Amperometric systems for the determination of oxidase enzyme dependent reactions by continuous flow and flow injection analysis

Introduction There are two primary types of flow analysis. One is continuous flow analysis (CFA). In Skegg’s classic CFA system the attainment of a steady state and air segmentation, which serves as a regulator to the flow [1] as well as a physical barrier to dispersion, are essential features 1,2,3 ]. The signals, commonly in the form of a steady state plateau, are usually dependent on the sample dispersion and mixing [31. A second type of flow analysis is flow injection analysis (FIA), described first by Bergmeyer and Hagen [4], and later by Ruzicka and Hansen 5 and Stewart and co-workers [6 ]. In this system the sample is introduced as a plug into a flowing stream via a valve or syringe, and mixing is accomplished by diffuon [2]. In contrast to CFA, the signal in FIA does not reach a steady state plateau, but gives sharp peaks [21. FIA has the advantage of a sampling rate commonly over 120 per hour, and as high as 300 per hour [7] or more [8,9]. The several applications of flow injection analysis have been reviewed by Betteridge [2]. While optical detection methods have been very successful in continuous flow and flow in3ection analysis, many electrochemical techniques have also been used for detection [2,10]. Voltammetric measurements with tubular platinum electrodes have been used in flow systems 11]. Procedures have been described for the determinaton of lactate [12], triglycerides [13], and ethanol [14] using open tubular carbon electrodes by biamperometric monitoring of hexacyanoferrate [11] ion. Glucose and glucose oxidase


Introduction
There are two primary types of flow analysis. One is continuous flow analysis (CFA). In Skegg's classic CFA system the attainment of a steady state and air segmentation, which serves as a regulator to the flow [1] as well as a physical barrier to dispersion, are essential features 1,2,3 ].
The signals, commonly in the form of a steady state plateau, are usually dependent on the sample dispersion and mixing [31.
A second type of flow analysis is flow injection analysis (FIA), described first by Bergmeyer and Hagen [4], and later by Ruzicka and Hansen 5 and Stewart and co-workers [6 ].
In this system the sample is introduced as a plug into a flowing stream via a valve or syringe, and mixing is accomplished by diffuon [2]. In contrast to CFA, the signal in FIA does not reach a steady state plateau, but gives sharp peaks [21. FIA has the advantage of a sampling rate commonly over 120 per hour, and as high as 300 per hour [7] or more [8,9]. The several applications of flow injection analysis have been reviewed by Betteridge [2].
While optical detection methods have been very successful in continuous flow and flow in3ection analysis, many electrochemical techniques have also been used for detection [2,10]. Voltammetric measurements with tubular platinum electrodes have been used in flow systems 11]. Procedures have been described for the determinaton of lactate [12], triglycerides [13], and ethanol [14] using open tubular carbon electrodes by biamperometric monitoring of hexacyanoferrate [11] ion. Glucose and glucose oxidase wires. [15,16] have been determined in serum, plasma, and whole blood by reacting with the H2 02 produced in the enzymatic reaction with iodide. The iodine produced is measured biamperometrically in static or stirred solutions.
In the present study several electrode systems were incorporated into a flow system to amperometrically or biamperometrically detect iodine in the presence of excess Flow-cell and electrodes: i) Two platinum wires approximately 0.002 in in diameter were inserted through the walls of a 0.04 in i.d. tygon tube and bent so that they protruded about 0.12 in down the centre of the tube. The two wires were separated by 0.31 in ( Figure 1). Leakage through the walls of the tubing through which the wires were inserted did not occur. ii) A single platinum wire as above was used with a saturated calomel electrode (SCE). The SCE reference was placed in a beaker into which the sample stream is fed directly.
iii) A platinum wire was cathodised for 30 minutes at 2mA in a AuClsolution containing KCN. The gold plated wire was then inserted into the flowing stream as above with an SCE reference. iv) A platinum wire working electrode was inserted into the tubing as above. A silver tube served as the counter electrode ( Figure 2).
Voltage source: A Princeton Applied Research Polarographic Analyser (PAR 174) was used to provide a constant potential between the electrodes and to monitor the steady state current output. A Moseley X-Y recorder was used to record the signals and insure current levelling between standards and samples.
Enzyme column: Alcohol oxidase was immobilised by covalent attachment via glutaraldehyde on silanised glass beads and also on the inside walls of nylon tubing as previously  Ethanol-CFA twin platinum electrodes The flow system design for the determination of ethanol is shown in Figure 3.   Alternatively, a sample loop bypass injector was constructed to allow sample injection as a discrete plug of reproducible volume into the reagent stream. This injector and flow system is shown in Figure 4 together with the tube dimensions. The injected volume was 30 pl. The position of the stopcock had no effect on the flowing conditions at the detection cell since reagent was able to flow freely through the bypass tube when the main tube was closed. Using this injector, Hz0: standards wer,e injected directly into the reagent stream and mixing was achieved solely by diffusion of the sample plug.
The third method of injection was by syringe through a septum directly into the reagent stream. This method suffered contamination of the septum port, and gave the lowest precision of the three methods.
The iodine produced by the H202 by reaction 2 was measured in the excess iodide at an applied potential of H Oz CFA platinum wire versus silver/silver chloride Hz 02 standards in pH 5.0 acetate buffer were injected with the stopcock valve into a reagent stream containing 0.1 M acetate, pH 5.0, 0.2 M KI, 10 -a M heptamolybdate, and 10-aM KI. The H2 02 oxidised the iodide to iodine, and the iodine was then reduced at the platinum electrode. The potential of the platinum electrode was +100 mV versus the silver/silver chloride electrode.

