Chapter 2. Materials and methods
2.3 Methods
Research scheme
In order to build up a biomedical sensor system, we assembled four main components: optimal conditions of bio-reactions, CMOS chip consist of photodiode and circuit, optical devices, and data process software, is considered reasonably as usual. Fig. 1 is the flowchart of our research.
Reagents preparation
Sodium phosphate buffer (0.1 M stock)
0.1 M dibasic phosphate solution was prepared by add ddH2O to 7.1 g Na2HPO4
until final volume 500 ml was obtained. 0.1 M monobasic phosphate solution was prepared by add ddH2O to 6.9g NaH2PO4‧H2O until final volume 500 ml was obtained. Then added NaH2PO4 solution to Na2HPO4 solution until the desired pH 7.0 was obtained. Phosphate buffer was stored at room temperature.
ABTS (2,2'-azino-bis(3-ethylbenzthiazoline-6-sulphonic acid) / 36 mM stock)
36mM ABTS solution was prepared by add phosphate buffer to 9 tabs ABTS (10
mg/tab) until final volume 4.556 ml was obtained. And then add NaOH to ABTS solution until the desired pH 7.0 was obtained. ABTS solution had to be prepared fresh and keep it from light lest it would go off.
H2O2 (1M stock)
Added ddH2O to 1ml H2O2 (30%, W/W) until final volume 9 ml was obtained. And then, we diluted this stock with ddH2O to obtained 0.1M and 0.01 M solution.
Glucose solution (1 M stock)
Glucose solution was prepared by adding ddH2O to dissolve 9 mg glucose powder until final volume 50ml was obtained. And then, we diluted this stock with ddH2O to obtained 0.1 M and 0.01 M solution.
Histamine solution (0.2 M stock)
Histamine solution was prepared by adding phosphate buffer t o dissolve 222 mg histamine powder until final volume 10ml was obtained. And then, we diluted this stock with dd H2O to obtained 20mM and 2mM solution.
Enzyme preparation
Horseradish peroxidase (HRP, EC 1.11.1.7)
The powder of HRP (4mg or 1000 units) was dissolved in 10ml sodium phosphate buffer (0.1 M at pH 7.0), and the stock (100 unit/ml) was stored at -80 . ℃ One unit of HRP will form 1.0 mg of purpurogallin from pyrogallol in 20 seconds at pH 6.0 at 20℃. This purpurogallin (20 seconds) unit is equivalent to approximately 18 μM units per minute at 25℃.
Glucose oxidase (GOx, EC 1.1.3.4)
The powder of GOx (6mg or 1200 units) was dissolved in 10ml sodium phosphate buffer (0.1 M at pH7.0), and the stock (120 unit/ml) was stored at -80 . One unit ℃ corresponds to the amount of enzyme which oxidizes 1 μmol glucose per minute at pH 7.0 and 25 °C.
Diamine oxidase (DAO, EC 1.4.3.6)
The powder of DAO (200mg or 10 units) was dissolved in 1ml sodium phosphate buffer (0.1 M at pH7.0), and the stock (10 unit/ml) was stored at -80 . One unit will ℃ oxidize 1.0 μmole of putrescine per hour at pH 7.2 and 37 °C.
Optimal condition for enzyme assay
Absorption wavelength of product
The “wavelength scan” function of UV-3310 was used to measure the range of absorption wavelength of reaction product. The scan parameters of UV-3310 are as follows: start wavelength is 700 nm, end wavelength is 300 nm, delay time is 120 sec, scan speed is 300 nm/min and silt is 2 nm. Reaction condition is: 25 , pH 7.0, 100 ℃ mM phosphate buffer, 30 mM ABTS, 0.2 mM H2O2 and 0.2 unit/ml HRP.
Optimal concentration of reagents
Optimization of substrate amount (H2O2, glucose, and histamine) played an important role in enzyme assay before the determination in the solution. The “time scan” function of UV-3310 was used to find out the optimal concentration of reagents.
The scan parameters of UV-3310 were: wavelength is 430nm (H2O2 / histamine) and 650nm (H2O2 / glucose), sampling interval is 1 sec; total times are 60 sec for H2O2, 120sec for glucose, and 1200sec for histamine, and delay time is zero. Several reaction conditions were kept under 25℃(H2O2 and glucose) and 37℃(histamine), pH 7.0, 100 mM phosphate buffer.
