3.5.1 Optimization of enzymatic assay condition
3.5.1.1 Optimal wavelength for product absorbance
Figure9, the result of wavelength scan from 400 nm to 700 nm, shows that reaction product, quinoneimine, has absorption peak at 495 nm and this wavelength was set to be used in the following experiments.
3.5.1.2 Optimal pH value of phosphate buffer
Figure 28 shows the effect of different and pH value (6.0, 6.5, 7.0, 7.4, 7.5, 8.0,
and 8.5) working on assay efficiency. This experiment is under the standard condition that contains HRP 0.16Unit, phenol 14mM, 4-AAP 0.82mM, and PBS 0.2M.
According to the result of this experiment, we realize that initial rate in lower pH value performs better than higher pH value in this closed interval (from 6.0 to 8.0).
Therefore, we choose a pH 6.0 phosphate buffer for coupled enzyme assay.
3.5.1.3 Optimization of temperature
The relationship between different temperatures (from 20℃ to 40℃) and their initial rates of cholesterol assay is shown in figure 29 which tells that higher
temperature tends to speed the catalysis rate of cholesterol oxidase more efficiency. It is able to discern that the initial rate rises high at 37℃ and then returns to the original pattern that ascends gently.
Since there is no temperature controller integrated to our CMOS chip device, we do any experiment on traditional UV/Vis spectrometer in 25℃, which is selected to be the condition temperature that closes room temperature and thus we could compare the experimental results of our CMOS chip device and traditional UV/Vis
spectrometer.
3.5.1.4 Optimization of incubation time
From Figure 30 of different incubation time (0 min, 2 min, 3 min, 5 min, and 8 min) observers can see that while it takes longer time for incubation, the initial rate
value goes higher. To short the total reaction time and to find a proper time point to start a better condition simultaneously, we experiment enzyme assay after 3 minute for incubation and compare it with the one with no incubation time (0 min) later.
3.5.1.5 Optimal concentration of reagents 3.5.1.5.1 HRP activity standard curve
The linear range of HRP falling between 0.016~1 2 unit/ml was found in figure 31, thus appropriate amount of HRP is chosen to catalyze
phenol/4-aminoantypyrine/H2O2 chromogenic system.
3.5.1.5.2 Cholesterol oxidase activity standard curve
The cholesterol oxidase activity standard curve is shown in Figure 32. Studying the graph, we find out the linear range of cholesterol oxidase. The linear range falls between 0.0025~0.25 unit/ml, and then we choose the concentration falling in this interval to catalyze phenol/4-aminoantypyrine/H2O2 system. The concentration of CHOD is selected with 0.02 unit per milliliter.
3.5.1.5.3 Optimization of phenol concentration
Enough 4-aminoantypyrine is added in 4-aminoantypyrine/ phenol/H2O2
chromogenic system, the phenol concentration differs to ascertain the optimal concentration of phenol. By varying the concentration of phenol, we investigate the
optimal concentration of phenol in phenol/4-aminoantypyrine/H2O2 chromogenic system. Observing the kinetic graph, figure33, overmuch amount of phenol will inhibit the activity of HRP and decrease the initial rate of whole assay. The initial rate of absorbance accumulation tends to maximum and turns to decay gradually near 10 mM. Otherwise, to compare the result detected by CMOS chip with the former one by integrated photodiode system, we follow the phenol concentration used before which lies in the rational range of the kinetic graph. For this reason, we pick up 14 mM phenol for the following tests.
3.5.1.5.4 Optimization of 4-aminoantypyrine concentration
By the same way, we try to find out the adequate concentration for
4-aminoantypyrine. Figure34 displays that the initial rate of 4-aminoantypyrine tends to decrease as 4-AAP concentration increases in our detection range. In lower
concentration, the initial rate of whole assay is higher than it in higher 4-AAP
concentration. Considering that the stability of reaction and to compare our data with the experiment result from former tests, therefore we choose a concentration of 0.82 mM to be the fixed concentration of 4-aminoantypyrine. The 4-AAP concentration of 0.82 mM falls in the efficient zone of the kinetic graph as well. Besides, the delay in attained color for higher 4-AAP concentrations could be due to competition between phenol and 4-AAP in reaction with the oxidized enzyme (51).
After the amount of reagents is fixed, we couple the
phenol/4-aminoantypyrine/H2O2 system with cholesterol oxidation reaction to find out the linear range of cholesterol oxidase.
