyun Lee a,∗ a b
Center for BioMicrosystems, Korea Institute of Science and Technology, Seoul 136-791, Republic of Korea Department of Electronic Engineering, Sogang University, 1 Shinsoo-dong, Mapo-gu, Seoul 121-742, Republic of Korea
a r t i c l e
i n f o
Article history: Received 18 May 2015 Received in revised form 29 July 2015 Accepted 20 August 2015 Available online 21 August 2015 Keywords: Enzymeless Glucose sensor Nerve cuff electrode Black Pt Nafion Inflammation Chronic implantation
a b s t r a c t Using glucose concentration as an inflammation responsive element, we newly established an enzymeless glucose sensor integrated with a chronically implantable peripheral nerve cuff electrode for continuous and in-situ monitoring of local inflammation. The glucose sensor integrated with a nerve cuff electrode was fabricated on a polyimide substrate side-by-side, then the glucose sensor and nerve cuff electrode were reversely folded, and were located inside and outside, respectively. The experimental results reveal that the electroplated black Pt working electrode of the glucose sensor shows an enhancive surface roughness factor of 16.41 and had a good distribution on the flexible polyimide surface, which exhibits distinctly enhanced electro-catalytic activity compared to that obtained with plain Pt. Amperometry and electrochemical impedance spectroscopy indicated that the fabricated sensor had a sensitivity of 7.17 A/mM cm2 , an outstanding detection limit of 10 M, significant selectivity, and excellent recovery performance for enzymeless glucose detection. In order to evaluate the feasibility for inflammation monitoring in the immediate vicinity of the implantable peripheral nerve cuff electrode, the association of an evoked nerve signal recording and glucose concentration was investigated through ex-vivo test using the sciatic nerve of a SD rat. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Nerve cuff electrodes have been widely used for chronic nerve signal recording and stimulation [1–3]. However, much of the implanted device itself tends to induce a foreign body reaction and morphological changes such as the growth of surrounding connective tissue, nerve reshaping, epineurial fibrosis, and degeneration/regeneration of myelinated fibers [4]. In particular, fibrotic reaction is preceded by an important epineurial inflammation, which can alter the interface between the nerve and cuff electrode. Therefore, the development of inflammation suppression and ability to monitor in a local area is vital for chronically implanted devices. Generally, an inflammatory reaction is the body’s attempt to protect, maintain and try to remove something harmful or that is irritating in the body, specifically, acute inflammation reflects that the body’s attempt to heal itself. Sometimes, long-term inflammation (chronic inflammation) can be a contributing factor to many serious and chronic diseases such as cancer, heart disease, diabetes,
∗ Corresponding author. Tel.: +82 2 958 6755; fax: +82 2 958 6910. E-mail addresses: [email protected], [email protected] (S.H. Lee). http://dx.doi.org/10.1016/j.snb.2015.08.091 0925-4005/© 2015 Elsevier B.V. All rights reserved.
autoimmune diseases, and Alzheimer’s disease [5]. Generally, when assessing the extent or activity of inflammation, serum C-reactive protein (CRP) level provides a rough guide to the amount of tissue involved in inflammation and to the integrity of the inflammatory response. The CRP method used in the laboratory is considered a more direct measure of the inflammatory process, while erythrocyte sedimentation rate (ESR) is a more indirect measure. ESR reflects the concentration of several plasma proteins including fibrinogen, ␣-globulins, -globulins, immunoglobulins and albumin. Therefore, any condition (pathological or non-pathological) that affects any of the contributing proteins can alter the ESR. Recently, several studies regarding the association of inflammation and glucose disorders have been reported [6,7]. The possibility that the inflammation affects the concentration of glucose in the immediate vicinity of a glucose sensor has also been suggested [8]. Specifically, a previous study reported increased glucose consumption that was double in wounded rat tissue compared to that of normal tissue [9]. Additionally, the degradation in the sensitivity of implanted devices may be caused by changes in the surrounding tissue due to an inflammatory response or from the formation of a fibrotic capsule that surrounds the implanted device. Therefore, monitoring inflammation in the immediate vicinity of the implanted device is essential for the stable operation of
426
Y.J. Lee et al. / Sensors and Actuators B 222 (2016) 425–432
chronically implanted devices, e.g., a nerve cuff electrode. Although there are many types of glucose sensors, such as a micro-needle, planar, and patch [10,11], most have only focused on monitoring body status through the glucose level in blood or interstitial fluid. The enzymeless glucose sensors are largely classified as potentiometric sensors, e.g., polymer membrane including boronic acid derivatives coated sensor, voltammetric sensors, e.g., sensor using ferrocene unit for voltammetric read-out-unit, and amperometric sensors using applying programmed potential pulsed or to monitor the current at a fixed potential [12]. More specifically in amperometric methods, enzymeless glucose sensors have been classified that use fixed potential amperometric analysis [13], medium fixed potential amperometry coupled with of flow injection analysis [14] and liquid chromatography, and variable potential amperometric methods such as potential sweep and pulsed voltammetry [15]. Amperometric analysis based on high catalytic nano-materials is one of the most widely used methods due to enhanced amperometric detection avoiding electroactive interference [16]. To our knowledge, this is the first study that directly applies a robust glucose sensor using black Pt integrated with an implantable nerve cuff electrode for inflammation monitoring of a peripheral nerve. The enzymeless glucose sensor with black Pt is operated by direct electro-oxidation of glucose via fixed potential amperometric analysis, which is considerably depending on the electrode material used. Pt is one of the most studied noble metals in the field of sensors and catalysts. However, conventional Pt has drawbacks of low sensitivity, selectivity, often suffer from losing activity due to easily poisoned by adsorbed intermediates. While noble metal nanoparticles are the most widely used nanomaterials in electrochemical sensor configuration due to their enhanced electro-catalytic properties. Particularly, electroplated black Pt is an enhancive material for sensor sensitivity due to its improved oxidation rate by enlarged real surface area and has excellent biocompatibility [17,18]. The electro-catalytic oxidation of glucose on Pt nanoparticles is a kinetic controlled reaction, whereas the responses from the common interfering species such as ascorbic acid and uric acid will not be changed significantly since their oxidation are diffusion controlled [19]. In this study, an enzymeless glucose sensor integrated with a chronically implantable peripheral nerve cuff electrode was designed, fabricated and characterized for in-situ inflammation monitoring. For sensitive and selective detection of glucose that was stable and unaffected by the denaturing of an immobilized enzyme, black Pt was selectively electroplated onto a plain Pt working electrode of an enzymeless glucose sensor. Subsequently, it was deep-coated with Nafion to avoid anionic interfering species. Glucose sensitivity was compared with and without a Nafion coating. The electrochemical properties of a fabricated enzymeless glucose sensor integrated with a nerve cuff electrode were investigated by cyclic voltammetry, electrochemical impedance spectroscopy, and chronoamperometry. In addition, the fabricated nerve cuff electrode integrated with a glucose sensor was tested to confirm whether the nerve cuff electrode could receive a stable nerve signal transfer at various glucose concentrations using a rat sciatic nerve through ex-vivo test.
2. Experimental 2.1. Reagents and apparatus HCPA (hexachloroplatinic acid hydrate, 99.9% purity, Aldrich), hydrochloric acid, and lead acetate (reagent grade, 95%) were prepared to make the Pt black electrode. The fabricated electrode was measured to check its surface roughness factor (RF) in a 1 M sulfuric acid (H2 SO4 , 95–98%, ACS, Sigma-Aldrich) solution using cyclic
voltammetry. The 1 M H2 SO4 was prepared by dilution in deionized water. The -d(+) glucose (99.5%, Sigma) stock solution was prepared by diluting it in a 0.1 M PBS solution and allowing it to stand for 24 h before use, in order to create equilibration. The 0.1 M urea (98%, Sigma), uric acid (UA, 99%, Sigma), sucrose (99.5%, Sigma), and ascorbic acid (AA, 98%, Sigma) solution were prepared by diluting them in a 0.1 M phosphate buffered saline (PBS, pH 7.0) solution. Nafion (5 wt.% solution) as a coating material for selectivity was obtained from Sigma Aldrich, Korea. The electrochemical experiments on the fabricated electrodes were performed by using Autolab (PGSTAT 302N, NOVA software, Ecochemie, Utrecht, The Netherlands) employing a three-electrode configuration with a fabricated black Pt for the working electrode (WE), a sputtered Pt as a counter electrode (CE), and an Ag/AgCl reference electrode (RE) at room temperature. The cyclic voltammograms of the fabricated sensor was recorded from −0.5 to +1.0 V in a range with a 50 mV/s scan rate. Chronoamperometry was performed in a 0.1 M PBS (10 mL volume) solution under continuous stirring to dilute the glucose with a constant concentration (0.5 M). The response current was recorded after stabilizing the background current at a nearly constant value and the prepared glucose were consecutively injected at a regular interval of 100 s. Surface morphology was characterized using a Hitachi S-2300 scanning electron microscope (SEM). 2.2. Fabrication of an enzymeless glucose sensor integrated with a nerve cuff electrode Fig. 1 provides conceptual drawings of an enzymeless glucose sensor integrated with a nerve cuff electrode for inflammation monitoring in a peripheral nerve. As shown in Fig. 1(a), the nerve cuff electrode consisted of four Pt line electrodes (0.25 mm × 1 mm) and the glucose sensor consisted of black Pt as a WE (geometry area: 4 mm2 ), sputtered Pt as a CE, and Ag/AgCl as a RE. The fabricated glucose sensor integrated with a nerve cuff electrode was reversely folded to the nerve cuff electrode (Fig. 1b). Fig. 1(c and d) shows a wrapped glucose sensor (outside) integrated with a nerve cuff electrode (inside) on a peripheral nerve and local inflammation monitoring of glucose sensor via fluctuation of glucose concentration by the inflammatory activation. The fabrication process for the enzymeless glucose sensor integrated with a nerve cuff electrode is described in Supplementary data Fig. S1 and in our previous research [20]. The first polyimide (VTEC 1388, Richard Blaine International, Inc., Philadelphia, PA, USA) layer of a 20 m thickness was formed on a 4-in. silicon wafer by spin-coating (Supplementary data Fig. S1a). After curing the wafer at 90 ◦ C for 10 min, 110 ◦ C for 10 min, and 220 ◦ C for 60 min in a convection oven, sputtered Ti/Pt (50/300 nm) layers on top of the 1st polyimide layer were formed by using a lift-off process (Supplementary data Fig. S1b). Next, the 2nd polyimide (5 m thickness) layer for insulation was formed by spin-coating. In order to selectively open the electrode site and the connector pads, a positive photoresist (AZ 9260, AZ Electronic Materials, NJ, USA) was patterned on the 2nd polyimide layer and the exposed polyimide patterns were etched by reactive ion etching (RIE) (Plasma Therm, St. Petersburg, FL, USA) (Supplementary data Fig. S1c). A laser dicing machine (M-2000, Exitech, Oxford, UK) was used to cut the perimeter of the electrode substrate (Supplementary data Fig. S1d). To achieve a cuff shape with a 1 mm diameter capable of wrapping around the sciatic nerve of a rat, the flat nerve electrode and glucose sensor electrode were reversely folded and reformed using a metal rod, and then cured in a convection oven at 220 ◦ C for 2 h to permanently fix the shape of the cuff (Supplementary data Fig. S1e). Then the black Pt was selectively electroplated on the sputtered Pt WE using a potentiostatic mode (−200 mV, 100 s).