Results and discussion
Measurement of iodine H20 is a product of oxidase enzyme catalysed reactions, and indeed, the direct measurement of H:02 has been studied and used to determine glucose and other substances [19]. However, at a platinum electrode a potential of at least +0.35 V versus SCE (depending on the pH) is required for the anodic decomposition of the H2 02. The 12-1-couple is electroactively reversible, and at twin platinum electrodes poised at a small potential, the I-is oxidised at the anode while the 12 is reduced at the cathode.
A low, 200 mV, potential poised between twin platinum electrodes gives a low background current and a high sensitivity to 12 in the presence of excess I-Linearity was obtained for the measurement of I2 in unbuffered aqueous solution over 2 orders of magnitude. There is some upward drift in the current as the standards are successively measured. For instance, the net signals for 10 -s M Iz in triplicate analysis are 0.18, 0.21, and 0.23 #A, respectively, when the set of I2 standards between 10-6 and 10 -4 M is run three times in sequence. This problem is discussed in more detail below.
The pH had little effect on the measurement of I2 by our system. However, the blank current was affected, rising smoothly from 25 nA at pH to 50 nA at pH 4 and falling back to 20 nA by pH 7.
Measurement of 1"12 02 The first series of measurements of H2 02 was carried out at pH 7.8 because many enzymic reactions have their optimum range around this pH. H2 02 concentrations down to 4 x 10 -6 M were measured, and linearity was found over 1/2 orders of magnitude( Figure 5, curve A). The accuracy below 10 -s M was unsatisfactory using this pH.
Since the reaction 2 consumes H + it was expected that the reaction would favour a low pH. Indeed the rate of reaction 2 is seen to fall off rapidly above pH 5.5 ( Figure  6) and to have its maximum value at pH 5.0. Therefore, another calibration curve for determining H202 was con-structed with all standards and the reagent buffered at pH 5.0 ( Figure 5, curve B). The detection limit for H2 0 was lowered by 2 orders of magnitude over the same determination at pH 7.8. The data are linear over the range of concentrations measured (7 x 10 "8 to 7 x 10 "-s M). Application to ethanol analysis An immobilised alcohol oxidase column had been previously used to determine blood ethanol in a flow system by monitoring the oxygen concentration of the effluent from the enzyme column with an oxygen electrode in a flow cell [17]. The same column was used here to analyse aqueous ethanol standards (Figure 8). The biamperometric detection system showed almost an order of magnitude improvement in sensitivity over the former oxygen monitoring system. While the enzyme column used here had much lower activity than previous enzyme columns prepared, (approximately a factor of 5 lower than a freshly prepared column [20]), a Since the pH optimum of immobilised alcohol oxidase is 8.2 [21 and the pH optimum for reaction 2 is 5.0, the flow system arrangement in Figure 3 was used. The effluent from the enzyme column was a 0.05 M, pH 8.2 buffer. This was mixed with a 0.2 M, pH 4.7 buffer in approximately a 2 to ratio (determined by the relative flow rates in the two inlet tubes) so that the indicating reaction (reaction 2) occurred at pH 5.0.