3.5.2 Chromogenic peroxidase enzyme reaction
The peroxidase enzyme reaction produces the chromogen, quinoneimine, with the detectable absorbance in 495 nm (concluded in the former experiment). Steps by steps, before we test the coupled enzyme reaction, the single equation responded to chromogen production also needs to be worked under the standard condition we set.
Peroxidase 2 H O + 4-AAP + Phenol ---Æ Quinoneimine + H O 2 2 2
Summing up the total experiments of assay optimization, we select the adequate reagent concentrations, instrument parameters and experimental conditions to be the standard condition. Under the standard condition, the phenol/4-aminoantypyrine/H2O2
chromogenic system is established to get the enzyme kinetic graph as standard curve for analysis. All of the curves and Km value are obtained with Michaelis-Menten equation in enzyme kinetics analysis of Sigma Plot 2001.
After testing peroxidase catalysis reaction by CMOS chip and UV/Vis
spectrometer, the kinetic graphs are compared with each other and result in similar Km
values. They both display a typical enzyme kinetic curve with the variation of H2O2.
As we see in Figure 35, the Km value of peroxidase is 1.61 × 10-1 mM detected by the
CMOS sensor chip and the one detected by UV/Vis spectrometer is 1.91 × 10-1 mM.
The data means that the CMOS chip is able to detect chromogenic signal with the variation of H2O2. The similar Km values of HRP can also be the reference to support CMOS chip as a multifunctional CMOS chip since the ability of chemiluminescene reorganization has already been proved in the former part of the report.
3.5.3 Cholesterol Coupled Enzyme Reaction
After the half part of the coupled enzyme reaction is proved working solidly, the test of the whole coupled enzyme reaction is performed in sequence. Coupled enzyme reaction is performed under the same condition as before with the variation of
cholesterol concentration.
3.5.3.1 Cholesterol oxidase-HRP coupled enzyme reaction with 2.5 Volts LED
supplied voltage
As cholesterol coupled enzyme reaction works, an external light resource is necessary to be set to offer the incident radiation source for absorption. Since the product shows a peak of absorption at 495 nm (figure9), a blue light emitting device is chosen to be coupled with the CMOS sensor.
A blue LED is assembled to our CMOS system served as light source plus 2.5 volts to drive it. Under this standard condition, we operate the cholesterol coupled enzyme reaction to compare the result with the one detected by traditional UV/Vis
spectrometer.
Figure 36 displays the kinetic curve of CHOD-HRP coupled enzyme reaction.
The Km value of cholesterol oxidase is 3.746 × 10-1 mM detected by CMOS chip, which did not approaches the Km = 1.067 × 10-1 mM obtained with traditional UV-vis spectrophotometer (UV-3310). To discuss the possible reason which causes the error between these two systems, the supplied voltage of LED is brought into discussion. A supplied voltage of 2.5 Volts added to LED leads to strong light source that may cause the saturation of the CMOS sensor and thus output voltage couldn’t respond the difference over the saturation threshold. To verify this point, a suitable supplied voltage is tested to get the optimal experimental parameter. We adjusted the supplied voltage of LED around 2.5 volts and found that the output voltage of CMOS chip keeps saturation till the LED supplied voltage is lower than 2.48 volts. Hence, the threshold of LED supplied voltage adaptable to the CMOS chip output voltage limitation is 2.48 volts and that is the reason why the variation of light intensity between 2.48 volts and 2.5 volts can not be distinguished and this phenomenon may causes errors.
3.5.3.2 Cholesterol Coupled Enzyme Reaction by LED supplied voltage adjusted to 2.48 Volts with incubation = 0 minute and 3 minutes
To demonstrate the points of above, we set the supplied voltage of LED to 2.48 volts and record the output data under the same condition. According to the figures of
the experiments under standard condition, we found that the adjustment of supplied voltage instead of 2.48 volts works out. Enzyme kinetic analysis of figure 37,38 illustrates that the Km value of cholesterol oxidase with 2.48 volts supplied voltage is approaches the Km value obtained with traditional UV-vis 3310 spectrophotometer (UV-3310). The outcome proved that the setting of appropriate voltage influence the accuracy of the sensor and that need to be concerned.
Incubation time alters while the experiment goes and we set two different incubation time while we performs our coupled enzyme reaction to find out the relation between incubation time and initial rate.
