Carbon Nanotube–Bilirubin Oxidase Bioconjugate as a New Biofuel

Article pubs.acs.org/ac

Carbon Nanotube−Bilirubin Oxidase Bioconjugate as a New Biofuel Cell Label for Self-Powered Immunosensor Jiashun Cheng,† Yajing Han,† Liu Deng,*,† and Shaojun Guo*,‡ †

College of Chemistry and Chemical Engineering, Central South University, Changsha, Hunan People’s Republic of China 410083 Physical Chemistry and Applied Spectroscopy, Los Alamos National Laboratory, Los Alamos, New Mexico 87545, United States

‡

ABSTRACT: We demonstrated a biofuel cells (BFCs)-based self-powered sensing system for the detection of Nε(carboxymethyl)lysine (CML), in which the bilirubin oxidase (BOD)−carbon nanotube (CNT) bioconjugate modified with antibody acted as a biocatalyst for enhancing O2 reduction in the biocathode, as well as the transducing enzyme for signaling magnification. With an increase in the concentration of CML, the amount of BOD labels on biocathode surface increases, thus leading to the higher output of the as-prepared BFCs. This novel BFCs-based self-powered sensor showed a wide linear range for analyzing CML from 1 nM to 100 μM with a detection limit of 0.2 nM, which was 50 times more sensitive than that determined from the conventional ELISA. Most importantly, our new self-powered sensing platform can determine the level of CML in serum samples from multiple healthy donors and multiple sclerosis patients, being well in accordance with that from the commercial ELISA analysis.

A

provides the power for the sensing.7−9 Compared with traditional biosensors, one of the most significant advantages of BFCs-based biosensing systems is their ability to detect target analytes integrated with chemical-to-electrochemical energy transformation, resulting in no requirement for external power sources or control electronics. This concept has been well demonstrated in developing different kinds of self-powered biosensors for glucose, ethanol, and fructose, for example, because of the amplifying nature imparted by substrate turnover and the specificity of the biological recognition between biocatalyst and the substance.10−15 Furthermore, through the inhibitive and decoupler effect between the modulator and the biocatalyst (bilirubin oxidase; laccase and mitochondrial), the self-powered biosensor was utilized to detect the toxic substrate, such as cyanide, Hg2+, EDTA, and nitroaromatic compounds, among others.16−21 To explore the new scope of the self-powered biosensor, the combination of immunoreaction with glucose dehydrogenase (GDH) as bioanode was used to fabricate a miniaturized self-powered immunosensor for the detection of carcinoembryonic antigen (CEA) recently.22 However, the relatively low sensitivity and stability in the existing BFCs-based self-powered biosensors cannot meet the needs for higher detection sensitivity in earlier diagnosis. In this regard, rational design of electrode with excellent electrocatalytic performance for magnifying the sensitivity is highly desirable for achieving better BFCs-based self-powered sensors.

dvanced glycation end products (AGEs) are a diverse group of highly oxidant compounds through a nonenzymatic reaction between reducing sugars and free amino groups of proteins, lipids, or nucleic acids.1 AGEs are able to covalently cross-link with proteins, altering their structure and function in cellular matrix, basement membranes and vesselwall components. In addition, AGEs can also interact with a specific receptor for advanced glycation end products (RAGE), a member of the immunoglobulin superfamily, initiating signaling pathways that amplify inflammation and oxidative stress. Abundant evidence suggests that AGEs are of pathogenic significance in diabetes and several other chronic diseases, such as diabetes mellitus (DM), rheumatoid arthritis (RA) and Alzheimer’s disease (AD).2−5 To better understand the role of AGE in the modification of proteins in patients with diabetes, predict diabetic complication, and provide the monitoring indicators of clinical medicine to guide individual therapy, the accurate and high-sensitivity detection of AGEs in serum and tissue has become a very popular and challenging topic in clinic therapy and research area. To date, various methods have been explored to detect and quantify AGEs, including cation exchange chromatography, affinity chromatography, gas chromatography−mass spectrometry, electrospray ionization liquid chromatography mass spectrometry, or enzyme-linked immunosorbent assay (ELISA).5,6 However, the use of these traditional techniques for the detection of AGEs as earlier diagnosis is highly limited by their high cost and complicated operation process. To develop more interesting biosensing platforms, recently, a new concept based on biofuel cells (BFCs) was proposed to fabricate a self-powered biosensor, where the sensor itself © XXXX American Chemical Society