The optimization studies began with ABTS/H2O2 colorimetric system to find out the linear range of HRP, and then appropriate amount of HRP, excess ABTS, were
added to Fig. out the optimal concentration of reagents. After finding out the suiTABLE amount of other reagents, we coupled ABTS/H2O2 colorimetric system with glucose oxidation reaction to find out the linear range of glucose oxidase (650 nm). Repeating previous process, we also found out the linear range of diamine oxidase (430 nm).
Optimal pH value of buffer
The “time scan” function of UV-3310 was used to find out the optimal pH value of phosphate buffer. In the experiment of glucose, value of pH ranged from 3.0 to 9.0 with an internal of 1.0 and three kinds of buffer system, citrate buffer (pH 3.0~5.0), phosphate buffer (pH 5.0~6.0), Tris buffer (pH 7.0~9.0) were used. The scan parameters of UV-3310 are: wavelength is 650 nm, sampling interval is 1 sec, total time is 120 sec, and delay time is zero. Constant reaction conditions were: 25 , 30 ℃ mM ABTS, 20 mM glucose, 0.48unit/ml GOD and 0.2 unit/ml HRP. For determine histamine, value of pH range is from 6.5 to 7.5 with an internal of 0.5 and phosphate buffer (pH 6.5~7.5), was used. The scan parameters of UV-3310 are: wavelength is 430 nm, sampling interval is 1 sec, total time is 1200 sec, and delay time is zero.
Constant reaction condition is: 37 , 30 mM ABTS, 2 mM histamine, 0.8 unit/ml ℃ DAO and 0.2 unit/ml HRP.
Optimal temperature
The “time scan” function of UV-3310 was used to find out the optimal temperature for the coupled enzyme assay. Temperature ranges from 20 to 40 . All reagents ℃ ℃ except enzymes were mixed, and then enzymes were added into mixed reagents immediately before we began to measure the process of time scan. The scan parameters of UV-3310 were: wavelength is 650 nm (glucose), 430 nm (histamine), sampling interval is 1 sec, total times are 120 sec (glucose), 1200 sec (histamine) and delay time is zero. Constant reaction condition is: pH 7.0, 30 mM ABTS, 20 mM Glucose/ 2 mM Histamine, 0.48 unit/ml GOD / 0.8 unit/ml DAO, and 0.2 unit/ml HRP.
Coupled enzyme assay to determine concentration of substrate
Glucose determination
ABTS/H2O2 colorimetric system coupled to GOD catalyzed reaction could determine the concentration of glucose. The reaction conditions for this coupled enzymes system contained 0.48 unit/ml GOD, 0.2 unit/ml HRP, 30 mM ABTS and phosphate buffer (100 mM, pH 7.0) in a final volume of 1 ml at 25 .℃
Histamine determination
ABTS/H2O2 colorimetric system coupled to DAO catalyzed reaction can determine the concentration of histamine. The reaction conditions for this coupled enzymes system contained 0.8 unit/ml DAO, 0.2 unit/ml HRP, 30Mm ABTS and phosphate buffer (100 mM, pH 7.0) in a final volume of 1 ml at 37 .℃
Mixing procedure
Aliquot amount of enzyme is added into the cuvette following by the injection of all other necessary reagents and sample. These processes are to make sure that the whole compounds are well mixed in cuvette without extra shaking and the data can be collected in a short period of time [20].
Data processing
Each data point was calculated from the mean value of triplicate assay. The curve and Km value were obtained with Michaelis-Menten equation in enzyme kinetics analysis of Sigma Plot 2001.
CMOS chip design and biomedical sensor system setup
The flowchart of the proposed CMOS optical biosensing system is illustrated in Fig.
2. A narrowband LED was used as the light source, and a bandpass filter allowed the
light of a specific wavelength to travel through the cuvette, in which the light was modulated by the biochemical materials. The light, after passing through the cuvette, was detected by the CMOS biochip. The CMOS photo detector converted the modulated optical signals into current signals, and then the transimpedance amplifier (TIA) converted them into voltage signals. Finally, the signals were recorded by a multimeter and processed by personal computer.
CMOS photodiode
We applied the optical sensing theorem of light transition. And a novel design of photodiode with finger structure (p -n well) shortened the distance of electron drift, and increased the area of depletion layer closed to the surface. Therefore, the photodiode receipted carriers more efficiently, and then it could produce higher photosensitive current. When the interval of each bar is the twice width of depletion layer, the photodiode has the best sensitivity.