The two different incubation time conditions before reaction begins are 0 minute and 3 minutes. The figure in the first panel shows the result of the 0 minute incubation time and reveals that the Km value of cholesterol oxidase is 1.19 × 10-1 mM by this method, which approaches the Km = 1.067 × 10-1 mM obtained with traditional UV-vis spectrophotometer (UV-3300) (as figure 37 displayed).
After 3 minutes incubation, comparing the result operated under
spectrophotometer with the one resulted from CMOS chip detection as figure 38 shows. The comparison exhibits that the Km value with the longer incubation time did altered ( Km = 6.7 × 10-1 mM obtained by the CMOS chip and Km = 6.68 × 10-1 mM while detected by the UV-vis spectrophotometer), but these two parallel experiments
both results in similar Km with reability.
Besides, the Vmax of the initial rate with 3 minutes incubation time increases obviously from 0.06 delta A/ min to 0.37 delta A/ min detected by CMOS chip. That illustrates that longer experiment incubation time produce more products for the next going equation to cause reaction product, quinoneimine. The accumulation of
products during incubation time leads to the appearance of higher initial rate.
3.5.4 Comparison of Chrmogenic Reaction Detection Limitation
Table 7 shows the detection limitations of chromogenic reaction of three devices used for detectors, UV/ vis 3300 spectrophotometer, commercial photodiode, and the improved CMOS chip. The results are listed in the following table.
There are two tables of sensitivity about chromogenic assay. While table7 (a) displays the detection limit of the H2O2 -4-AAP – Phenol HRP reaction, table7 (b) shows the detection limit of the coupled enzyme reaction (cholesterol oxidation coupled with H2O2 -4-AAP – Phenol HRP reaction. Reading these tables, we realize that the detection limitations differs under two condition detected by different sensors.
In Table 8(a), the sensitivity of both UV-vis 3300 spectrometer and CMOS chip achieved are in the level of micro molar. Where the spectrometer has a minimum H2O2 concentration detected, 2μM; the CMOS chip is also with the minimum
concentration in the same level, 5μM. While detecting the absorbance of chromogen,
CMOS chip performs well even if it is compared with the large instrument.
Coupled with cholesterol oxidation for cholesterol examination, the sensitivities of all devices decrease. The efficiency of the conversion from cholesterol to H2O2
isn’t completed, so H2O2 provided to produce chromogen isn’t as much as singular reaction operated before. It is approximately 1 order of magnitude lower than the limit reported in the singular reaction. The detection limit of assembled semiconductor device is 0.02 mM while the cholesterol detection limit of UV-vis spectrophotometer is 0.01 mM. The performance of novel CMOS chip sensitivity is 0.1mM and it is almost 1 order of magnitude lower than the limit of spectrophotometer. But it works successfully in cholesterol examination and serves as a useful biosensor to measure the product absorbance.
The sensitivity got from chromogenic assay is relative to luminescence assay.
Although the CMOS chip performs less sensitively than the commercial photodiode S2387, the comparison of the sensing areas of devices indicates that the CMOS chip did have better unit sensitivity. If the sensor size is restricted for being easy to carry and use, the CMOS chip will be more suitable for scale diminution.
3.5.5 Results of the method for HRP-GOD coupled enzyme reaction
In the same way, glucose oxidation reaction could also be coupled to
same as luminescence method. The standard condition is detected by UV/vis 3310 and the result is shown in Figure 39 (a.). The Km of glucose is determined to be 0.05mM with PMT instrument (uv/vis 3310). The same reaction detected by CMOS photodiodes gives similar result as shown in Figure 39 (b.), (c.), and (d.) (Frequency parameters: 45hz, 50 hz, and 60 hz). Otherwise, among these detections under diverse frequency parameters, they both got a similar Km value as uv/vis 3310 but the graph with 45 Hz frequency setting displays a more similar Km with the one set in the same frequency parameter used for cholesterol detection (50 Hz). The kinetics of different frequency parameter are displayed in Figure 40(b.).
The result told that the quantification ability of our CMOS chip can be more accurate by modulate the frequency parameter. Like the luminescence test, Figure 40(a.), different frequency parameter resulted in different sensitivity. Notability, unlike luminescence test, lower frequency parameter inversely leaded into lower sensitivity. There is an opposite point against luminescence test and the difference could be explained by Figure 18. As Figure 18 illustrated, a higher frequency setting offered a higher sensitivity when a stronger light signal is being detected. The slope is steeper than that of a lower frequency parameter. Contrarily, a lower frequency setting gave a better sensitivity to detect dim signal. The results of the sensitivity, Km value, and saturated concentration of different frequency parameter are displayed in Table 10.