Y.J. Lee et al. / Sensors and Actuators B 222 (2016) 425–432
427
Fig. 1. (a) Conceptual drawings of the enzymeless glucose sensor integrated with a nerve electrode, (b) the reversely folded glucose sensor and nerve electrode, (c) the wrapped glucose sensor integrated with nerve cuff electrode at peripheral nerve, and (d) the local inflammation monitoring of glucose sensor via fluctuation of glucose concentration by the inflammatory activation.
Fig. 2 shows the fabricated enzymeless glucose sensor integrated with a nerve electrode and SEM images of the glucose sensor. As shown in Fig. 2(a), the fabricated nerve electrode and glucose sensor were located equally, then each of the parts were reversely folded and cured into a cuff shape (inside: nerve electrode, outside: glucose sensor). The selectively electroplated black Pt on top of sputtered Pt WE was relatively well distributed with a similar grain size as shown in Fig. 2(b). 2.3. Ex-vivo nerve signal recordings in the sciatic nerve Ex-vivo experiments were performed using a three-month-old Sprague-Dawley (SD) rat. The animal was kept and handled in accordance with the regulations of the Institutional Animal Care and Use Committee (IACUC) of the Korea Institute of Science and Technology (KIST). The SD rat was kept in a temperature-controlled room at 24 ◦ C, and was fed food and water in an ad libitum feeder. The experimental protocol used for this study was approved by the Ethical Committee (KIST IACUC approval #2014-019) for Animal Experiments. During surgical extraction, the SD rat was anesthetized with an anesthetic mixed with Zoletil 50 (Virbac, Carros, France) and Rompun (Bayer, Seoul, Korea).
The sciatic nerve in the right leg of the SD rat was dissected. Then, the dissected sciatic nerve was placed in a PBS solution. The stimulation cuff electrode and recording cuff electrodes integrated with a glucose sensor were wrapped around the extracted sciatic nerve, respectively. The gap between the stimulating and recording cuff electrodes was 5 mm when wrapped around the extracted sciatic nerve. The ground electrode was grounded in another PBS solution. The stimulating cuff electrode was connected to a pulse stimulator (Isolated Pulse Stimulator, Model 2100, A-M Systems, Sequim, WA, USA). The recording cuff electrode integrated with a glucose sensor was partially connected to a differential amplifier (Differential AC amplifier, Model 1700, A-M systems, Sequim, WA, USA) and AutoLab. The amplified neural signals of the sciatic nerve were collected in a data acquisition (DAQ) device (NI USB-6356, National Instruments, Seoul, Korea). Then, the collected signals were processed using LabVIEW software (National Instruments, Seoul, Korea) and displayed on a laptop computer. The electrical stimulation was applied with pulse amplitude of 100, 200, and 300 A and a pulse width of 100 s. The differential amplifier gain was a factor of 1000.
Fig. 2. Photographs of the fabricated enzymeless glucose sensor integrated with nerve cuff electrode (a) and the SEM images of the glucose sensor (gap size of among WE, RE, and CE: 10 m).
428
Y.J. Lee et al. / Sensors and Actuators B 222 (2016) 425–432
the consecutive increment of 2 mM glucose, changes of the oxidation current were observed and the voltammetric response also increased along with increments of glucose concentration. The glucose oxidation peak currents might be attributed to typical Pt oxidation as described in the following reactions: Pt + 2OH− − 2e− → Pt(OH)2 Pt(OH)2 + 2OH− → PtO(OH)2 + H2 O + 2e− PtO(OH)2 + glucose → Pt(OH)2 + gluconolactone Especially, the oxidation current around 0.4 V increased significantly with an increase of glucose concentration, indicating the response of the sensor to glucose. Thus, 0.4 V was selected for an amperometric test to minimize the oxidation of interfering species. Electrochemical impedance spectroscopy and voltammetric responses of the fabricated enzymeless glucose sensor with plain Pt and black Pt WE are exhibited in Supplementary data Fig. S2(a and b). The measured impedance of the glucose sensor with black Pt WE was extremely lowered due to an enlarged surface area (plain Pt: 600 , black Pt: 40 at 1 kHz). Unlike the black Pt WE, the distinguished response peak current as increment of glucose concentration was not observed in cyclic voltammograms of plain Pt WE (Supplementary data Fig. S2b). 3.2. Amperometric behavior of the glucose sensor integrated with a nerve cuff electrode
Fig. 3. Cyclic voltammograms of the plain Pt WE and black Pt WE in 1 M sulfuric acid solution (a) and for the black Pt WE to various glucose concentrations (no glucose, 2, 4, 6, 8, 10 mM) in 0.1 M PBS (b). Scan rate: 50 mV/s.
3. Results and discussions 3.1. Voltammetric behavior of the glucose sensor integrated with a nerve cuff electrode Fig. 3(a) shows a cyclic voltammogram of the fabricated sensor with black Pt and plain Pt WE in 1 M sulfuric acid solution to compare the roughness factor at a hydrogen adsorption region. The dual peaks area (0 ∼ −0.45 V) of the black Pt WE in the cathodic scan indicates a reduction charge of hydrogen at the electrode surface. Since the charge was necessary to form a monolayer of adsorbed hydrogen and the WE area was covered by each hydrogen atom, the electrochemical roughness factor (RF) could be calculated by dividing the adsorption charge by 0.21 mC/cm2 [21]. The accumulated charge density in a region of Fig. 3(a) of the plain Pt WE and black Pt WE were 0.21 and 3.45 mC/cm2 , respectively, which were divided by the aforementioned charge density of the adsorbed hydrogen monolayer (0.21 mC/cm2 ). The corresponding RF of the plain Pt and black Pt WE of the sensor were 1 and 16.41(n = 3), respectively. The glucose sensor with black Pt WE showed an enhanced current response due to an extremely enlarged surface activation area. Fig. 3(b) shows cyclic voltammograms of the black Pt WE to various glucose concentrations in 0.1 M PBS. The voltammetric behaviors of the glucose sensor with black Pt WE was obtained at a scan rate of 50 mV/s and the potential ranged from −0.5 to 1 V vs. Ag/AgCl in a 0.1 M PBS solution without and with the addition of glucose (2, 4, 6, 8, and 10 mM). As shown in Fig. 3(b), upon
In order to achieve practical use, anti-fouling, and avoid reactions to anionic interfering species surrounding a peripheral nerve [22], the fabricated glucose sensor with black Pt WE was coated with Nafion. Glucose sensitivity with and without Nafion coating was compared. Nafion is the most widely used biocompatible material due to its simple handling and commercial availability [23]. The amperometric current response of the glucose sensor with black Pt WE and a Nafion coated black Pt WE are shown in Fig. 4(a and b). Consecutive addition of different concentrations of glucose (0.01, 0.025, 0.05, 0.1, 0.5, 1, and 2 mM) while stirring PBS at an applied potential of 0.4 V after stabilizing the background current at a nearly constant value with 100 s intervals. The current response as increment of glucose concentration was proportionally increased due to covered black Pt nanoparticles, which indicates that the Pt nanoparticles provide an high effective area and the glucose are electrolyzed as if in a thin layer cell. The shorten diffusion distance to electrode by electroplated Pt nanoparticles might be attributed to highly enhanced current response as well as change of diffusion regime [24]. Furthermore, the electroplated black Pt nanoparticles as porous-like surface in the SEM image of Fig. 2(b) might be lead to enhanced glucose oxidation current. Unlike semiinfinite diffusion of plain Pt electrode, the formed porous-like surface of the black Pt nanoparticles might be slightly affected on selectivity of the enzymeless glucose sensor (see Fig. 5a). The linear regression curve of the sensor with black Pt WE and Nafion coated black Pt WE in Fig. 4(c) is linear over a glucose concentration up to 22 mM (R2 = 0.93835 and 0.97936) and had a high sensitivity of 7.17 (286. 8) and 2.57 (103.1) A/mM cm2 (nA/mM), respectively. The sensitivity of the glucose sensor with plain Pt was only 0.02 A/mM cm2 (Supplementary data Fig. S2c). The detection limit of the sensor with black Pt and Nafion coated black Pt WE was estimated to be about 10 and 50 M, respectively. The detection limit was calculated from signal-to-noise (S/N = 3) ratio method. The estimated sensitivity of the fabricated glucose sensor is comparable with those of glucose sensors tested through a similar approach as previous reported 465.9 ± 48 nA/mM (Gr Ox /Pt-black) [25], 0.531 ± 0.149 nA/mM (Pt black/MWNT/Nafion/GOx ) [26]. The
Y.J. Lee et al. / Sensors and Actuators B 222 (2016) 425–432
429
Fig. 5. Amperometric response of 1 mM glucose, 0.1 mM urea, 0.1 mM UA, 0.1 mM sucrose, and 0.1 mM AA at the sensor with black Pt WE and Nafion coated black Pt WE (a) and the amperometric response recordings for recovery characteristic of glucose sensor with black Pt WE and Nafion coated black Pt WE in step change of glucose concentration (b).