Received: September 1, 2014 Accepted: November 5, 2014

A

dx.doi.org/10.1021/ac503277w | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

Figure 1. (A) Configuration of immuno-regulated membraneless glucose/O2 BFC-based self-powered immunosensor and its detection mechanism for CML. (B) Configuration of immune-regulated biocathode.

accordance with commercial ELISA analysis. We expect this self-powered chip biosensor integrated with other electronics has the potential to fit the requirement of the totally selfsufficient lab-on-a-chip devices.

Herein, we demonstrated a new concept that a BFCs-based self-powered sensing system was used to develop the highly sensitive immnuosensor for the detection of Nε(carboxymethyl)lysine (CML) (a model of AGE), in which a carbon nanotube (CNT)−bilirubin oxidase (BOD) bioconjugate modified with biorecognition element (antibody) acted as a biocatalyst for enhancing O2 reduction in the biocathode as well as the transducing enzyme for signaling magnification. Figure 1A shows our BFC-based self-powered sensing strategy for the detection of CML. In the bioanode, the GDH was immobilized on the multilayer of CNTs/thionine/gold nanoparticles (AuNPs))8 for promoting the oxidation of glucose. In the biocathode (Figure 1B), CNTs were chosen as the carrier to load a higher amount of BOD to realize more efficient direct electron transfer (DET) for O2 reduction. Then, chitosanprotected graphene was used as electrochemical conducting matrix for the immobilization of capture antibody Ab1. A “sandwich” configuration between BOD/MWNT/mAb2 and Chi-GN/Ab1 in the presence of target (CML) was used to prepare the biocathode. With increasing concentration of CML, the amount of CNT−BOD bioconjugate on biocathode increases, thus leading to the higher power output of the asprepared BFCs. We found the as-prepared BFCs-based selfpowered biosensors can detect CML with a wide linear range of 1 nM to 100 μM and a detection limit of 0.2 nM. More importantly, our newly designed self-powered biosensor was 50 times more sensitive and had a 100 times wider detection range than that determined from the conventional CML ELISA kit. To demonstrate the clinical relevance of our biosensors, the levels of CML in serum samples from multiple healthy donors and multiple sclerosis patients were measured, being well in

■

EXPERIMENTAL SECTION Reagents and Materials. CNTs were purchased from Shenzhen Nanotech Port Co. Ltd. (NTP, China), and further treated with mixed acid (HNO3 and H2SO4) to get COOH group on their surface.23 Bilirubin Oxidase (BOD) from Myrothecium verrucaria, lyophilized 99% bovine serum albumin (BSA), lyophilized powder 96% human serum albumin (HSA), poly lysine (PLL), branched Polyethylenimine (B-PEI), 1ethyl-3-(3-dimethyllaminopropyl)carbodiimide hydrochloride (EDC), N-hydroxysuccinimide (NHS), KH 2 PO 4 , and K2HPO4 were purchased from Sigma-Aldrich. Nε-(1-Carboxymethyl)-L-lysine, Nε-(1-carboxyethyl)-L-lysine, and pentosidine (PTS) were purchased from Santa Cruz Biotechnology. The argpyrimidine was obtained from BOC science. CML polyclonal antibody was purchased from R&D Systems (Minneapolis, MN). Human carboxymethyl lysine and CML ELISA Kit (Abnova) were purchased from A&D Technology (Beijing). Antigens and antibodies were dissolved in pH 7.0 phosphate buffered saline (PBS) (0.01 M in phosphate, 0.14 M NaCl) unless otherwise noted. Glutaraldehyde and poly-Llysine (PLL) were purchased from Sigma. Graphene was prepared from graphite oxide (GO) through a thermal exfoliation method.24,25 The chitosan-protected graphene was prepared by dispersing 1 mg of graphene in 1 mL of 1% chitosan solution (pH 5.2). The mixture was sonicated for 1 h B

dx.doi.org/10.1021/ac503277w | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

Figure 2. (A) Electrochemical impedance spectroscopy of GN/GCE (blue), Ab1/GN/GCE (red), BOD/MWNT/mAb2/CML/Ab1/GN/GCE (black). (B) SEM image of the BOD/MWNT/mAb2/CML/Ab1/GN/GC in the presence of 1 mM CML.