The characteristic is useful for us to quantify the light signals in different intensity level. Thus, the detections of different optical reactions might be more reliable under specific parameter setting.
4. Conclusion
This work reveals that the CMOS-based photodiode chip can be served as an optical biosensor applied to multiple bio-molecule recognition. Integrated with commercial LED device, the application of CMOS chips could be more flexible and broad and it is also tested in distinct optical assays such as luminescence and
colorimetric method. The new generation CMOS- based chips are also compared with the traditional large sized instrument, Hitachi UV/vis 3300 spectrometer and Hitachi F4500 luminescence spectrometer, and the commercial photodiode device.
The characteristics of frequency modulation of the CMOS chip helps users to detect diverse medical targets via different optical reactions. Since the different scanning rates of the function wave supplied to the CMOS chip are dominant in different light intensity level and lead into different sensitivity, the specific parameter setting optimized to different assays is helpful to make a multifunctional biosensor more precise and accurate.
The novel CMOS chip has been proved to detect at least two different optical signal: chemiluminescence and absorbance. Else, it has also been tested two kinds of biomedical targets: glucose and cholesterol. The tests of the novel chip confirm the potential to develop the novel CMOS chip into a multifunctional optical CMOS sensor. The less sample volume requirement, low cost, miniature, flexibility and the
disposable plastic cuvette used for CMOS chip are more benefit for a potable biosensor applied not only to clinical diagnosis but also blood composition analysis, environment monitoring and food quality control. (52-57).
In another hand, while the sensitivity of new generation CMOS chip is contrasted with the commercial photodiode device, though the commercial
photodiode contributes to higher sensitivity with its large sensing area. Comparatively, via conversion, the sensitivity of unit sensing area of both CMOS chip and
commercial photodiode is equal. It exhibits that the light transferring efficiency of integral CMOS chip performs as good as commercial photodiode device. Concerning the enzyme amount used for detection, our CMOS chip even take less enzyme amount than other silicon photodiode such as amorphorous photodiode (50). Besides, the characteristics of CMOS chips such as mass product, low cost, and integration are advantaged in practical application.
Else, the CMOS chip with multiple reaction detection ability will be designed to collect data of array composed of divided cells. These individual cells will set up a multifunctional diagnostic chip sensing different signals simultaneously.
In conclusion, by industrial fabrication, a more useful, sophisticated, powerful, and flexible optical sensor could be manufactured by combining both photodiodes and particular integrated circuits, for next generation biosensor market and study (58).
5. Reference
1. http://www.piribo.com/publications/medical_devices/biosensor_report.html 2. Mehrab Mehrvar, Mustafe ABDI, 2004. Recent developments, characteristics, and
potential applications of electrochemical biosensors. Analytical sciences. 20, 1113-1127.
3. Hofmann, O., Voirin, G., Niedermann, P., Manz, A., 2002. Three dimensional microfluidic confinement for efficient sample delivery to biosensor surfaces:
application to immunoassays on planar optical waveguides. Anal. Chem. 74, 5243–5250.
4. Kwakye, S., Baeumner, A., 2003. A microfluidic biosensor based on nucleic acid sequence recognition. Anal. Bioanal. Chem. 376, 1062–
1068.
5. Askari, M., Alarie, J.P., Moreno-Bondi, M., Vo-Dinh, T., 2001. Application of an antibody biochip for p53 detection and cancer diagnosis. Biotechnol. Prog. 17,
543–552.
6. Baeumner, A.J., Cohen, R.N., Miksic, V., Min, J., 2003. RNA biosensor for the rapid detection of viable Escherichia coli in drinking water.
Biosens. Bioelectron. 18, 405–413.
7. Choi, S.H., Gu, M.B., 2002. A portable toxicity biosensor using
freeze-dried recombinant bioluminescent bacteria. Biosens. Bioelectron.
17, 433–440.
8. DeBusschere, B.D., Kovacs, G.T., 2001. Portable cell-based biosensor system using integrated CMOS cell-cartridges. Biosens. Bioelectron.
16, 543–556.