Fig. 4. Comparison of amperometric response for the sensor with black Pt WE (a) and Nafion coated black Pt WE (b) to consecutive addition of glucose (0.01, 0.025, 0.05, 0.1, 0.5, 1, 2 mM) with interval of 100 s and linear regression for current to glucose concentration (error bars represent standard error of the mean) (c). The applied potential: 0.4 V.
comparison for the properties of enzymatic/enzymeless glucose sensors and this work was shown in Supplementary data Table S1. This result could be attributed to the large surface-to-volume ratio, high conductivity and facilitation of electron transfer by the black Pt basement, indicating it is a good platform for construction of an enzymeless sensor. The maximum relative standard deviation (RSD) of the current responses to 20 mM glucose is 12.3 (black Pt) and 14.4% (Nafion coated black Pt) for three successive
measurements, respectively (Fig. 4c). From the adsorbed glucose electro-oxidation on black Pt electrode at 0.4 V, gluconolactone is obtained as byproduct, which hydrolysis to gluconic acid on standing [16]. The response current of the glucose sensor with black Pt WE as increment of glucose concentration was slightly affected by generated oxidation products and intermediates during glucose oxidation. On the contrary, the glucose sensor with Nafion coated black Pt WE might be effectively avoided from the competitive adsorbed other anions such as phosphate anions, chloride anions. Several typical interfering species including urea, UA, sucrose, and AA were chosen to test the selectivity of the enzymeless glucose sensor. Although the glucose sensor with black Pt WE was slightly affected by the AA, the sensor with Nafion coated black Pt WE effectively avoided interference from AA (Fig. 5a). However, the consideration for non-glucose sugars (mono and polysaccharides) such as maltose, icodextrin, xylose, and galactose on fabricated glucose sensor is additionally required due to their severe interference with glucose analysis, e.g., overestimation of glucose level. The securing of glucose selectivity to non-glucose sugars is common problem for solution on enzymatic sensors in spite of their high selectivity as well as enzymeless sensors that goes without saying [16]. Recently, glucose dehydrogenase (GDH)-nicotinamide adenine dinucleotide (NAD) [27] or GDH-flavin adenine dinucleotide (FAD) based sensor [28], and enzymeless sensor with three
430
Y.J. Lee et al. / Sensors and Actuators B 222 (2016) 425–432
Fig. 6. Experimental setup of stimulus cuff electrode and recording cuff electrode integrated with enzymeless glucose sensor with dissected rat’s sciatic nerve for acute ex-vivo sciatic nerve signal recording (a), acute ex-vivo recordings to the biphasic electrical stimulation (100, 200, and 300 A) (b), acute ex-vivo recordings to the various glucose concentrations (from no glucose to 20 mM) at 300 A electrical stimulation (c), and plot for concentration of glucose versus evoked nerve compound action potential peak to peak (d). Input pulse duration was 100 s.
dimensional network of ZnO–CuO nanocomposites [29] were reported as platforms for free of these interferences. In order to prove that the enzymeless glucose sensor with black Pt WE and Nafion coated black Pt WE are a good platform for the construction of in-situ glucose monitoring of inflammation in tissues, an amperometric current to step-change glucose concentration (no glucose, 2, 5, 10 mM) was used to study the recovery characteristics of the as-prepared sensor with black Pt WE and Nafion coated black Pt WE as shown in Fig. 5(b). All measurements were performed by a stepwise change of glucose concentration at 100 sec regular intervals. The current responses of black Pt WE and Nafion coated black Pt WE were very stable toward both increments and decrements of glucose concentration. The recovery rates of the black Pt WE and Nafion coated black Pt WE for no glucose, 2, and 5 mM were 127.5, 91.6, 85.2% (black Pt) and 108.1, 111.9, 104.2% (Nafion coated black Pt), respectively. Recovery rate was defined by the following equation: recouvery rate =
current response by 2nd glucose concentration × 100 current response by 1st glucose concentration (1)
3.3. Ex-vivo test As shown in Fig. 6(a), a fabricated enzymeless glucose sensor integrated with a nerve cuff electrode was employed to record the
compound action potential of a rat’s sciatic nerve in increments of electrical stimulation (100, 200, and 300 A) and glucose concentration up to 20 mM. The dissected sciatic nerve was wrapped by a stimulating cuff electrode and a recording cuff electrode integrated with a glucose sensor to record compound action potential. The stimulating cuff electrode (5 mm width) had the same electrode configuration (4 line electrodes) as the recording cuff electrode (12 mm width). When the sciatic nerve was excited by the stimulating cuff electrode, a compound action potential through the nerve was transferred and recorded by the recording cuff electrode integrated with a glucose sensor. Fig. 6(b) shows the typical responses of the sciatic nerve recorded from the cuff electrode integrated with a glucose sensor at a 6 mM glucose concentration. The compound action potential of the sciatic nerve was simultaneously generated after the electrical stimulation given by the stimulation electrode. The insets of Fig. 6(b) represent a zoomed-in view of the evoked nerve signal and applied biphasic pulse current, respectively. In Fig. 6(b), the stimulus artifact signal and evoked nerve signal in the sciatic nerve were clearly and distinctly recorded as the stimulation intensity was increased from 100 to 300 A. The evoked nerve signal amplitudes to 100, 200, and 300 A peak current amplitudes were 0.608, 0.982, and 1.248 mV peak to peak, respectively. This demonstrated that the cuff electrode integrated with a glucose sensor was capable of functioning as a simultaneous nerve signal recording electrode. According to the results in Fig. 6(c and d), the transferred compound action potential can be simultaneously recorded
Y.J. Lee et al. / Sensors and Actuators B 222 (2016) 425–432
by the nerve cuff electrode integrated with a glucose sensor at various glucose concentrations. The experimental range for the glucose concentration was defined by normal ranges (Normal range: 4.2–5.3 mM, approximate short-term nonlethal limit: 1.1–83.3 mM) of important extracellular fluid constituents [30]. An electrical biphasic stimulation current was applied at 300 A with a 100 s duration. Since the evoked nerve signal was affected by the glucose concentration, and nerve injury is affected by hypoglycemia or hyperglycemia [31], the amplitude of the compound action potential was sharply increased up to a glucose concentration of 6 mM, then the potential was slightly decreased in a glucose concentration range from 8 to 18 mM. Finally, the evoked potential was sharply decreased at a high glucose concentration of 20 mM (Fig. 6d). This result indicates that the evoked nerve signal might be affected by the glucose concentration in the immediate vicinity of the cuff electrode. Therefore, local glucose concentration fluctuation in the immediate vicinity of the cuff electrode can be sufficiently considered as an indirect inflammation marker, because glucose capacity to use glucose as a fuel is particularly correlated to the inflammatory status of macrophages [32]. Although the activity of macrophages in adipose tissue is not yet clear, there is considerable evidence that they participate in the onset of insulin resistance and the development of diabetes associated with obesity via the production of an inflammatory reaction [33]. When onset of inflammatory activation in the local area of an implanted nerve cuff electrode occurs, an integrated glucose sensor in the local area can be used to directly monitor any remarkably decreased glucose level due to macrophage activation.
4. Conclusion This study introduced an enzymeless glucose sensor integrated with a nerve cuff electrode. The sensor was newly designed, fabricated, and characterized for in-situ and local inflammation monitoring in the immediate vicinity of a chronically implanted nerve cuff electrode. The fabricated glucose sensor with black Pt as a working electrode had a relatively large surface RF value when compared with plain Pt, which means black Pt has a larger electrode activation area over the other type. Furthermore, the sensor with black Pt had high sensitivity and a low detection limit in various glucose concentrations without any enzyme and mediators. The glucose sensor with a Nafion coated black Pt electrode exhibited excellent interference removal characteristic for selective glucose detection and recovery properties for continuous glucose concentration fluctuations. In addition, the fabricated nerve cuff electrode integrated with a glucose sensor was investigated for feasibility and nerve signal transduction in terms of variance in glucose concentration via ex-vivo test using a SD rat’s sciatic nerve. Since drastic local fluctuations of glucose concentration can be considered as triggering pro-inflammatory activation, sensitive and stable glucose monitoring in the immediate vicinity of the nerve cuff electrode is imperative for chronic stability in nerve signal recording. In the near future, the multilateral relevance of glucose concentration for in vivo inflammatory status will be investigated. It is likely that an understanding of glucose metabolism via local implantable devices will aid in fine-tuning treatment for proinflammation or chronic inflammation.
Acknowledgements This research was partially supported by the Happy Tech Program through the National Research Foundation of Korea (NRF) (grant number: 2009-0091918, 2010-0020786), which was funded by the Ministry of Education, Science and Technology (grant
431
number: 2009-0091918, 2010-0020786) and the KIST Institutional Program (2E25474).
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.snb.2015.08.091.