μL of BOD/MWNT/mAb2 solution was dropped onto the electrode surface and incubated for 1 h at 37 °C. Fabrication of Microchip-Based BFCs. ITO glass plates were cleaned by ultrasonication in Milli-Q water, ethanol, NaOH ethanol solution, and ethanol to provide a clean, negatively charged surface. The chip electrodes were prepared on indium-doped tin oxide (ITO) glass using standard microfabrication technique.18 The negatively charged patterned ITO was first immersed in B-PEI aqueous solution (2 mg/mL) with 0.1 M NaCl for 15 min. The CNTs/thionine/Au NPs multilayers was fabricated the same way in our previous paper.26 The biocathode was fabricated as follows. The B-PEIterminated ITO electrode was first activated by EDC and NHS, then modified with Ab1 (10 μL of 10 μg/mL). The cell was covered with a clean glass dish and incubated overnight at room temperature, then washed with PBS buffer twice. Then, the cathode was blocked by adding 10 μL of blocking buffer (1% BSA in PBS, pH 7.4), further incubated at 37 °C for 1 h, and washed for three times using PBS buffer. To detect the CML, the cathode was drop casted by 10 μL of CML with different concentrations from 1 nM to 1 μM, followed by incubating for 1 h at 37 °C. Finally, 10 μL of BOD/MWNT/mAb2 was added to get a biocathode. All modification processes were carried out in a PDMS cell with Φ = 3 mm (the same size as the electrode surface) × 60 mm (height). Then, a PDMS flow cell with dimensions of 30 mm × 5 mm × 60 mm covered the two modified electrodes. Then, the electrolyte was added into the cell to generate the power output. BFCs-Based Self-Powered Sensors for Analyzing CML in Practical Samples. To assess the practical applicability of the present system, the detection of CML in plasma of multiple sclerosis (MS) and healthy controls (HCs) was carried out. Plasma samples were obtained from patients (confirmed by pathological examinations) at Xiangya Hospital (Hunan, China), and samples were stored at −80 °C before use. To prevent formation of AGEs from early glycation products during the sample preparation, plasma samples were reduced by sodium borohydride borate buffer (200 mM, pH 9.2) before precipitation. This mixture was allowed to stand for 2 h at room temperature. Proteins were then precipitated by the addition of 20% trichloroacetic acid and centrifuged for 20 min (4 °C) at 10 000 rpm. The protein pellet hydrolyzed by 6 M HCl and incubated for 18 h at 110 °C. After hydrolysis, these samples were used for the CML measurements. Instrument. The electrochemical measurements of bioelectrodes were performed using an CHI832C (Shanghai, Chenhua). Coiled platinum wire and an Ag/AgCl (saturated KCl) electrode were used as the counter electrode and the reference electrode, respectively. Current and potential output

