Research On The Fabrication And Application Of Several Novel Electrochemical Biosensors

Repeatedly occurring food safety affairs make food security become an attention focus, which turns into a world issue. Some of the foodborne pathogens, such as E. coli0157:H7, has become the main major cause of food safety. How to quickly and accurately detect these pathogens is primary task to ensure food safety. Electrochemical biosensor is a rapid detection methodology for pathogens. As one of electrochemical test methods, electrochemical impedance spectroscopy (EIS) has great potential to be used in the fast, low-cost, non-marking and non-invasive biological detection methodology. Electrochemical impedance measurement is also suitable for the development of small, easy to operate electronic detection device. However, the expense of the electrode and the sensitivity of the sensor would limit out-of-laboratory applications for economic and fast screening. As a consequence, the development of "user-friendly" biosensor which is lower cost, higher practicability and easier to operate is very vital and unavoidable.Considering the difficult problems of the electrochemical biosensor for rapid detection in food safety, this dissertation was devoted to the study of novel electrochemical biosensors which were combined with screen printing technology and nano-materials in the electrodes, biomolecules binding, signal processing, etc. Several different biosensor were built and the feasibility of the technology for rapid detection in food safety have been confirmed by the detection of E. coli O157:H7, Brucella, and cow mastitis which was caused by various of bacteria infection. The main work of this dissertation could be summarized as follows:(1) In order to find sensitive and quantitative methods to detect foodborne pathogens for food security, a sandwich immunochromatographic test strip was developed trying to integrate the electrical impedance analysis with immunochromatography technology. Screen-printed technology and nano-materials were used to improve the performance of the sensor. Firstly, three kinds of acid proton doped polyaniline were synthesized. The product doped by polystyrene sulfonic acid, with a better solubility (0.370g) and conductivity (2258.8μs/cm) than4-hydroxy acid and perchloric acid products, was chosen to label the antibody. And then the screen-printed electrodes were used to assemble the rapid detection of E. coli0157:H7immunochromatographic test strip. The feasibility of this test strip to qualitative detection of E. coli O157:H7was exploded by EIS. Impedance changes during the immunochromatography were characterized by impedance-time measurement to determine the time when immune responses became stabilized, and the electrical responses at multi-frequencies were investigated by EIS method to get the best test frequency. The results showed that, there was a good uniformity for screen-printed electrodes. The immune response became stabilized after180seconds after applying the analyte. The impedance of blank sample was significant different from samples with bacteria at the frequency less than100Hz, and the total impedance of the samples with bacteria were less than the blank solution, which proved the feasibility of this test strip to qualitative detection of E. coli O157:H7.(2) Based on the method of qualitative detection of E. coli0157:H7, the electrical responses at multi-frequencies were investigated to get a quantitative detection further. The E. coli O157:H7was detected quantitatively by the normalized impedance changes (NICz) of solutions with bacteria to the solutions without bacteria after immune responses became stabilized. The results showed that, the NICZ was linearly related to the logarithm of bacteria concentration in the range of1.15×103-1.15×106CFU/mL at1Hz. The regression function was NICz=0.0741Lg(C)+0.0108. Its determination coefficient was0.9616, and the minimum detection limit was7.08×102CFU/mL (S/N=3). The total test process (exception of the time for strip manufacturing) took about6min for each test. The method, for the advantages of rapidness, and simple operation, would allow the development of easy-to-use and portable devices in the future for the rapid monitoring of E. coli O157:H7. However, the detection limit needs to be further improved.(3) In order to improve the precision of the immunochromatographic strip, the complex Td correction method was first time applied to reduce the influence of parasitic capacitance of this strip under the condition of negligible electrode polarization effects for stimulation by a small voltage of AC. Firstly, the impedance of the "quasi-tissue"(a bio-system constrcted by E.coli O157:H7antigen, antibody, solution, the biofilm, and others) was corrected by the complex Td correction method. And then, the corrected data were fitted using the least-square method to get the Cole-Cole parameters in the admittance plane. A regression model between the concentration of E.coli O157:H7and the optimum values of the Cole-Cole parameters were selected by stepwise regression method. The result showed that the interference of the parasitic capacitance of the sensor was larger with wide randomness as the frequency increased. The minimum detection limit of the sensor was higher with the error caused by the parasitic capacitance, the electrode polarization and other factors. While after corrected by the complex Td factor correction method, the complex impedance data for the biosensor were rectified effectively at all frequencies, thereby the interference of parasitic capacitance was reduced. The Corrected data were in good agreement with the Cole-Cole track fitted by nonlinear least-square method. The Cole-Cole parameters were very sensitive to the antigen concentrations, which would provide enriched and reliable information for the qualitative and quantitative test of bacteria with the biosensor. The optimum characteristic parameters selected by stepwise regression in SAS software were G0、a and G02. The determination coefficient of the regression model was0.9792, and the limit of detection was able to reach39CFU/mL.