References [1] J. Badia, T. Boretius, D. Andreu, C. Azevedo-Coste, T. Stieglitz, X. Navarro, Comparative analysis of transverse intrafascicular multichannel, longitudinal intrafascicular and multipolar cuff electrodes for the selective stimulation of nerve fascicles, J. Neural Eng. 8 (Jun) (2011) 036023. [2] E.L. Foldes, D.M. Ackermann, N. Bhadra, K.L. Kilgore, N. Bhadra, Design, fabrication and evaluation of a conforming circumpolar peripheral nerve cuff electrode for acute experimental use, J. Neurosci. Methods 196 (Mar) (2011) 31–37. [3] N.S. Korivi, P.K. Ajmera, Self-closing cuff electrode for functional neural stimulation and recording, J. Med. Biol. Eng. 31 (2011) 353–357. [4] V. Vince, M.E. Brelen, J. Delbeke, I.M. Colin, Anti-TNF-alpha reduces the inflammatory reaction associated with cuff electrode implantation around the sciatic nerve, J. Neuroimmunol. 165 (Aug) (2005) 121–128. [5] M.M. Bosma-den Boer, M.L. van Wetten, L. Pruimboom, Chronic inflammatory diseases are stimulated by current lifestyle: how diet, stress levels and medication prevent our body from recovering, Nutr. Metab. 9 (Apr) (2012) 32. [6] J.I. Barzilay, L. Abraham, S.R. Heckbert, M. Cushman, L.H. Kuller, H.E. Resnick, et al., The relation of markers of inflammation to the development of glucose disorders in the elderly—the cardiovascular health study, Diabetes 50 (Oct) (2001) 2384–2389. [7] Y. Pan, Y. Wang, L. Cai, Y. Cai, J. Hu, C. Yu, et al., Inhibition of high glucose-induced inflammatory response and macrophage infiltration by a novel curcumin derivative prevents renal injury in diabetic rats, Br. J. Pharmacol. 166 (Jun) (2012) 1169–1182. [8] K. Rebrin, U. Fischer, H. Hahn von Dorsche, T. von Woetke, P. Abel, E. Brunstein, Subcutaneous glucose monitoring by means of electrochemical sensors: fiction or reality? J. Biomed. Eng. 14 (Jan) (1992) 33–40. [9] J. Forster, A.S. Morris, J.D. Shearer, B. Mastrofrancesco, K.C. Inman, R.G. Lawler, et al., Glucose uptake and flux through phosphofructokinase in wounded rat skeletal muscle, Am. J. Physiol. 256 (Jun) (1989) E788–E797. [10] S. Ichimori, K. Nishida, S. Shimoda, T. Sekigami, Y. Matsuo, K. Ichinose, et al., Development of a highly responsive needle-type glucose sensor using polyimide for a wearable artificial endocrine pancreas, J. Artif. Organs 9 (2006) 105–113. [11] M. Yasuzawa, S. Sato, H. Nakanishi, K. Edagawa, Fabrication of minimally-invasive patch type glucose sensors, Chem. Sens. 10 Mems/Nems 10 50 (2012) 83–87. [12] S. Park, H. Boo, T.D. Chung, Electrochemical non-enzymatic glucose sensors, Anal. Chim. Acta 556 (2006) 46–57. [13] T. Alizadeh, S. Mirzagholipur, A Nafion-free non-enzymatic amperometric glucose sensor based on copper oxide nanoparticles–graphene nanocomposite, Sens. Actuators, B: Chem. 198 (2014) 438–447. [14] C.-L. Sun, W.-L. Cheng, T.-K. Hsu, C.-W. Chang, J.-L. Chang, J.-M. Zen, Ultrasensitive and highly stable nonenzymatic glucose sensor by a CuO/graphene-modified screen-printed carbon electrode integrated with flow-injection analysis, Electrochem. Commun. 30 (2013) 91–94. [15] L. Rassaei, F. Marken, Pulse-voltammetric glucose detection at gold junction electrodes, Anal. Chem. 82 (2010) 7063–7067 (2010). [16] K.E. Toghill, R.G. Compton, Electrochemical non-enzymatic glucose sensors: a perspective and an evaluation, Int. J. Electrochem. Sci. 5 (Sep) (2010) 1246–1301. [17] Y. Li, C. Sella, F. Lemaitre, M.G. Collignon, L. Thouin, C. Amatore, Highly sensitive platinum-black coated platinum electrodes for electrochemical detection of hydrogen peroxide and nitrite in microchannel, Electroanalysis 25 (Apr) (2013) 895–902. [18] S.Q. Gu, Y.L. Lu, Y.P. Ding, L. Li, H.S. Song, J.H. Wang, et al., A droplet-based microfluidic electrochemical sensor using platinum-black microelectrode and its application in high sensitive glucose sensing, Biosens. Bioelectron. 55 (May) (2014) 106–112. [19] L.-Q. Rong, C. Yang, Q.-Y. Qian, X.-H. Xia, Study of the nonenzymatic glucose sensor based on highly dispersed Pt nanoparticles supported on carbon nanotubes, Talanta 72 (2007) 819–824. [20] S.H. Lee, J.H. Jung, Y.M. Chae, J.K.F. Suh, J.Y. Kang, Fabrication and characterization of implantable and flexible nerve cuff electrodes with Pt, Ir and IrOx films deposited by RF sputtering, J. Micromech. Microeng. 20 (Mar) (2010) 035015. [21] T. Biegler, D.A.J. Rand, R. Woods, Limiting oxygen coverage on platinized platinum—relevance to determination of real platinum area by hydrogen adsorption, J. Electroanal. Chem. 29 (1971) 269–270. [22] B. Akle, W. Habchi, R. Abdelnour, J. Blottman Iii, D. Leo, Biologically inspired highly efficient buoyancy engine, Proc. SPIE 8339 (Apr) (2012) 833900.
432
Y.J. Lee et al. / Sensors and Actuators B 222 (2016) 425–432
[23] H.B. Noh, M.S. Won, Y.B. Shim, Selective nonenzymatic bilirubin detection in blood samples using a Nafion/Mn–Cu sensor, Biosens. Bioelectron. 61 (Nov) (2014) 554–561. [24] B.R. Kozub, N.V. Rees, R.G. Compton, Electrochemical determination of nitrite at a bare glassy carbon electrode; why chemically modify electrodes? Sens. Actuators, B: Chem. 143 (2010) 539–546. [25] J. Shi, H.Y. Zhang, A. Snyder, M.X. Wang, J. Xie, D.M. Porterfield, et al., An aqueous media based approach for the preparation of a biosensor platform composed of graphene oxide and Pt-black, Biosens. Bioelectron. 38 (Oct–Dec) (2012) 314–320. [26] E.S. McLamore, J. Shi, D. Jaroch, J.C. Claussen, A. Uchida, Y. Jiang, et al., A self referencing platinum nanoparticle decorated enzyme-based microbiosensor for real time measurement of physiological glucose transport, Biosens. Bioelectron. 26 (Jan) (2011) 2237–2245. [27] D.M. Kim, M.Y. Kim, S.S. Reddy, J. Cho, C.H. Cho, S. Jung, et al., Electron-transfer mediator for a NAD-glucose dehydrogenase-based glucose sensor, Anal. Chem. 85 (Dec) (2013) 11643–11649. ˇ ´ K. Luˇspai, P. Magdolen, E. Sturdík, [28] R. Monoˇsík, M. Stred’ansky, Amperometric glucose biosensor utilizing FAD-dependent glucose dehydrogenase immobilized on nanocomposite electrode, Enzyme Microb. Technol. 50 (2012) 227–232. [29] C. Zhou, L. Xu, J. Song, R. Xing, S. Xu, D. Liu, et al., Ultrasensitive non-enzymatic glucose sensor based on three-dimensional network of ZnO–CuO hierarchical nanocomposites by electrospinning, Sci. Rep. 4 (2014) 7382. [30] M. V. T. D. S. R. Anura Kurpad and E. India, Guyton & Hall: Textbook of Medical Physiology—A South Asian Edition, Elsevier India, 2013. [31] M.M. Stecker, M. Stevenson, Effect of glucose concentration on peripheral nerve and its response to anoxia, Muscle Nerve 49 (Mar) (2014) 370– 377. [32] A. Johnson, A. Freemerman, E. Abel, J. Rathmell, L. Makowski, Glucose metabolism is linked to the inflammatory status of macrophages, BMC Proc. 6 (2012) P62. [33] A. Miranville, A.W. Herling, G. Biemer-Daub, M.D. Voss, Reversal of inflammation-induced impairment of glucose uptake in adipocytes by direct effect of CB1 antagonism on adipose tissue macrophages, Obesity 18 (Dec) (2010) 2247–2254.
Biographies Yi Jae Lee received the B.S.E.E., M.S.E.E., and Ph.D. degrees from Kwangwoon University, Seoul, Korea in 2007, 2009 and 2013, respectively. He is a postdoctoral researcher at Korea Institute of Science and Technology (KIST), Seoul, Korea. His current research interests include biosensors, bioelectronics, neural electrodes, glucose fuel cells for implantable system applications. Sung Jin Park will obtain his Master’s degree in 2015 February and issued by Sogang University of Republic of Korea. Mr. Park is studying the functional nerve cuff electrode. Kwang-Seok Yun received his B.S. degree in Electronics Engineering from Kyungpook National University in 1996, M.S. and Ph.D. degrees in Electronics Engineering from Korea Advanced Institute of Science and Technology (KAIST) in 1997 and 2002, respectively. He was a postdoctoral researcher at the University of California, Los Angeles, from 2005 to 2007. He joined the Department of electronic Engineering at Sogang University, Korea, in 2007, where he is now an associate professor. His current research area includes micro total analysis systems, Lab on-a-chip, MEMS, and micro sensors and actuators. Ji Yoon Kang received the B.S., M.S., and Ph.D. degrees in Mechanical Engineering from Seoul National University, Korea, in 1990, 1992, and 1997, respectively. From 1997 to 2001, he was a Research scientist at Samsung Advanced Institute of Science and Technology, Korea. From 2003 to 2004, he was a visiting scholar at the University of Cincinnati. In 2001, he joined the Korea Institute of Science and Technology, Seoul, Korea, where he is currently a Principal Researcher. His research interests include microfluidics, biosensor, and implantable micro-devices. Soo Hyun Lee received the B.S. degree from the Department of Electrical Engineering, Korea University, Seoul, Korea, in 2002, the M.S. degree and Ph.D. degree from the Department of Electrical and Computer Engineering and Computer Science, University of Cincinnati, in 2004 and 2007. In 2006, he joined the Korea Institute of Science and Technology (KIST), Seoul, Korea, where he is currently a Senior Researcher. His research interests include implantable sensors, neural electrodes, biosensors, and cell chips.