to obtain a homogeneous dispersion. All solutions were prepared with double-distilled water. Preparation of CNT-BOD Bioconjugate Modified with mAb2. A mixture with 1 mg/mL CNT, 1.5 mg/mL PLL, and 1 M NaCl was sonicated at room temperature for 15 min, followed by centrifugation at 10 000 rpm at 4 °C for 10 min and then discarding the supernatant. The centrifugation/ discarding process was repeated three times to remove excessive PLL. Then, 50 μL of 1 M glutaraldehyde, 75 μL of 8 mg/mL of BOD, and 75 μL of 0.1 mg/mL detection antibody (mAb2) were added to the 50 μL PLL-functionalized CNT and stirred in a small vial overnight at room temperature. The reaction mixture was then centrifuged at 10 000 rpm at 4 °C for 15 min, and the supernatant was discarded to remove any free BOD and mAb2. The above wash process was repeated for four times. Finally, 100 μL of 0.01 M pH 7.0 PBS was added to the collected precipitate to form a homogeneous dispersion. The solution was stored at 4 °C and diluted with PBS buffer immediately before use. Optimization of CNT-BOD Bioconjugate Modified with mAb2. To optimize the amount of BOD used in probe preparation, the mAb2 and BOD concentration was fixed at 0.1 and 8 mg/mL respectively, and 10, 20, 25, 30, 40, 50, 60, 70, 75, 80, 100, 125, or 150 μL of the mAb2/BOD solution was added to 50 μL of a PLL-functionalized CNT solution. Then, the optimized BOD amount was fixed, and the mAb2/BOD ratios were changed to 1/50, 1/70, 1/80, 1/100, 1/150, or 1/ 200 to obtain the optimized ratio of BOD to mAb2, while the concentration of BOD was fixed at 8 mg/mL. The performance of CNT-BOD bioconjugate modified with mAb2 was evaluated using the polarization current of the prepared biocathode at 0.4 V in the presence of 100 nM CML. Fabrication of Biocathode. The glassy carbon electrode (GCE, 3 mm diameter) was first polished using 0.3 μm alumina powder and then thoroughly cleaned before use. Five microliters of the graphene-chitosan (GN-Chi) suspension (1 mg/mL) was dropped onto GCE. After the GCE was dried, 10 μL of freshly prepared 400 mM EDC and 100 mM NHS were dropped onto the GN-Chi/GCE for 30 min. Then, after removing the excess EDC and NHS by washing GN-Chi/GCE in PBS, the interface was further incubated in 25 μg/mL of Ab1 solution for 1 h at 37 °C. After the electrode was washed, it was blocked by adding 300 μL of blocking buffer (1%BSA in PBS, pH 7.4) at 37 °C for 40 min and then washed three times to block excess active groups and nonspecific binding between the antigen and the electrode surface. Thereafter, the Ab1/GNChi/GCE was immersed into the CML buffer solution with a different concentration and incubated for 1 h at 37 °C, followed by extensively removing unbounded CML molecules. Finally, 5 C

dx.doi.org/10.1021/ac503277w | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

Figure 3. (A) CVs of BOD/MWNT/mAb2/CML/Ab1/GN/GC electrode in a nitrogen-saturated (black line) and air-saturated (red line) solution at pH 7.0 with 10 μM CML. (B) Polarization curves of BOD/MWNT/mAb2/CML/Ab1/GN/GC electrode in air-saturated solution at pH 7.0 with 1 mM CML (red line) and 10 mM CML (black line).

Figure 4. (A) Polarization curves of the (CNT/thinione/AuNPs)8GDH anode (red line) and the immuno-regulated cathode (black line). (B) Dependence of the power density on the cell voltage in air-saturated quiescent solution with 100 μM CML, 15 mM glucose, and 0.2 M PBS (pH 7.5). (C, D) Optimization condition for the preparation of BOD/MWNT/mAb2 labels. (A) Different amounts of BOD were mixed with pretreated CNT solution. The volume ratios of BOD to CNT were as follows: 1/5, 2/5, 1/2, 3/5, 4/5, 1/1, 6/5, 7/5, 8/5, 3/2, 2/1, 5/2, and 3/1. (B) mAb2 and BOD with different mass ratios were mixed with pretreated CNT solution while the concentration of BOD was fixed. The molar ratios of mAb2 to BOD were as follows: 1/50, 1/70, 1/80, 1/100, 1/150, and 1/200.