+
The CMOS biochip is mainly composed of a photo detector and a TIA. Fig. 3 illustrates the top view and cross section of the proposed P+/Nwell finger photodiode. The quantum efficiency of the photodiode is enhanced by adopting interdigitated shallow P+/Nwell junctions to extend the depletion regions
[27]. Each P+ stripe is 1μm wide with 4 μm spacing, and the total area of this finger photodiode is 100 μm ×100 μm. The photodiode is surrounded by double guard rings to keep the substrate noise off; the P+ guard ring is connected to ground, and the N+
guard ring is connected to a high potential [28]. Photodiode with finger structure (p+-n well), current amplifier, TIA circuit with variable resistance are designed by Yu-Wei Chang who is a Ph.D. student of the department of electronic engineering at NCTU.
Circuit design
As shown in Fig. 4, the subsequent circuits include a current source, current mirrors, a TIA, and output buffers. The TIA consists of a common-source stage (M0), two source followers (M1 and M2), and an external feedback resistance (RF). M1 serves in the feedback loop to isolate R0 from the loading effect, and M2 drives the load capacitance to alleviate the stability issue [29].
Light source module
We choose two high lightness red LED and a color filter with 650nm to form a
light source module, and select R2/1 MΩ (normal light intensity) on TIA. With preceding setup, we determinated glucose amount in the reagent. On the other hand,
we choose three high lightness violet LED and a color filter with 430 nm to form a light source module, and select R1/10 MΩ (low light intensity) on TIA.
Data analysis
Data analysis consists of hardware and software; therefore, we have to connect them with our biomedical sensor system. E3646A power supply provides CMOS chip 3.0 V bias voltage, and 4401A multimeter collects the output voltage data and transfers it to a personal computer (PC) through GPIB connection.
And we use Labview 6.0 to record the output voltage data transferred to PC.
CHAPETR 3. RESULTS 3.1 CMOS photodiode chip performance
The biochip is implemented in a standard 0.35-μm CMOS technology. The finger photodiode is reverse-biased moderately to ensure that the photocurrent is proportional to the absorbed photons. The photocurrent is then converted into voltage signals by the subsequent TIA. With a 3.0 V power supply, the lower bound and upper bound of the output voltage are 0.49 V and 2.02 V, respectively. (See Fig. 5 (a)
－(c)) Therefore, we obtained a linear range of output voltage to compare with the input light intensity which was proportioned to the concentration of -D(+)-glucose or histamine.
Circuit design
When photodiode absorbs photons, it will generate a flow of current in an external circuit, proportional to the incident light. The photocurrent was amplified by current amplifier, and then the current was translated into voltage output within 0.49~2.02 V by Transimpendence Amplifier (TIA). The circuit design is showed in Fig. 4. Three variable resistances were designed for low light intensity (R1/10 MΩ), normal light intensity (R2/1 MΩ), and high light intensity (R3/0.1 MΩ).
Before we know the corresponding concentration of reagents by measuring the
voltage value, the liner relation between light intensity and output voltage of CMOS chip must be confirmed. To affirm this, the first step, the module of LED was gave various input voltages, and a color filter was used to sift the specific-wavelength light out. In the meanwhile, the corresponding values of light intensity were measured by power meter. Then we could build up a curve of LED input voltage versus light intensity such as Fig. 5(a). The second step, we gave various LED input voltages as serious light intensity, and measure the corresponding output voltages of CMOS chip.
Then we could also build up another curve of LED input voltages versus CMOS chip output voltage such as Fig. 5(b). The third step, with merging two curves previously, we get a new curve of light intensity versus CMOS chip output voltage. It is showed as Fig. 5(c). According to the linear relation showed as Fig. 5(c), we can translate the CMOS chip output voltage collected by 4401A multimeter into transiting light intensity.
3.2 Optimization of enzyme assay
Absorption wavelength of product in glucose and histamine oxidation reaction
It illustrates the absorption peak of photodiode between 400 and 1000 nm, and it is obvious that the maximum absorption peak of photodiode within the visible light ranges between 600 and 700nm. Furthermore, it shows that the absorption sepctrum
of the product of reaction, ABTS
.
+, and it has three absorption peaks at 415 nm, 640 nm and 720 nm, separately (Fig. 7). Light 1 in Fig. 6 reveals the weaker absorption peak of photodiode and Light 2 shows the stronger one. After all we will confirm that our photodiode could have the corresponding performance and it could also determine the glucose amount in the reagent. Therefore, we choose 650 nm as the optimal absorptive wavelength in the glucose oxidation reaction.