Contents lists available at ScienceDirect
Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb
Enzymeless glucose sensor integrated with chronically implantable nerve cuff electrode for in-situ inflammation monitoring Yi Jae Lee a , Sung Jin Park a,b , Kwang-Seok Yun b , Ji Yoon Kang a , Soo Hyun Lee a,∗ a b
Center for BioMicrosystems, Korea Institute of Science and Technology, Seoul 136-791, Republic of Korea Department of Electronic Engineering, Sogang University, 1 Shinsoo-dong, Mapo-gu, Seoul 121-742, Republic of Korea
a r t i c l e
i n f o
Article history: Received 18 May 2015 Received in revised form 29 July 2015 Accepted 20 August 2015 Available online 21 August 2015 Keywords: Enzymeless Glucose sensor Nerve cuff electrode Black Pt Nafion Inflammation Chronic implantation
a b s t r a c t Using glucose concentration as an inflammation responsive element, we newly established an enzymeless glucose sensor integrated with a chronically implantable peripheral nerve cuff electrode for continuous and in-situ monitoring of local inflammation. The glucose sensor integrated with a nerve cuff electrode was fabricated on a polyimide substrate side-by-side, then the glucose sensor and nerve cuff electrode were reversely folded, and were located inside and outside, respectively. The experimental results reveal that the electroplated black Pt working electrode of the glucose sensor shows an enhancive surface roughness factor of 16.41 and had a good distribution on the flexible polyimide surface, which exhibits distinctly enhanced electro-catalytic activity compared to that obtained with plain Pt. Amperometry and electrochemical impedance spectroscopy indicated that the fabricated sensor had a sensitivity of 7.17 A/mM cm2 , an outstanding detection limit of 10 M, significant selectivity, and excellent recovery performance for enzymeless glucose detection. In order to evaluate the feasibility for inflammation monitoring in the immediate vicinity of the implantable peripheral nerve cuff electrode, the association of an evoked nerve signal recording and glucose concentration was investigated through ex-vivo test using the sciatic nerve of a SD rat. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Nerve cuff electrodes have been widely used for chronic nerve signal recording and stimulation [1–3]. However, much of the implanted device itself tends to induce a foreign body reaction and morphological changes such as the growth of surrounding connective tissue, nerve reshaping, epineurial fibrosis, and degeneration/regeneration of myelinated fibers [4]. In particular, fibrotic reaction is preceded by an important epineurial inflammation, which can alter the interface between the nerve and cuff electrode. Therefore, the development of inflammation suppression and ability to monitor in a local area is vital for chronically implanted devices. Generally, an inflammatory reaction is the body’s attempt to protect, maintain and try to remove something harmful or that is irritating in the body, specifically, acute inflammation reflects that the body’s attempt to heal itself. Sometimes, long-term inflammation (chronic inflammation) can be a contributing factor to many serious and chronic diseases such as cancer, heart disease, diabetes,
∗ Corresponding author. Tel.: +82 2 958 6755; fax: +82 2 958 6910. E-mail addresses: [email protected], [email protected] (S.H. Lee). http://dx.doi.org/10.1016/j.snb.2015.08.091 0925-4005/© 2015 Elsevier B.V. All rights reserved.
autoimmune diseases, and Alzheimer’s disease [5]. Generally, when assessing the extent or activity of inflammation, serum C-reactive protein (CRP) level provides a rough guide to the amount of tissue involved in inflammation and to the integrity of the inflammatory response. The CRP method used in the laboratory is considered a more direct measure of the inflammatory process, while erythrocyte sedimentation rate (ESR) is a more indirect measure. ESR reflects the concentration of several plasma proteins including fibrinogen, ␣-globulins, -globulins, immunoglobulins and albumin. Therefore, any condition (pathological or non-pathological) that affects any of the contributing proteins can alter the ESR. Recently, several studies regarding the association of inflammation and glucose disorders have been reported [6,7]. The possibility that the inflammation affects the concentration of glucose in the immediate vicinity of a glucose sensor has also been suggested [8]. Specifically, a previous study reported increased glucose consumption that was double in wounded rat tissue compared to that of normal tissue [9]. Additionally, the degradation in the sensitivity of implanted devices may be caused by changes in the surrounding tissue due to an inflammatory response or from the formation of a fibrotic capsule that surrounds the implanted device. Therefore, monitoring inflammation in the immediate vicinity of the implanted device is essential for the stable operation of
426
Y.J. Lee et al. / Sensors and Actuators B 222 (2016) 425–432
chronically implanted devices, e.g., a nerve cuff electrode. Although there are many types of glucose sensors, such as a micro-needle, planar, and patch [10,11], most have only focused on monitoring body status through the glucose level in blood or interstitial fluid. The enzymeless glucose sensors are largely classified as potentiometric sensors, e.g., polymer membrane including boronic acid derivatives coated sensor, voltammetric sensors, e.g., sensor using ferrocene unit for voltammetric read-out-unit, and amperometric sensors using applying programmed potential pulsed or to monitor the current at a fixed potential [12]. More specifically in amperometric methods, enzymeless glucose sensors have been classified that use fixed potential amperometric analysis [13], medium fixed potential amperometry coupled with of flow injection analysis [14] and liquid chromatography, and variable potential amperometric methods such as potential sweep and pulsed voltammetry [15]. Amperometric analysis based on high catalytic nano-materials is one of the most widely used methods due to enhanced amperometric detection avoiding electroactive interference [16]. To our knowledge, this is the first study that directly applies a robust glucose sensor using black Pt integrated with an implantable nerve cuff electrode for inflammation monitoring of a peripheral nerve. The enzymeless glucose sensor with black Pt is operated by direct electro-oxidation of glucose via fixed potential amperometric analysis, which is considerably depending on the electrode material used. Pt is one of the most studied noble metals in the field of sensors and catalysts. However, conventional Pt has drawbacks of low sensitivity, selectivity, often suffer from losing activity due to easily poisoned by adsorbed intermediates. While noble metal nanoparticles are the most widely used nanomaterials in electrochemical sensor configuration due to their enhanced electro-catalytic properties. Particularly, electroplated black Pt is an enhancive material for sensor sensitivity due to its improved oxidation rate by enlarged real surface area and has excellent biocompatibility [17,18]. The electro-catalytic oxidation of glucose on Pt nanoparticles is a kinetic controlled reaction, whereas the responses from the common interfering species such as ascorbic acid and uric acid will not be changed significantly since their oxidation are diffusion controlled [19]. In this study, an enzymeless glucose sensor integrated with a chronically implantable peripheral nerve cuff electrode was designed, fabricated and characterized for in-situ inflammation monitoring. For sensitive and selective detection of glucose that was stable and unaffected by the denaturing of an immobilized enzyme, black Pt was selectively electroplated onto a plain Pt working electrode of an enzymeless glucose sensor. Subsequently, it was deep-coated with Nafion to avoid anionic interfering species. Glucose sensitivity was compared with and without a Nafion coating. The electrochemical properties of a fabricated enzymeless glucose sensor integrated with a nerve cuff electrode were investigated by cyclic voltammetry, electrochemical impedance spectroscopy, and chronoamperometry. In addition, the fabricated nerve cuff electrode integrated with a glucose sensor was tested to confirm whether the nerve cuff electrode could receive a stable nerve signal transfer at various glucose concentrations using a rat sciatic nerve through ex-vivo test.
2. Experimental 2.1. Reagents and apparatus HCPA (hexachloroplatinic acid hydrate, 99.9% purity, Aldrich), hydrochloric acid, and lead acetate (reagent grade, 95%) were prepared to make the Pt black electrode. The fabricated electrode was measured to check its surface roughness factor (RF) in a 1 M sulfuric acid (H2 SO4 , 95–98%, ACS, Sigma-Aldrich) solution using cyclic
voltammetry. The 1 M H2 SO4 was prepared by dilution in deionized water. The -d(+) glucose (99.5%, Sigma) stock solution was prepared by diluting it in a 0.1 M PBS solution and allowing it to stand for 24 h before use, in order to create equilibration. The 0.1 M urea (98%, Sigma), uric acid (UA, 99%, Sigma), sucrose (99.5%, Sigma), and ascorbic acid (AA, 98%, Sigma) solution were prepared by diluting them in a 0.1 M phosphate buffered saline (PBS, pH 7.0) solution. Nafion (5 wt.% solution) as a coating material for selectivity was obtained from Sigma Aldrich, Korea. The electrochemical experiments on the fabricated electrodes were performed by using Autolab (PGSTAT 302N, NOVA software, Ecochemie, Utrecht, The Netherlands) employing a three-electrode configuration with a fabricated black Pt for the working electrode (WE), a sputtered Pt as a counter electrode (CE), and an Ag/AgCl reference electrode (RE) at room temperature. The cyclic voltammograms of the fabricated sensor was recorded from −0.5 to +1.0 V in a range with a 50 mV/s scan rate. Chronoamperometry was performed in a 0.1 M PBS (10 mL volume) solution under continuous stirring to dilute the glucose with a constant concentration (0.5 M). The response current was recorded after stabilizing the background current at a nearly constant value and the prepared glucose were consecutively injected at a regular interval of 100 s. Surface morphology was characterized using a Hitachi S-2300 scanning electron microscope (SEM). 2.2. Fabrication of an enzymeless glucose sensor integrated with a nerve cuff electrode Fig. 1 provides conceptual drawings of an enzymeless glucose sensor integrated with a nerve cuff electrode for inflammation monitoring in a peripheral nerve. As shown in Fig. 1(a), the nerve cuff electrode consisted of four Pt line electrodes (0.25 mm × 1 mm) and the glucose sensor consisted of black Pt as a WE (geometry area: 4 mm2 ), sputtered Pt as a CE, and Ag/AgCl as a RE. The fabricated glucose sensor integrated with a nerve cuff electrode was reversely folded to the nerve cuff electrode (Fig. 1b). Fig. 1(c and d) shows a wrapped glucose sensor (outside) integrated with a nerve cuff electrode (inside) on a peripheral nerve and local inflammation monitoring of glucose sensor via fluctuation of glucose concentration by the inflammatory activation. The fabrication process for the enzymeless glucose sensor integrated with a nerve cuff electrode is described in Supplementary data Fig. S1 and in our previous research [20]. The first polyimide (VTEC 1388, Richard Blaine International, Inc., Philadelphia, PA, USA) layer of a 20 m thickness was formed on a 4-in. silicon wafer by spin-coating (Supplementary data Fig. S1a). After curing the wafer at 90 ◦ C for 10 min, 110 ◦ C for 10 min, and 220 ◦ C for 60 min in a convection oven, sputtered Ti/Pt (50/300 nm) layers on top of the 1st polyimide layer were formed by using a lift-off process (Supplementary data Fig. S1b). Next, the 2nd polyimide (5 m thickness) layer for insulation was formed by spin-coating. In order to selectively open the electrode site and the connector pads, a positive photoresist (AZ 9260, AZ Electronic Materials, NJ, USA) was patterned on the 2nd polyimide layer and the exposed polyimide patterns were etched by reactive ion etching (RIE) (Plasma Therm, St. Petersburg, FL, USA) (Supplementary data Fig. S1c). A laser dicing machine (M-2000, Exitech, Oxford, UK) was used to cut the perimeter of the electrode substrate (Supplementary data Fig. S1d). To achieve a cuff shape with a 1 mm diameter capable of wrapping around the sciatic nerve of a rat, the flat nerve electrode and glucose sensor electrode were reversely folded and reformed using a metal rod, and then cured in a convection oven at 220 ◦ C for 2 h to permanently fix the shape of the cuff (Supplementary data Fig. S1e). Then the black Pt was selectively electroplated on the sputtered Pt WE using a potentiostatic mode (−200 mV, 100 s).