conductive GN can form good electronic and ionic conduction pathways between the electrode and electrolyte. After the introduction of Ab1, the Ab1-immobilized GN/GC gives the biggest Ret, suggesting that the Ab1 layer formed an additional barrier for reducing the electron transfer of redox probe. After the BOD/MWNT/mAb2 was linked to electrode surface in the presence of target, CML, the interface resistance decreased because of the contribution of conducting BOD/MWNT/ mAb2. The assembly process was further confirmed by the scanning electron microscopy (SEM) (Figure 2B), showing the many randomly oriented CNTs in the assembly interface. Figure 3A (red line) shows cyclic voltammograms (CVs) of BOD/MWNT/mAb2/CML/Ab1/GN/GC electrode in an airsaturated solution (red line) and under anaerobic condition (black line) at pH 7.0 in the presence of 10 μM CML under anaerobic condition. It is found that the BOD/MWNT/mAb2/ CML/mAb1/GN/GC electrode does not produce an observable redox response under anaerobic condition, being consistent with previous reports that the direct electron transfer

of the biofuel cell system was measured by using a digital multimeter (Keithley 2700). We connected the two electrodes through an external 50 Ω resistor for assement of the BFC performance. All tests were conducted in a 25 °C temperaturecontrolled room. The scanning electron microscopy (SEM) image was determined with a Philips XL-30 ESEM. The accelerating voltage was 20 kV. The ELISA assay performed with a 96-well plate reader (The Sunrise Microplate, Tecan Trading AG, Switzerland) was based on a “sandwich enzymeimmunoassay” protocol involving two monoclonal antibodies directed against CML.

■

RESULTS AND DISCUSSION Characterization of Biocathode. The assembly process for immunoreaction-related biocathode was first monitored by impedance spectroscopy of the GN/GCE, Ab1/GN/GCE and BOD/CNT/mAb2/CML/Ab1/GN/GC using K4Fe(CN)6 as electrochemical redox probe, as shown in Figure 2A. The GN/ GCE shows the lowest resistance (Ret), indicating the highly D

dx.doi.org/10.1021/ac503277w | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

Figure 5. (A) Optimization condition for the working pH of the immuno-regulated membraneless glucose/O2 BFCs in the presence of 10 nM CML. (B) Polarization curves obtained at the immuno-regulated membraneless glucose/O2 BFCs in the presence of CML with different concentration: 1 nM (blue), 3 nM (red), 10 nM (menga), 100 nM (green), 1 μM (dark blue), and 100 Mm (black). Other conditions are the same as those in Figure 4B. (C) Linear relationship between Pmax and the logarithm of CML concentrations in as-prepared BFCs-based self-powered microchip sensor (black line) and the linear relationship between the absorbance and the logarithm of CML concentrations in the commercial CML ELISA kit (red line). (D) The specificity of the prepared self-powered immunosensor in the presence of CML (10 nM), CEL (10 μM), PTS (10 μM), and APR (10 μM).

(RSD = 1.3%, n =8). Compared with the reported miniaturized enzyme BFCs, the maximum power density is almost 10 times higher than that of the integrated bioanode combined with an external platinum cathode,28 four times higher than that of the three-dimensional microfluidic origami-based BFCs,22and 1.5 times higher than our previous microchip biofuel cell.18 Herein, the higher power density can be ascribed to the following facts: (a) the suitable microenvironment provided by CNT and PLL can well retain the BOD activity.29,30 (b) The strong interaction between proteins and CNT surface may allow the BOD to orientate in favorable fashion.23,31 (c) The abundant oxygen functionalities on the surface of CNT can work as the surface tethered mediators, thus facilitating the electron transfer from the electrode to surface bound BOD.23,25 High-Sensitivity Detection of CML. The process for preparation of the BOD/MWNT/mAb2 was optimized to achieve the best cathode performance for BFCs-based sensing. The volume ratio of BOD to CNT was first studied by mixing different amounts of BOD with pretreated CNT solution. We found that a volume ratio of 3 to 2 gave the highest O2 reduction current for the BOD/CNT/mAb2/CML/Ab1/GN/ GC electrode (Figure 4C). Because the linking efficiency of BOD to CNT decreases with the increase in mixture volume, the higher BOD/MWNT ratio does not give higher current output. The ratios of mAb2 to BOD were further optimized (Figure 4D). A ratio of 1/80 for mAb2/BOD results in the highest output current. When the molar ratio is higher than 1/ 80, more antibody molecules were linked to a CNT. In this regard, less BOD amount on the electrode interface results in less efficient catalysis of the BOD/CNT/mAb2 in reducing oxygen. It should be noted when the mass ratio is less than 1/ 80, a lower amount of antibody molecules was bound to CNT, thus decreasing the ability of bind the BOD/CNT/mAb2 to CML. In the immunoassays, various experimental parameters such as antigen−antibody interaction time and pH of medium for the detection of CML were optimized to achieve the best