Reproducibility
Poor reproducibility of the steady state biamperometric signals was obtained in the continuous flow measurement of 12 in the presence of excess I-. As mentioned previously, a 10-s M 12 solution gave signals of 0.18, 0.21, and 0.23 pA when measured at three different times in sequence with other 12 concentrations. On the other hand, the same solution gave net signals of 0.19, 0.19, 0.18, 0.19 and 0.18pA (relative standard deviation of 1.6%) when the standard was alternated with the blank. These data suggest iodine poisoning of the electrodes. Iodide has been reported to adsorb onto platinum [18,22]. With this adsorption as a possible contributor to the irreproducibility of the 12 measurement and H2 02 determination when coupled to I2-I-,it was decided to employ flow injection analysis of H202 standards was used to determine if the transient presence of I2 would alleviate some of the adsorption problems. Figure 9 shows the data obtained when H202 standards from 3.5 x 10-7 to 3.5 x 10-4 M were injected, and the resulting I2 was measured biamperometrically. The average relative deviation of the data is 15%. The data fit a straight line with a correlation coefficient of 0.991. However, the signals did not vary randomly around the best fit line. There was a definite increasing trend in the signals from a particular H202 concentration when the samples were run singularly from lower to higher concentrations and then the process repeated one or two times. The FIA signals obtained in constructing a calibration curve, similar to that shown in Figure 9, are shown in Figure 10. The problem described above is evident.
A gold plated platinum wire was used as the cathode in an attempt to determine H2 02 by measuring 12 amperometrically versus an SCE. The same problem as when a platinum cathode was used was encountered.
If iodide adsorption was the true cause of the irreproducibility of the amperometric signals when iodine was reduced at a platinum or gold cathode, a high concentration of chloride (0.2 M) in the reagent buffer containing less iodide (10-3 M rather than 10-M) might alleviate the problem.
With a tubular silver electrode at-100mV relative to a platinum wire electrode and a reagent containing the chloride and the iodine a very low (20 nA) background current was obtained. Standards of H2 02 oxidised iodide to iodine which was then reduced at the platinum electrode. A 6 x 10-6 M H2 02 solution resulted in a biamperometric current 5 nA above the 20 nA background current ( Figure 11). The low (1 x 10-3 M) concentration of I-appears to prohibit the complete reduction of H2 02 by I-in the 22 second reaction time allowed. Thus the response of the detection system is non-linear above 2 x 10-s M H2 02 The sensitivity of this detection system was a factor of 10 worse than that of the twin platinum electrodes but the reproducibility proved to be superior to any of the other I2-1-detection systems utilised, with an average relative deviation of 6%. However, it must be noted that air bubbles trapped at either electrode causes spurious readings.

Conclusion
It has been demonstrated that the electrochemical detection of iodine in the presence of excess iodide can be reliable and sensitive, when the appropriate solution environment is used with the appropriate working and counter electrodes. While, certain electrode systems have a very low detection limit for iodine, they can suffer poor reproducibility due to absorption onto or possibly oxidation, of the electrode surfaces. One electrode system has been found to offer good sensitivity and repeatability.
Hz02, a product of oxidase enzyme reactions, can be sensitively coupled to the reversible Iz-I-redox pair, offering a method for determining certain substrates of interest.  [6], Keller-Richter [7], and Bartscher [8] incremental methods have been discussed, there is little information in the literature on practical comparisons. The automated titrator described here has enabled hundreds *Correspondence to this author 194 of unbiased titration results to be printed out rapidly for the various incremental techniques. Results are presented and discussed for a weak acid-strong base and a precipitation titration. The incremental methods are compared on the basis of the experimental results obtained. Instrumentation A block diagram of the automated titrator is shown in Figure 1. An ADD-8080 microcomputer [9] provides the "intelligence" for the titrator. It is based on the 8080A microprocessor (Intel Corp., Santa Clara, Calif. 95051, USA). The microcomputer has 10K bytes of programmable read only memory (PROM) which contain a BASIC interpreter (a modification of BASIC/5, Processor Technology Corp., Emeryville, Calif. 94608, USA), a monitor program to facilitate machine language programming, and several utility programs. Also present are 15K bytes of read-write memory (RWM) which are used to store BASIC user programs in machine language and data. An arithmetic processing unit (AM9511 APU, Advanced Micro Devices, Inc., Sunnyvale, Calif. 94086, USA),is available to perform calculations that would be too time-consuming or cumbersome if done on the microprocessor. A conventional teletypewriter provides interaction with the operator.