Y.J. Lee et al. / Sensors and Actuators B 222 (2016) 425–432
427
Fig. 1. (a) Conceptual drawings of the enzymeless glucose sensor integrated with a nerve electrode, (b) the reversely folded glucose sensor and nerve electrode, (c) the wrapped glucose sensor integrated with nerve cuff electrode at peripheral nerve, and (d) the local inflammation monitoring of glucose sensor via fluctuation of glucose concentration by the inflammatory activation.
Fig. 2 shows the fabricated enzymeless glucose sensor integrated with a nerve electrode and SEM images of the glucose sensor. As shown in Fig. 2(a), the fabricated nerve electrode and glucose sensor were located equally, then each of the parts were reversely folded and cured into a cuff shape (inside: nerve electrode, outside: glucose sensor). The selectively electroplated black Pt on top of sputtered Pt WE was relatively well distributed with a similar grain size as shown in Fig. 2(b). 2.3. Ex-vivo nerve signal recordings in the sciatic nerve Ex-vivo experiments were performed using a three-month-old Sprague-Dawley (SD) rat. The animal was kept and handled in accordance with the regulations of the Institutional Animal Care and Use Committee (IACUC) of the Korea Institute of Science and Technology (KIST). The SD rat was kept in a temperature-controlled room at 24 ◦ C, and was fed food and water in an ad libitum feeder. The experimental protocol used for this study was approved by the Ethical Committee (KIST IACUC approval #2014-019) for Animal Experiments. During surgical extraction, the SD rat was anesthetized with an anesthetic mixed with Zoletil 50 (Virbac, Carros, France) and Rompun (Bayer, Seoul, Korea).
The sciatic nerve in the right leg of the SD rat was dissected. Then, the dissected sciatic nerve was placed in a PBS solution. The stimulation cuff electrode and recording cuff electrodes integrated with a glucose sensor were wrapped around the extracted sciatic nerve, respectively. The gap between the stimulating and recording cuff electrodes was 5 mm when wrapped around the extracted sciatic nerve. The ground electrode was grounded in another PBS solution. The stimulating cuff electrode was connected to a pulse stimulator (Isolated Pulse Stimulator, Model 2100, A-M Systems, Sequim, WA, USA). The recording cuff electrode integrated with a glucose sensor was partially connected to a differential amplifier (Differential AC amplifier, Model 1700, A-M systems, Sequim, WA, USA) and AutoLab. The amplified neural signals of the sciatic nerve were collected in a data acquisition (DAQ) device (NI USB-6356, National Instruments, Seoul, Korea). Then, the collected signals were processed using LabVIEW software (National Instruments, Seoul, Korea) and displayed on a laptop computer. The electrical stimulation was applied with pulse amplitude of 100, 200, and 300 A and a pulse width of 100 s. The differential amplifier gain was a factor of 1000.
Fig. 2. Photographs of the fabricated enzymeless glucose sensor integrated with nerve cuff electrode (a) and the SEM images of the glucose sensor (gap size of among WE, RE, and CE: 10 m).
428
Y.J. Lee et al. / Sensors and Actuators B 222 (2016) 425–432
the consecutive increment of 2 mM glucose, changes of the oxidation current were observed and the voltammetric response also increased along with increments of glucose concentration. The glucose oxidation peak currents might be attributed to typical Pt oxidation as described in the following reactions: Pt + 2OH− − 2e− → Pt(OH)2 Pt(OH)2 + 2OH− → PtO(OH)2 + H2 O + 2e− PtO(OH)2 + glucose → Pt(OH)2 + gluconolactone Especially, the oxidation current around 0.4 V increased significantly with an increase of glucose concentration, indicating the response of the sensor to glucose. Thus, 0.4 V was selected for an amperometric test to minimize the oxidation of interfering species. Electrochemical impedance spectroscopy and voltammetric responses of the fabricated enzymeless glucose sensor with plain Pt and black Pt WE are exhibited in Supplementary data Fig. S2(a and b). The measured impedance of the glucose sensor with black Pt WE was extremely lowered due to an enlarged surface area (plain Pt: 600 , black Pt: 40 at 1 kHz). Unlike the black Pt WE, the distinguished response peak current as increment of glucose concentration was not observed in cyclic voltammograms of plain Pt WE (Supplementary data Fig. S2b). 3.2. Amperometric behavior of the glucose sensor integrated with a nerve cuff electrode
Fig. 3. Cyclic voltammograms of the plain Pt WE and black Pt WE in 1 M sulfuric acid solution (a) and for the black Pt WE to various glucose concentrations (no glucose, 2, 4, 6, 8, 10 mM) in 0.1 M PBS (b). Scan rate: 50 mV/s.
3. Results and discussions 3.1. Voltammetric behavior of the glucose sensor integrated with a nerve cuff electrode Fig. 3(a) shows a cyclic voltammogram of the fabricated sensor with black Pt and plain Pt WE in 1 M sulfuric acid solution to compare the roughness factor at a hydrogen adsorption region. The dual peaks area (0 ∼ −0.45 V) of the black Pt WE in the cathodic scan indicates a reduction charge of hydrogen at the electrode surface. Since the charge was necessary to form a monolayer of adsorbed hydrogen and the WE area was covered by each hydrogen atom, the electrochemical roughness factor (RF) could be calculated by dividing the adsorption charge by 0.21 mC/cm2 [21]. The accumulated charge density in a region of Fig. 3(a) of the plain Pt WE and black Pt WE were 0.21 and 3.45 mC/cm2 , respectively, which were divided by the aforementioned charge density of the adsorbed hydrogen monolayer (0.21 mC/cm2 ). The corresponding RF of the plain Pt and black Pt WE of the sensor were 1 and 16.41(n = 3), respectively. The glucose sensor with black Pt WE showed an enhanced current response due to an extremely enlarged surface activation area. Fig. 3(b) shows cyclic voltammograms of the black Pt WE to various glucose concentrations in 0.1 M PBS. The voltammetric behaviors of the glucose sensor with black Pt WE was obtained at a scan rate of 50 mV/s and the potential ranged from −0.5 to 1 V vs. Ag/AgCl in a 0.1 M PBS solution without and with the addition of glucose (2, 4, 6, 8, and 10 mM). As shown in Fig. 3(b), upon
In order to achieve practical use, anti-fouling, and avoid reactions to anionic interfering species surrounding a peripheral nerve [22], the fabricated glucose sensor with black Pt WE was coated with Nafion. Glucose sensitivity with and without Nafion coating was compared. Nafion is the most widely used biocompatible material due to its simple handling and commercial availability [23]. The amperometric current response of the glucose sensor with black Pt WE and a Nafion coated black Pt WE are shown in Fig. 4(a and b). Consecutive addition of different concentrations of glucose (0.01, 0.025, 0.05, 0.1, 0.5, 1, and 2 mM) while stirring PBS at an applied potential of 0.4 V after stabilizing the background current at a nearly constant value with 100 s intervals. The current response as increment of glucose concentration was proportionally increased due to covered black Pt nanoparticles, which indicates that the Pt nanoparticles provide an high effective area and the glucose are electrolyzed as if in a thin layer cell. The shorten diffusion distance to electrode by electroplated Pt nanoparticles might be attributed to highly enhanced current response as well as change of diffusion regime [24]. Furthermore, the electroplated black Pt nanoparticles as porous-like surface in the SEM image of Fig. 2(b) might be lead to enhanced glucose oxidation current. Unlike semiinfinite diffusion of plain Pt electrode, the formed porous-like surface of the black Pt nanoparticles might be slightly affected on selectivity of the enzymeless glucose sensor (see Fig. 5a). The linear regression curve of the sensor with black Pt WE and Nafion coated black Pt WE in Fig. 4(c) is linear over a glucose concentration up to 22 mM (R2 = 0.93835 and 0.97936) and had a high sensitivity of 7.17 (286. 8) and 2.57 (103.1) A/mM cm2 (nA/mM), respectively. The sensitivity of the glucose sensor with plain Pt was only 0.02 A/mM cm2 (Supplementary data Fig. S2c). The detection limit of the sensor with black Pt and Nafion coated black Pt WE was estimated to be about 10 and 50 M, respectively. The detection limit was calculated from signal-to-noise (S/N = 3) ratio method. The estimated sensitivity of the fabricated glucose sensor is comparable with those of glucose sensors tested through a similar approach as previous reported 465.9 ± 48 nA/mM (Gr Ox /Pt-black) [25], 0.531 ± 0.149 nA/mM (Pt black/MWNT/Nafion/GOx ) [26]. The
Y.J. Lee et al. / Sensors and Actuators B 222 (2016) 425–432
429
Fig. 5. Amperometric response of 1 mM glucose, 0.1 mM urea, 0.1 mM UA, 0.1 mM sucrose, and 0.1 mM AA at the sensor with black Pt WE and Nafion coated black Pt WE (a) and the amperometric response recordings for recovery characteristic of glucose sensor with black Pt WE and Nafion coated black Pt WE in step change of glucose concentration (b).