of BOD was unable to be observed with CV on CNTs or carbon-based electrodes.27 However, in the presence of air, a much larger cathodic wave for oxygen reduction with the onset potential of 520 mV vs Ag/AgCl in the absence of mediator is observed. The onset potential of 520 mV is close to the redox potential of the T1 Cu site of BOD, which is the most plausible electron-accepting site of BOD.19,27 These results indicate that BOD immobilized on immunosensing platform can work as an efficient DET-type biocatalyst for O2 reduction. Further experiments also show that the catalytic peak current increases with the amount of target (CML) increasing (Figure 3B), indicating that the novel immune-regulated biocathode have a good potential in the development of new-concept immunosensors. Characterization of Biofuel Cells. Figure 4A shows the polarization curves of the (CNTs/thionine/AuNPs)8GDH film modified ITO anode (red curve) and also the BOD/CNT/ mAb2/CML/Ab1/GN/GC cathode in the presence of 100 μM CML, in a quiescent 15 mM glucose solution in air at 25 °C in 0.2 M PBS (pH 7.5). The onset potential for the electrooxidation of glucose is observed at −0.1 V, and the peak current density can reach a plateau at 140 μA cm−2 near 0.2 V. The onset potential for the oxygen electroreduction is observed at +0.51 V, and the current density reaches a plateau at 125 μA cm−2 near 0.1 V. These results show excellent biocatalytic activities of both bioanode and biocathode, and no crossover reaction between them can promise a high-performance onecompartment BFCs. This membraneless structure can simplify the chip fabrication, allow the cell stack up, and reduce the cost. As an extension, a microchip-based BFCs with both the anode and cathode on an ITO slide was also designed (Figure 1A). Figure 4B displays the power curve of the assembled microchip glucose/O2 BFCs in the presence of 100 μM CML. The open circuit voltage (Voc) of the cell is estimated to be 625 mV (relative standard deviation (RSD) =0.7%, n = 8) and the maximum power density (Pmax) reaches 455 μWcm2 at 410 mV E

dx.doi.org/10.1021/ac503277w | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

performance. The Pmax of the immune-regulated BFC toward 10 nM CML increases with an increase in the incubation time and reaches a plateau at 40 min, indicating that 40 min is enough time to finish the immunoreaction on the electrode surface. However, longer incubation time can result in a large nonspecific signal, and 40 min was chosen as the interaction time between antigen and antibody. The effect of pH of the antigen−antibody binding medium is examined over the pH of 5−9 (Figure 5A). The Pmax gradually increases from pH 5 to 7.5 and then rapidly decreases at pH values higher than 8. The Pmax decrease below 6 and over pH 8 might be due to the weak immunointeraction in acid and alkaline solution. The maximum reduction current was observed at the pH of 7.5. Thus, the optimum pH was chosen as 7.5. Under the optimal conditions, our self-powered BFC-based sensing platform was carried out via the polarization curve to confirm the feasibility of quantitative determination for CML. First, the biocathode reacts with different amounts of CML, and then it is treated with the BOD/MWNT/mAb2 to get the different power output of BFCs. Figure 5B shows the polarization curves of the sensor in the presence of various CML concentrations. It is observed that the highest power output values of the cell increases with increasing the CML concentration. As shown in Figure 5C, there is a linear relationship between Pmax and CML concentration from 1 nM to 100 μM (R2 = 0.999). Figure 6C also shows the comparison between our self-powered BFC-based sensing platform and conventional ELISA. Our self-powered immunosensor can detect at least 0.2 nM CML, which is 50 times more sensitive than commercial CML ELISA kit (10 nM CML). The higher sensitivity of BFC-based self-powered sensing platform should be attributed to the high signal amplification ability of our newly designed immuno-regulated biocathode. Figure 5D shows the selectivity of our self-powered immunosensor by the use of the carboxyethyl lysine (CEL), pentosidine (PTS), and argpyrimidine (APR), as interfering agents. The results show that these interfering agents lead to almost no signal increase and thus a satisfactory selectivity. The reproducibility of the self-powered immunosensor was examined by detecting 10 nM of CML with five immunosensors. The relative standard deviation (RSD), estimated from the slopes of the calibration plots of five different and freshly prepared immunosensors, was 7.6%, suggesting the good precision and reproducibility of the immunosensor. Here, to demonstrate the amenability of self-powered immunosensor in real samples, we carried out the assays of CML in serum samples from multiple healthy donors and multiple sclerosis (MS) patients. Accumulative data suggests that CML are extensively believed to be a biomarker of oxidative stress, which can be increased in MS patients, thus leading to axonal injury and neurodegeneration. As shown in Table 1, the CML levels from healthy donors and MS patients are estimated to be around 0.63 μM and 1.46 μM, respectively, being well in accordance with those from the ELISA analysis.