9. Armin W. Wieder Franz Neppl, 1992. CMOS Technology Trends and Economics . IEEE micro. 12, 10-19.
http://ieeexplore.ieee.org/search/wrapper.jsp?arnumber=149732 10. http://www.dalsa.com/markets/ccd_vs_cmos.asp
11. Blum LJ, Gautier SM. Bioluminescence- and chemiluminescence-
based fiberoptic sensors. In: Blum LJ, Coulet PR, editors, Biosensor principles and applications. New York: Marcel Dekker, 1991:213–247.
12. Golden, J.P., Anderson, G.P., Rabbany, S.Y., Ligler, F.S., 1994. IEEE Trans.
Biomed. Eng. 41, 585–591.
13. Chen, Z., Kaplan, D.L., Gao, H., Kumar, J., Marx, K.A., Tripathy, S.K., 1996.
Mater. Sci. Eng. C 4, 155–159.
14. Marquette, C.A., Blum, L.J., 1999. Anal. Chim. Acta 381, 1–6.
15. Tsafack, V.C., Marquette, C.A., Pizzolato, F., Blum, L., 2000. Biosens.
Bioelectron.15, 125–133.
16. S.M.Sze (1985) Semiconductor Devices (physics and technology), 1st Ed., John Wiley & Sons, New York.
17. Marlene A.Deluca (1978) Preface. In Marlene A.Deluca, editor. Methods in Enzymology: Bioluminescence and Chemiluminescence, Academic Press.
18. Larry J.Kricka and Gary H.G.Thorpe (1990) Bioluminescent and
Chemiluminescent Detection of Horseradish Peroxidase Labels in Ligand Binder Assay. In Knox Van Dyke and Richard Van Dyke, editors. Luminescence
Immunoassay and Molecular Applications, CRC press.
19. Larry J.Kricka, Irena Bronstein, and Jhon C.Voyta (2000) Chemiiluminescence Methods for Detecting and Quantitating Enzyme Activity. In Mirian M.Ziegler and Thomas O.Baldwin, editors. Methods in Enzymology: Bioluminescence and Chemiluminescence, Acadamic Press.
20. Tan,A., Xiao,C., 1997. An automatic photometric method with hogh precsion by gradually diluting a sample. Talanta. 44, 2081-2086.
21. Copeland, R.A., ENZYMES A Practical Introduction to Structure, Mechanism, and Data Analysis 2nd, Wiley-Vch, New York, 2000 .
22. He,M.F., Clinical Biochemistry, He-Ji, Taipei, 1992.
23. Gaw,A., Murphy,M.J., Cowan,R.A., O’Reilly,D.St.J., Stewart,M.J.,
Shepherd,J.,Clinical BioChemistry 3rd, Churchill Livingstone, London,1995.
24. Nelson,D.L. and Cox,M.M., Lehninger Principles of Biochemistry 3rd, Worth, New York, 2000.
25. J. Zicha, J. Kunes, M.A. Devynek, 1999. Abnormalities of membrane function And lipid metabolism in hypertension. Am. J. Hypertens. 12, 315–331.
26. R. G. Bowden et al.,2006. A Cholesterol Primer for Health Professionals.
Californian Journal of Health Promotion 2006, 4, 12-21.
27. Charpentier, L., Murr, I.E., 1995. Amperometric determination of cholesterol in serum with use of a renewable surface peroxidase electrode. Anal. Chim. Acta 318, 89–93.
28. Martin, S.P., Lamb, D.J., Lynch, J.M., Reddy, S.M., 2003. Enzyme-based determination of cholesterol using the quartz crystal acoustic wave
sensor. Anal. Chim. Acta 487, 91–100.
29. Situmorang, M., Alexander, P.W., Hibbert, D.B., 1999. Flow injection
potentiometry for enzymatic assay of cholesterol with a tungsten electrode sensor.
Talanta. 49, 639–649.
30. M.D.Marazuela, B. Cuesta,M.C.Moreno-Bondi, 1997. A. Quejido, Free Cholesterol fiber-optic biosensor for serum samples with simplex optimization.
Biosens. Bioelectron. 12, 233–240.
31. G. Li, J.M Lia, G.Q. Hu, N.Z. Ma, P.J. Wu, 2005. Study of carbon nanotubes modified biosensor for monitoring total cholesterol in blood. Biosens. Bioelectron.
20. 2140–2144.
32. S. Singh, A. Chaubey, B.D. Malhotra, 2004. Amperometric cholesterol biosenor
32. S. Singh, A. Chaubey, B.D. Malhotra, 2004. Amperometric cholesterol biosenor