(4) Brucellosis, especially that arose from Brucella melitensis, is often foodborne, and unpasteurized milk and dairy products are common causes in propagation of disease. In order to prevention of Brucellosis timely, minimizing economic losses and ensuring the supply of good quality milk, a label-free impedance immunosensor was developed based on GNPs-SPCE for Brucella melitensis detection. The disposable GNPs-SPCE was used for sensing by electrochemical impedance spectroscopy (EIS) in the presence of [Fe(CN)6]3-/4-]. A general electronic equivalent model was introduced for modeling the performance of the immunosensor. Among these impedance components, the greatest change was found in electron-transfer resistance due to the cell binding. The label-free impedance biosensor showed linearity from4×104to4x106CFU/mL with a detection limit of1x104CFU/mL in pure culture. When testing real-life samples, the biosensor was able to detect as low as6×104CFU/mL of Brucella melitensis in milk samples in less than1.5h. Interference from other bacteria was eliminated by the use of specific antibodies. This biosensing method uses a low-cost disposable electrode. It is hand-held operation, and no complex pretreatment is needed, thus would allow in the future the development of easy-to-use and portable devices for the rapid monitoring of Brucella melitensis.(5)The issue of food safety, such as mastitis, often caused by a variety of bacteria infections, and would resulted in somatic cell count (SCC) increasing significantly in milk which jeopardized our public health. Therefore, rapid, sensitive, and reliable detection of SCC in milk was significantly necessary in prevention of mastitis timely, minimizing economic losses and ensuring the supply of good quality milk. A sensor based on interdigitated microelectrode was developed for the rapid detection of SCC. Electric responses of this biosensor at multi-frequencies were explored by EIS method. And a model based on electrical parameters and support vector regression for quantitative detection of SCC was exploded. It was showed that there was a relationship between the electrical parameters and the grade of mastitis. One of the electrical parameters of RS was decreased with increasing of SCC, while other electrical parameters such as Q and a presented a nonlinear trend. The SVR model was good for prediction of SCC with high precision except the negative samples. The average relative error was29.40%, and the bovine mastitis detection rate were100%for1(sub-clinical),2(relative serious infection), and3(serious infection) level of samples.(6) In order to improve the accuracy of the sensor to the negative samples, a four-wire interdigital electrode was designed. The effect of using of different interdigital electrode spacing and test patterns (such as two-wire (A-D),(B-C) measurements, and the four-wire measurements) to reduce the interference of electrode polarization and parasitic capacitance was exploded. The results showed that the resistance of the milk sample was the main contributor to the total impedance with two-wire electrode mode at a frequency greater than10kHz, and the same to the measurement with four-wire electrode mode at a frequency greater than1kHz.. For three different test patterns, the distinction of the milk samples at the high-frequency was better than at the low-frequency. The distinction of the milk samples with the two-wire (A-D) measurement was beteer than the two-wire (B-C) measurement. And four-wire electrode measurement was better than the two-wire electrode measurement, since the interference introduced by electrode polarization, contact impedace and parasitic capacitance, ect. were reduced significantly, which made the measured impedance was much more close to the bulk impedance of milk.(7) To further develop a low-cost instruments with the need of only single frequency or several number of frequencies impedance data, a SCC detection method based on characteristic frequency was exploded. The stepwise regression method was used to extract the characteristic frequency from the impedance data obtained with the four-wire interdigital electrode measurement. A support vector regression model between the impedance data at the selected characteristic frequencies and the SCC was established. It was showed that the characteristic frequencies were11720Hz,9668Hz,8105Hz and1172Hz. The mastitis detection rate for T (suspicious) level of samples was the same to two-wire (AD) electrode measurement, while the detection rate of the model for the negative samples were improved. The mastitis detection rate for2(relative serious infection) and3(serious infection) level of samples were100%. The method achieved the purpose of dimensionality reduction successfully,and would be in favor of development of portable and user friendly devices for rapid monitoring of SCC.