Table 1. Comparison of CML Level from Healthy Donors and MS Patients Determined by the As-Prepared SelfPowered Immunosensor and ELISA immunosensor (μM)

ELISA (μM)

sample no.

healthy donors

MS patients

healthy donors

MS patients

1 2 3 4 5 6 7 8

0.62 0.58 0.79 0.63 0.45 0.66 0.52 0.72

1.46 1.57 1.41 1.97 1.13 1.87 1.69 1.39

0.62 0.59 0.79 0.62 0.46 0.66 0.53 0.71

1.43 1.56 1.43 1.93 1.15 1.85 1.70 1.39

Under the new sensing mechanism, our self-powered immunosensor shows a wide linear range for the detection of CML from 1 nM to 100 μM with a detection limit of 0.2 nM. To demonstrate the clinical relevance of our biosensors, the levels of CML in serum samples from multiple healthy donors and multiple sclerosis (MS) patients were measured to be around 0.63 μM and 1.46 μM, respectively, being well in accordance with those from the ELISA analysis. It should be emphasized that the present self-powered BFC-based immunosensor can be actually accomplished only via a multimeter without the need of any large and expensive instruments in laboratories, which opens a simple, economical, sensitive, and portable device for the immunosensor. Most importantly, our miniaturized BFCs-based self-powered sensors show great potential in powering the small autonomous sensor-transmitter systems in animals and in plants, because its work condition is similar to the physiological one.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We are grateful to the National Natural Science Foundation of China (Nos. 21105126, 21076232, and 21276285) and the China Postdoctoral Science Foundation (Nos. 2011M500126, 2012T50656) for support.

■

REFERENCES

(1) Reddy, S.; Bichler, J.; Wells-Knecht, K. J.; Thorpe, S. R.; Baynes, J. W. Biochemistry 1995, 34, 10872. (2) Schleicher, E. D.; Wagner, E.; Nerlich, A. G. J. Clin. Invest. 1997, 99, 457. (3) Gaens, K. H.; Goossens, G. H.; Niessen, P. M.; van Greevenbroek, M. M.; van der Kallen, C. J.; Niessen, H. W.; Rensen, S. S.; Buurman, W. A.; Greve, J. W. M.; Blaak, E. E. Arterioscler., Thromb., Vasc. Biol. 2014, 34, 1199. (4) Hanssen, N. M.; Beulens, J. W.; van Dieren, S.; Scheijen, J. L.; Spijkerman, A. M.; van der Schouw, Y. T.; Stehouwer, C. D.; Schalkwijk, C. G. Diabetes 2014, DB_131864. (5) Sternberg, Z.; Hennies, C.; Sternberg, D.; Wang, P.; Kinkel, P.; Hojnacki, D.; Weinstock-Guttmann, B.; Munschauer, F. J. Neuroinflammation 2010, 7, 72. (6) Nagai, R.; Shirakawa, J.-i.; Ohno, R.-i.; Moroishi, N.; Nagai, M. Amino Acids 2014, 46, 261. (7) MacVittie, K.; Katz, E. Chem. Commun. 2014, 50, 4816.