Fig. 4. Comparison of amperometric response for the sensor with black Pt WE (a) and Nafion coated black Pt WE (b) to consecutive addition of glucose (0.01, 0.025, 0.05, 0.1, 0.5, 1, 2 mM) with interval of 100 s and linear regression for current to glucose concentration (error bars represent standard error of the mean) (c). The applied potential: 0.4 V.
comparison for the properties of enzymatic/enzymeless glucose sensors and this work was shown in Supplementary data Table S1. This result could be attributed to the large surface-to-volume ratio, high conductivity and facilitation of electron transfer by the black Pt basement, indicating it is a good platform for construction of an enzymeless sensor. The maximum relative standard deviation (RSD) of the current responses to 20 mM glucose is 12.3 (black Pt) and 14.4% (Nafion coated black Pt) for three successive
measurements, respectively (Fig. 4c). From the adsorbed glucose electro-oxidation on black Pt electrode at 0.4 V, gluconolactone is obtained as byproduct, which hydrolysis to gluconic acid on standing [16]. The response current of the glucose sensor with black Pt WE as increment of glucose concentration was slightly affected by generated oxidation products and intermediates during glucose oxidation. On the contrary, the glucose sensor with Nafion coated black Pt WE might be effectively avoided from the competitive adsorbed other anions such as phosphate anions, chloride anions. Several typical interfering species including urea, UA, sucrose, and AA were chosen to test the selectivity of the enzymeless glucose sensor. Although the glucose sensor with black Pt WE was slightly affected by the AA, the sensor with Nafion coated black Pt WE effectively avoided interference from AA (Fig. 5a). However, the consideration for non-glucose sugars (mono and polysaccharides) such as maltose, icodextrin, xylose, and galactose on fabricated glucose sensor is additionally required due to their severe interference with glucose analysis, e.g., overestimation of glucose level. The securing of glucose selectivity to non-glucose sugars is common problem for solution on enzymatic sensors in spite of their high selectivity as well as enzymeless sensors that goes without saying [16]. Recently, glucose dehydrogenase (GDH)-nicotinamide adenine dinucleotide (NAD) [27] or GDH-flavin adenine dinucleotide (FAD) based sensor [28], and enzymeless sensor with three
430
Y.J. Lee et al. / Sensors and Actuators B 222 (2016) 425–432
Fig. 6. Experimental setup of stimulus cuff electrode and recording cuff electrode integrated with enzymeless glucose sensor with dissected rat’s sciatic nerve for acute ex-vivo sciatic nerve signal recording (a), acute ex-vivo recordings to the biphasic electrical stimulation (100, 200, and 300 A) (b), acute ex-vivo recordings to the various glucose concentrations (from no glucose to 20 mM) at 300 A electrical stimulation (c), and plot for concentration of glucose versus evoked nerve compound action potential peak to peak (d). Input pulse duration was 100 s.
dimensional network of ZnO–CuO nanocomposites [29] were reported as platforms for free of these interferences. In order to prove that the enzymeless glucose sensor with black Pt WE and Nafion coated black Pt WE are a good platform for the construction of in-situ glucose monitoring of inflammation in tissues, an amperometric current to step-change glucose concentration (no glucose, 2, 5, 10 mM) was used to study the recovery characteristics of the as-prepared sensor with black Pt WE and Nafion coated black Pt WE as shown in Fig. 5(b). All measurements were performed by a stepwise change of glucose concentration at 100 sec regular intervals. The current responses of black Pt WE and Nafion coated black Pt WE were very stable toward both increments and decrements of glucose concentration. The recovery rates of the black Pt WE and Nafion coated black Pt WE for no glucose, 2, and 5 mM were 127.5, 91.6, 85.2% (black Pt) and 108.1, 111.9, 104.2% (Nafion coated black Pt), respectively. Recovery rate was defined by the following equation: recouvery rate =
current response by 2nd glucose concentration × 100 current response by 1st glucose concentration (1)
3.3. Ex-vivo test As shown in Fig. 6(a), a fabricated enzymeless glucose sensor integrated with a nerve cuff electrode was employed to record the
compound action potential of a rat’s sciatic nerve in increments of electrical stimulation (100, 200, and 300 A) and glucose concentration up to 20 mM. The dissected sciatic nerve was wrapped by a stimulating cuff electrode and a recording cuff electrode integrated with a glucose sensor to record compound action potential. The stimulating cuff electrode (5 mm width) had the same electrode configuration (4 line electrodes) as the recording cuff electrode (12 mm width). When the sciatic nerve was excited by the stimulating cuff electrode, a compound action potential through the nerve was transferred and recorded by the recording cuff electrode integrated with a glucose sensor. Fig. 6(b) shows the typical responses of the sciatic nerve recorded from the cuff electrode integrated with a glucose sensor at a 6 mM glucose concentration. The compound action potential of the sciatic nerve was simultaneously generated after the electrical stimulation given by the stimulation electrode. The insets of Fig. 6(b) represent a zoomed-in view of the evoked nerve signal and applied biphasic pulse current, respectively. In Fig. 6(b), the stimulus artifact signal and evoked nerve signal in the sciatic nerve were clearly and distinctly recorded as the stimulation intensity was increased from 100 to 300 A. The evoked nerve signal amplitudes to 100, 200, and 300 A peak current amplitudes were 0.608, 0.982, and 1.248 mV peak to peak, respectively. This demonstrated that the cuff electrode integrated with a glucose sensor was capable of functioning as a simultaneous nerve signal recording electrode. According to the results in Fig. 6(c and d), the transferred compound action potential can be simultaneously recorded
Y.J. Lee et al. / Sensors and Actuators B 222 (2016) 425–432
by the nerve cuff electrode integrated with a glucose sensor at various glucose concentrations. The experimental range for the glucose concentration was defined by normal ranges (Normal range: 4.2–5.3 mM, approximate short-term nonlethal limit: 1.1–83.3 mM) of important extracellular fluid constituents [30]. An electrical biphasic stimulation current was applied at 300 A with a 100 s duration. Since the evoked nerve signal was affected by the glucose concentration, and nerve injury is affected by hypoglycemia or hyperglycemia [31], the amplitude of the compound action potential was sharply increased up to a glucose concentration of 6 mM, then the potential was slightly decreased in a glucose concentration range from 8 to 18 mM. Finally, the evoked potential was sharply decreased at a high glucose concentration of 20 mM (Fig. 6d). This result indicates that the evoked nerve signal might be affected by the glucose concentration in the immediate vicinity of the cuff electrode. Therefore, local glucose concentration fluctuation in the immediate vicinity of the cuff electrode can be sufficiently considered as an indirect inflammation marker, because glucose capacity to use glucose as a fuel is particularly correlated to the inflammatory status of macrophages [32]. Although the activity of macrophages in adipose tissue is not yet clear, there is considerable evidence that they participate in the onset of insulin resistance and the development of diabetes associated with obesity via the production of an inflammatory reaction [33]. When onset of inflammatory activation in the local area of an implanted nerve cuff electrode occurs, an integrated glucose sensor in the local area can be used to directly monitor any remarkably decreased glucose level due to macrophage activation.
4. Conclusion This study introduced an enzymeless glucose sensor integrated with a nerve cuff electrode. The sensor was newly designed, fabricated, and characterized for in-situ and local inflammation monitoring in the immediate vicinity of a chronically implanted nerve cuff electrode. The fabricated glucose sensor with black Pt as a working electrode had a relatively large surface RF value when compared with plain Pt, which means black Pt has a larger electrode activation area over the other type. Furthermore, the sensor with black Pt had high sensitivity and a low detection limit in various glucose concentrations without any enzyme and mediators. The glucose sensor with a Nafion coated black Pt electrode exhibited excellent interference removal characteristic for selective glucose detection and recovery properties for continuous glucose concentration fluctuations. In addition, the fabricated nerve cuff electrode integrated with a glucose sensor was investigated for feasibility and nerve signal transduction in terms of variance in glucose concentration via ex-vivo test using a SD rat’s sciatic nerve. Since drastic local fluctuations of glucose concentration can be considered as triggering pro-inflammatory activation, sensitive and stable glucose monitoring in the immediate vicinity of the nerve cuff electrode is imperative for chronic stability in nerve signal recording. In the near future, the multilateral relevance of glucose concentration for in vivo inflammatory status will be investigated. It is likely that an understanding of glucose metabolism via local implantable devices will aid in fine-tuning treatment for proinflammation or chronic inflammation.
Acknowledgements This research was partially supported by the Happy Tech Program through the National Research Foundation of Korea (NRF) (grant number: 2009-0091918, 2010-0020786), which was funded by the Ministry of Education, Science and Technology (grant
431
number: 2009-0091918, 2010-0020786) and the KIST Institutional Program (2E25474).
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.snb.2015.08.091.