On the other hand, we want to confirm that the performance of our photodiode could fit the optimal light absorbance peak of ABTS/H2O2 colorimetry system with the range from 390 nm to 440 nm. As a matter of fact, the product of coupled enzyme reactions catalyzing histamine had only one light absorbance peak at 430 nm (Fig. 8).
Therefore, we choose 430 nm as the optimal absorption wavelength of histamine oxidation reaction.
Optimal concentration of reagents
ABTS/H2O2 colorimetry system
There is a linear range of HRP between 0.05 to 0.2 U/ml (Fig. 9); therefore, we choose 0.2 U/ml of HRP to catalyze in the ABTS/H2O2 colorimetry system. We added excess H2O2 in ABTS/H2O2 colorimetry system with various concentrations of ABTS to investigate the optimal concentration of ABTS. In the consideration of the
solubility of ABTS and the catalytic rate in Fig. 10, we choose 30 mM to be the optimal concentration of ABTS to make sure of its continuous working.
Determination of optimal amount of glucose oxidase and diamine oxidase
After we found out the optimal concentration of other reagents, we coupled ABTS/H2O2 colorimetry system with glucose oxidation reaction to find out the linear range of glucose oxidase ranged from 0.01 to 0.8 U/ml as shown in Fig. 11. Therefore, we choose appropriate amount, 0.48 U/ml, of glucose oxidase to catalyze glucose oxidation reaction. Additionally, we coupled ABTS/H2O2 colorimetry system with histamine oxidation reaction to find out the linear range of diamine oxidase ranged from 0.4 to 2 U/ml as shown in Fig. 12. Therefore, we choose appropriate amount, 0.8 U/ml, of diamine oxidase to catalyze histamine oxidation reaction.
Optimal pH value of phosphate buffer
Fig. 13 demonstrates that the effect of pH value (3.0, 4.0, 5.0, 6.0, 7.0, 8.0, and
9.0) on the assay. In order to fit the application in biomedical diagnosis, we choose appropriate pH 7.0 to be the optimal pH condition in glucose oxidation reaction. In the side of histamine oxidation reaction, owing to the enzyme activity limitation, we choose the range between 6.5~7.5 to find out optimal pH value according to the
suggestion of sigma enzyme product data sheet. Fig. 14 tells us that the effect of pH value (6.6, 6.8, 7.0, 7.2, and 7.4) on assay. Base on the result of this experiment, the initial rate in pH value of 7.4 is optimal. Considering these two optimal pH value as described above, we choose a pH 7.0 phosphate buffer for coupled enzyme assay.
Optimal temperature
The relationship between different temperatures (20 ~40 ) and initial rates of ℃ ℃ glucose assay shows as Fig.15, it tells us the variation of temperature affect unobviously the catalysis rate of glucose oxidase. On the other hand, there is no temperature controlled setup on our assembled biosensor system. While proceeding any experiment on UV/Vis spectrometer U-3310, 25 is choose to be the optimal ℃ temperature.
Different from glucose oxidation reaction, the variation of temperature does affect the catalysis rate of diamine oxidase obviously. The relationship between different temperatures (20 ~40 ) and initial rates of diamine oxidase assay shows as ℃ ℃ Fig. 16, and it tells us initial rates of diamine oxidase has higher value between 36 and 40 ℃ ℃ than on room temperature. Considering the temperature of human body, we choose 37℃ as the optimal temperature of histamine oxidation reaction.
3.3 The determination of hydrogen dioxide
The CMOS chip output voltage at different hydrogen dioxide concentrations in the 650 nm and 430 nm systems were showed in Fig. 17(a) and Fig. 17(b). The hydrogen dioxide detection limit of assembled system in the 650 nm and 430 nm were 1 μM and almost were the same as that from UV/Vis spectrophotometer (shown in Fig.
18(a)-(b) and Fig. 19(a)-(b)). The liner detection ranges of our assembled biosensing system and UV/Vis spectrophotometer ranges from 1 ~ 20 μM and 10 ~ 80 μM, respectively(shown in Fig. 20 (a)-(b) and Fig. 21 (a)-(b) ).