■

CONCLUSION In summary, we have for the first time developed an immunoregulated glucose/O2 BFCs (a Voc of 625 mV and maximum power density of 455 μW cm−2)-based microchip immunosensors for the highly sensitive and selective detection of CML. We found that the maximum power density of the as-prepared biofuel cells is highly dependent on the amount of BOD label in the biocathode, thus the added amount of analyte (CML). F

dx.doi.org/10.1021/ac503277w | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

(8) Zhou, M.; Dong, S. J. Acc. Chem. Res. 2011, 44, 1232. (9) Zhou, M.; Wang, J. Electroanalysis 2012, 24, 197. (10) Blackwell, A. E.; Moehlenbrock, M. J.; Worsham, J. R.; Minteer, S. D. J. Nanosci. Nanotechnol. 2009, 9, 1714. (11) Gao, F.; Viry, L.; Maugey, M.; Poulin, P.; Mano, N. Nat. Commun. 2010, 1, 2. (12) Arechederra, R. L.; Minteer, S. D. Anal Bioanal Chem. 2011, 400, 1605. (13) Pan, C. F.; Li, Z. T.; Guo, W. X.; Zhu, J.; Wang, Z. L. Angew. Chem., Int. Ed. 2011, 50, 11192. (14) Cheng, H. J.; Yu, P.; Lu, X. L.; Lin, Y. Q.; Ohsaka, T.; Mao, L. Q. Analyst 2013, 138, 179. (15) Liu, Z. H.; Cho, B.; Ouyang, T. M.; Feldman, B. Anal. Chem. 2012, 84, 3403. (16) Meredith, M. T.; Minteer, S. D. Anal. Chem. 2011, 83, 5436. (17) Germain, M. N.; Arechederra, R. L.; Minteer, S. D. J. Am. Chem. Soc. 2008, 130, 15272. (18) Deng, L.; Chen, C. G.; Zhou, M.; Guo, S. J.; Wang, E. K.; Dong, S. J. Anal. Chem. 2010, 82, 4283. (19) Wen, D.; Deng, L.; Guo, S. J.; Dong, S. J. Anal. Chem. 2011, 83, 3968. (20) Zloczewska, A.; Celebanska, A.; Szot, K.; Tomaszewska, D.; Opallo, M.; Jonsson-Niedziolka, M. Biosens. Bioelectron. 2014, 54, 455. (21) Zhang, L. L.; Zhou, M.; Dong, S. J. Anal. Chem. 2012, 84, 10345. (22) Wang, Y. H.; Ge, L.; Wang, P. P.; Yan, M.; Yu, J. H.; Ge, S. G. Chem. Commun. 2014, 50, 1947. (23) Deng, L.; Liu, Y.; Yang, G.; Shang, L.; Wen, D.; Wang, F.; Xu, Z.; Dong, S. Biomacromolecules 2007, 8, 2063. (24) Stankovich, S.; Dikin, D. A.; Piner, R. D.; Kohlhaas, K. A.; Kleinhammes, A.; Jia, Y.; Wu, Y.; Nguyen, S. T.; Ruoff, R. S. Carbon 2007, 45, 1558. (25) Zhou, M.; Zhai, Y. M.; Dong, S. J. Anal. Chem. 2009, 81, 5603. (26) Deng, L.; Wang, Y.; Shang, L.; Wen, D.; Wang, F.; Dong, S. Biosens. Bioelectron. 2008, 24, 951. (27) Yan, Y.; Zheng, W.; Su, L.; Mao, L. Adv. Mater. 2006, 18, 2639. (28) Moore, C. M.; Minteer, S. D.; Martin, R. S. Lab Chip 2005, 5, 218. (29) Wen, Z. H.; Li, J. H. J. Mater. Chem. 2009, 19, 8707. (30) Deng, L.; Shang, L.; Wang, Y. Z.; Wang, T.; Chen, H. J.; Dong, S. J. Electrochem. Commun. 2008, 10, 1012. (31) Li, J. H.; Cheng, G. J.; Dong, S. J. J. Electroanal. Chem. 1996, 416, 97.

G

dx.doi.org/10.1021/ac503277w | Anal. Chem. XXXX, XXX, XXX−XXX