References [1] J. Badia, T. Boretius, D. Andreu, C. Azevedo-Coste, T. Stieglitz, X. Navarro, Comparative analysis of transverse intrafascicular multichannel, longitudinal intrafascicular and multipolar cuff electrodes for the selective stimulation of nerve fascicles, J. Neural Eng. 8 (Jun) (2011) 036023. [2] E.L. Foldes, D.M. Ackermann, N. Bhadra, K.L. Kilgore, N. Bhadra, Design, fabrication and evaluation of a conforming circumpolar peripheral nerve cuff electrode for acute experimental use, J. Neurosci. Methods 196 (Mar) (2011) 31–37. [3] N.S. Korivi, P.K. Ajmera, Self-closing cuff electrode for functional neural stimulation and recording, J. Med. Biol. Eng. 31 (2011) 353–357. [4] V. Vince, M.E. Brelen, J. Delbeke, I.M. Colin, Anti-TNF-alpha reduces the inflammatory reaction associated with cuff electrode implantation around the sciatic nerve, J. Neuroimmunol. 165 (Aug) (2005) 121–128. [5] M.M. Bosma-den Boer, M.L. van Wetten, L. Pruimboom, Chronic inflammatory diseases are stimulated by current lifestyle: how diet, stress levels and medication prevent our body from recovering, Nutr. Metab. 9 (Apr) (2012) 32. [6] J.I. Barzilay, L. Abraham, S.R. Heckbert, M. Cushman, L.H. Kuller, H.E. Resnick, et al., The relation of markers of inflammation to the development of glucose disorders in the elderly—the cardiovascular health study, Diabetes 50 (Oct) (2001) 2384–2389. [7] Y. Pan, Y. Wang, L. Cai, Y. Cai, J. Hu, C. Yu, et al., Inhibition of high glucose-induced inflammatory response and macrophage infiltration by a novel curcumin derivative prevents renal injury in diabetic rats, Br. J. Pharmacol. 166 (Jun) (2012) 1169–1182. [8] K. Rebrin, U. Fischer, H. Hahn von Dorsche, T. von Woetke, P. Abel, E. Brunstein, Subcutaneous glucose monitoring by means of electrochemical sensors: fiction or reality? J. Biomed. Eng. 14 (Jan) (1992) 33–40. [9] J. Forster, A.S. Morris, J.D. Shearer, B. Mastrofrancesco, K.C. Inman, R.G. Lawler, et al., Glucose uptake and flux through phosphofructokinase in wounded rat skeletal muscle, Am. J. Physiol. 256 (Jun) (1989) E788–E797. [10] S. Ichimori, K. Nishida, S. Shimoda, T. Sekigami, Y. Matsuo, K. Ichinose, et al., Development of a highly responsive needle-type glucose sensor using polyimide for a wearable artificial endocrine pancreas, J. Artif. Organs 9 (2006) 105–113. [11] M. Yasuzawa, S. Sato, H. Nakanishi, K. Edagawa, Fabrication of minimally-invasive patch type glucose sensors, Chem. Sens. 10 Mems/Nems 10 50 (2012) 83–87. [12] S. Park, H. Boo, T.D. Chung, Electrochemical non-enzymatic glucose sensors, Anal. Chim. Acta 556 (2006) 46–57. [13] T. Alizadeh, S. Mirzagholipur, A Nafion-free non-enzymatic amperometric glucose sensor based on copper oxide nanoparticles–graphene nanocomposite, Sens. Actuators, B: Chem. 198 (2014) 438–447. [14] C.-L. Sun, W.-L. Cheng, T.-K. Hsu, C.-W. Chang, J.-L. Chang, J.-M. Zen, Ultrasensitive and highly stable nonenzymatic glucose sensor by a CuO/graphene-modified screen-printed carbon electrode integrated with flow-injection analysis, Electrochem. Commun. 30 (2013) 91–94. [15] L. Rassaei, F. Marken, Pulse-voltammetric glucose detection at gold junction electrodes, Anal. Chem. 82 (2010) 7063–7067 (2010). [16] K.E. Toghill, R.G. Compton, Electrochemical non-enzymatic glucose sensors: a perspective and an evaluation, Int. J. Electrochem. Sci. 5 (Sep) (2010) 1246–1301. [17] Y. Li, C. Sella, F. Lemaitre, M.G. Collignon, L. Thouin, C. Amatore, Highly sensitive platinum-black coated platinum electrodes for electrochemical detection of hydrogen peroxide and nitrite in microchannel, Electroanalysis 25 (Apr) (2013) 895–902. [18] S.Q. Gu, Y.L. Lu, Y.P. Ding, L. Li, H.S. Song, J.H. Wang, et al., A droplet-based microfluidic electrochemical sensor using platinum-black microelectrode and its application in high sensitive glucose sensing, Biosens. Bioelectron. 55 (May) (2014) 106–112. [19] L.-Q. Rong, C. Yang, Q.-Y. Qian, X.-H. Xia, Study of the nonenzymatic glucose sensor based on highly dispersed Pt nanoparticles supported on carbon nanotubes, Talanta 72 (2007) 819–824. [20] S.H. Lee, J.H. Jung, Y.M. Chae, J.K.F. Suh, J.Y. Kang, Fabrication and characterization of implantable and flexible nerve cuff electrodes with Pt, Ir and IrOx films deposited by RF sputtering, J. Micromech. Microeng. 20 (Mar) (2010) 035015. [21] T. Biegler, D.A.J. Rand, R. Woods, Limiting oxygen coverage on platinized platinum—relevance to determination of real platinum area by hydrogen adsorption, J. Electroanal. Chem. 29 (1971) 269–270. [22] B. Akle, W. Habchi, R. Abdelnour, J. Blottman Iii, D. Leo, Biologically inspired highly efficient buoyancy engine, Proc. SPIE 8339 (Apr) (2012) 833900.
432
Y.J. Lee et al. / Sensors and Actuators B 222 (2016) 425–432
[23] H.B. Noh, M.S. Won, Y.B. Shim, Selective nonenzymatic bilirubin detection in blood samples using a Nafion/Mn–Cu sensor, Biosens. Bioelectron. 61 (Nov) (2014) 554–561. [24] B.R. Kozub, N.V. Rees, R.G. Compton, Electrochemical determination of nitrite at a bare glassy carbon electrode; why chemically modify electrodes? Sens. Actuators, B: Chem. 143 (2010) 539–546. [25] J. Shi, H.Y. Zhang, A. Snyder, M.X. Wang, J. Xie, D.M. Porterfield, et al., An aqueous media based approach for the preparation of a biosensor platform composed of graphene oxide and Pt-black, Biosens. Bioelectron. 38 (Oct–Dec) (2012) 314–320. [26] E.S. McLamore, J. Shi, D. Jaroch, J.C. Claussen, A. Uchida, Y. Jiang, et al., A self referencing platinum nanoparticle decorated enzyme-based microbiosensor for real time measurement of physiological glucose transport, Biosens. Bioelectron. 26 (Jan) (2011) 2237–2245. [27] D.M. Kim, M.Y. Kim, S.S. Reddy, J. Cho, C.H. Cho, S. Jung, et al., Electron-transfer mediator for a NAD-glucose dehydrogenase-based glucose sensor, Anal. Chem. 85 (Dec) (2013) 11643–11649. ˇ ´ K. Luˇspai, P. Magdolen, E. Sturdík, [28] R. Monoˇsík, M. Stred’ansky, Amperometric glucose biosensor utilizing FAD-dependent glucose dehydrogenase immobilized on nanocomposite electrode, Enzyme Microb. Technol. 50 (2012) 227–232. [29] C. Zhou, L. Xu, J. Song, R. Xing, S. Xu, D. Liu, et al., Ultrasensitive non-enzymatic glucose sensor based on three-dimensional network of ZnO–CuO hierarchical nanocomposites by electrospinning, Sci. Rep. 4 (2014) 7382. [30] M. V. T. D. S. R. Anura Kurpad and E. India, Guyton & Hall: Textbook of Medical Physiology—A South Asian Edition, Elsevier India, 2013. [31] M.M. Stecker, M. Stevenson, Effect of glucose concentration on peripheral nerve and its response to anoxia, Muscle Nerve 49 (Mar) (2014) 370– 377. [32] A. Johnson, A. Freemerman, E. Abel, J. Rathmell, L. Makowski, Glucose metabolism is linked to the inflammatory status of macrophages, BMC Proc. 6 (2012) P62. [33] A. Miranville, A.W. Herling, G. Biemer-Daub, M.D. Voss, Reversal of inflammation-induced impairment of glucose uptake in adipocytes by direct effect of CB1 antagonism on adipose tissue macrophages, Obesity 18 (Dec) (2010) 2247–2254.
Biographies Yi Jae Lee received the B.S.E.E., M.S.E.E., and Ph.D. degrees from Kwangwoon University, Seoul, Korea in 2007, 2009 and 2013, respectively. He is a postdoctoral researcher at Korea Institute of Science and Technology (KIST), Seoul, Korea. His current research interests include biosensors, bioelectronics, neural electrodes, glucose fuel cells for implantable system applications. Sung Jin Park will obtain his Master’s degree in 2015 February and issued by Sogang University of Republic of Korea. Mr. Park is studying the functional nerve cuff electrode. Kwang-Seok Yun received his B.S. degree in Electronics Engineering from Kyungpook National University in 1996, M.S. and Ph.D. degrees in Electronics Engineering from Korea Advanced Institute of Science and Technology (KAIST) in 1997 and 2002, respectively. He was a postdoctoral researcher at the University of California, Los Angeles, from 2005 to 2007. He joined the Department of electronic Engineering at Sogang University, Korea, in 2007, where he is now an associate professor. His current research area includes micro total analysis systems, Lab on-a-chip, MEMS, and micro sensors and actuators. Ji Yoon Kang received the B.S., M.S., and Ph.D. degrees in Mechanical Engineering from Seoul National University, Korea, in 1990, 1992, and 1997, respectively. From 1997 to 2001, he was a Research scientist at Samsung Advanced Institute of Science and Technology, Korea. From 2003 to 2004, he was a visiting scholar at the University of Cincinnati. In 2001, he joined the Korea Institute of Science and Technology, Seoul, Korea, where he is currently a Principal Researcher. His research interests include microfluidics, biosensor, and implantable micro-devices. Soo Hyun Lee received the B.S. degree from the Department of Electrical Engineering, Korea University, Seoul, Korea, in 2002, the M.S. degree and Ph.D. degree from the Department of Electrical and Computer Engineering and Computer Science, University of Cincinnati, in 2004 and 2007. In 2006, he joined the Korea Institute of Science and Technology (KIST), Seoul, Korea, where he is currently a Senior Researcher. His research interests include implantable sensors, neural electrodes, biosensors, and cell chips.