3.4 Coupled enzyme assay to determine β-D(+)-glucose and histamine
Coupling glucose oxidase and diamine oxiadse catalyzed reaction with ABTS/H2O2 colorimetry system, we get the enzyme kinetic analysis of β-D(+)-glucose and histamine standard curves. All the curve and Km value are obtained with Michaelis-Menten equation in enzyme kinetics analysis of Sigma Plot 2001.
Enzyme kinetic analysis of glucose oxidase and diamine oxidase observed in our assembled biosensor system is shown in Fig. 22(a) and Fig. 23(a). By this process the Km value of β-D(+)-glucose and histamine were 4.9 ± 0.5 μM and 125 ± 3.21 μM, respectively, which approached to those obtained from UV/Vis spectrophotometer
(7.2 ± 0.2 μM and 160 ± 19.7 μM ). (See TABLE 3(a), TABLE3(b), Fig. 22(b) and
23(b)) The optimal reaction conditions of these two enzyme was showed in TABLE 2,
and the CMOS chip output voltage at different glucose and histamine concentrations were showed in Fig. 22(c) and Fig. 23(c). β-D(+)-glucose detection limit of assembled system is 1 μM and almost the same as that from UV/Vis spectrophotometer. On the other hand, the histamine detection limit of assembled system is 10 μM while the hitamine detection limit of UV-vis spectrophotometer is 20 μM. The liner detection ranges of our assembled biosensing system ranges from 1 μM to 5 mM and 10 ~ 80 μM, respectively.
CHAPTER 4. DISCUSSION Optimal absorptive wavelength
There are two factors to determine the optimal absorptive wavelength β-D(+)-glucose and histamine oxidation reactions: The absorption wavelength of the
products of β-D(+)-glucose and histamine oxidation reactions; and the absorption peak of CMOS photodiode chip. Owing to the maximum absorption peak of CMOS photodiode chip located between 600 and 700 nm and the absorption peaks of β-D(+)-glucose oxidation reaction were at 415 nm, 650 nm and 720 nm, the 650 nm
was chosen to be the optimal wavelength to prove the feasibility of our CMOS photodiode chip. On the other hand, the absorption peak of histamine oxidation reaction was only at 430 nm. Therefore, we follow the normal biochemical rule:
Chose the maximum absorption peak of reaction. As matter of fact, the only absorption peak of histamine oxidation reaction was at 430 nm based on the results of our experiment. Although 430nm might not the suitable range of the absorption peak of CMOS photodiode chip, we still wondered whether our CMOS photodiode chip could determinate histamine at 430 nm. Finally, we choose 650 nm and 430 nm to be the optimal absorptive wavelength of glucose and histamine oxidation reactions, respectively.
The performance of assembled biomedical system
Enzyme kinetic analysis results of glucose oxidase and diamine oxidase observed in our assembled biosensor system approached to that obtained from UV/Vis spectrophotometer. Our system was more superior than the traditional optical spectrophotometer whether volume or cost. In the meanwhile, the detection linear ranges of β-D(+)-glucose and histamine obtained from our assembled system approached to the amount in human bady. When we determine the β-D(+)-glucose and histamine. The liner detection ranges of our assembled biosensing system ranged from 1 μM to 5 mM and 10 ~ 80 μM, respectively, while the glucose concentration of human blood is 4 ~ 8 mM [30] and 1~8 mg/dl (90~720 μM) [31]. By the 2 to 10 folds sample dilution at most, the detection linear range of our assembled system can determine β-D(+)-glucose and histamine concentration in the human blood.
CHAPETR 5. CONCLUSION
We have set up an optical bio-sensing system for determination of glucose and histamine. A photodiode chip based on CMOS manufacture procedure was developed with the P+/Nwell finger structure and the subsequent current amplifier circuits, especially, include current mirrors and TIA. The photodiode with the P+/Nwell finger structure is surrounded by double guard rings to keep the substrate noise off. The optical transition theorem is used to be the detection method in our system. We designed colorimetrical reactions to be a biochemistry detection platform. Using the
We have set up an optical bio-sensing system for determination of glucose and histamine. A photodiode chip based on CMOS manufacture procedure was developed with the P+/Nwell finger structure and the subsequent current amplifier circuits, especially, include current mirrors and TIA. The photodiode with the P+/Nwell finger structure is surrounded by double guard rings to keep the substrate noise off. The optical transition theorem is used to be the detection method in our system. We designed colorimetrical reactions to be a biochemistry detection platform. Using the