Summary

The enzyme superoxide dismutase (SOD) catalyses the breakdown of superoxide anion (O₂^-) and provides the first line of defense against oxygen toxicity. The activity and the assay techniques are associated with diverse fields such as medicine, biochemistry, plant physiology and food chemistry. During the past few decades, various SOD assay methods have been developed. However, those methods have some drawbacks in the selectivity, rapidity, cost and convenience. Recently, we found that novel water-soluble tetrazolium salts such as XTT, WST-1 and WST-8 are suitable for the detection of O₂^- and are applicable to the SOD assay. Of these tetrazolium salts, WST-1 appears to be the most promising for the SOD assay due to its sensitivity, its low absorbance of the oxidized form and its water solubility. The SOD assay method based on the use of WST-1 can be applied to practical biochemical samples such as erythrocytes, liver and heart from rats. A novel flow injection assay system for SOD was also developed using WST-1. In this system, a rapid assay (sampling frequency=30 samples/hour) was achieved.

Key Words: Superoxide anion; tetrazolium salt; XTT; WST-1


1. Introduction

SOD is an enzyme that catalyzes dismutation of two superoxide anion (O₂^-) into hydrogen peroxide and molecular oxygen (Scheme 1).

2O₂^- + 2H⁺ -> H₂O₂ + O₂ (Scheme 1)

SOD is one of the most important enzymes in the front line of defense against oxidative stress. According to Dr. Cutler, it is also a factor that controls organisms' life-spans.¹⁾ Therefore, research on SOD activity will be important for the understanding of various mechanisms of life. Since the discovery of SOD by Dr. McCord and Dr. Fridovich in 1969, various SOD detection methods have been developed. However, there is no satisfactory method available from the point of selectivity, rapidity, simplicity and application range. In this article, I describe a new WST-1-based SOD assay developed in order to resolve the above-mentioned problems of conventional methods.

2. Reactive Oxygen Species and SOD

The determination of SOD activity is necessary for all research fields related with organism. First, I will describe involvement of reactive oxygen species (ROS) and SOD in research fields concerning human, plants and food, then consider the roles of SOD assay techniques performed in these research fields.

2.1 Humans

A given amount of the oxygen taken in by the body is always converted to O₂^-, H₂O₂, hydroxy radical (OH^.) and other molecules by various enzymatic metabolism systems. Among these molecular species, the life spans of OH^. and O₂^-, which has an unpaired electron, are the shortest. OH^. has the highest reactivity and it reacts with various molecules with diffusion controlled rate. Although it is believed that O₂^- may not directly react with lipids, proteins, sugars or nucleotides, it is transformed into OH^. when it interacts with metal ions (Fenton reaction) and it reacts with nitric oxide (NO) to quench the physiological activities such as vascular relaxation. At the same time, O₂^- generates peroxinitrite (ONOO^.), which causes oxidative damage. In order to protect the body from highly toxic ROS, the body has acquired anti-oxidative stress mechanisms including SOD. These anti-oxidative stress mechanisms are localized in tissues and inside the cells where ROS are generated. If the amount of ROS exceeds the limit of the defense mechanism of the body for any reason, serious disease may be induced. The typical example is cancer and life-style related diseases such as arteriosclerosis are also included. Researchers are currently exploring the role of oxidatively aggregated deposition of Beta-amyloid to senile plaques in causing Alzheimer disease and other vascular-damage-related brain diseases. Furthermore, cell damage by ROS is considered to be one of the main causes of various aging-related diseases. The use of radical trapping agents as a possible cure for these diseases is also being studied.
It is speculated that the level of O₂^- scavenging activity by SOD is related to the appearance of these diseases. So far, there are several examples that clearly indicate the correlation between SOD activity and human diseases such as Werner syndrome, amyotrophic lateral sclerosis (ALS, decrease in SOD activity), and Down syndrome (increase in SOD activity). Furthermore, since SOD activity is dramatically decreased by in vivo Maillard reaction, the SOD activity level of diabetes patients tends to be low. This decrease in SOD activity also decreases the diabetes patient's defense capability against oxidative stress, and various diseases may result as diabetes complications. It is expected that the correlation between these various diseases and SOD activity will continue to be revealed. Additionally, SOD activity determination will be utilized not only for the research of the mechanisms that cause diseases but also for the diagnosis and indication of health conditions.

2.2 Plants

Since plants cannot move around freely, they have highly sophisticated defense mechanisms toward environmental alterations. For example, the SOD activity of a plant is increased by the use of herbicides such as paraquat, by an increase in SO2 concentration in the atmosphere, by drought, or by exposure to high concentration of zinc and magnesium. These phenomena cause ROS to be generated in the plant. Findings also suggest that the reduction in the toxicity of O₂^- is a very important defense system against oxidative stress. Therefore, the role of SOD assay techniques is very important in the study of plant physiology.

2.3 Food

It is thought that the risk of ROS-related diseases is decreased by reinforcement of the defense mechanism against oxidative stress. Many people are interested in the anti-oxidation qualities of plants and their products such as red wine, tea and so on. There is an element in these plants that induces O₂^- quenching activities (also called SOSA (superoxide anion scavenging activity)), which are similar to those of SOD. Currently, the O₂^- scavenging activities of animal proteins are also being studied. It is expected that the development of food materials that have high SOD activity will continue to advance.

3. Conventional SOD Assays

Since SOD activity detection methods are used in many research fields, samples to be analyzed vary widely. Therefore, the most important technical aspect of SOD assay is the applicability of the assay to different samples. In other words, high selectivity with little interference from other components in a sample solution is desired. For the production of O₂^-, which is a substrate of SOD, a xanthine-xanthine oxidase reaction is utilized. A probe for the detection of O₂^- is included in the reaction solution. The change of the probe without any sample is indicated as a blank control, and the suppression ratio of the change of the probe by the sample solution is indicated as the inhibition ratio.
Generally, 50% inhibition by the sample solution is used for the activity determination (IC₅₀). On the other hand, O₂^- generated by the xanthine-xanthine oxidase reaction is spontaneously transformed to oxygen and hydrogen peroxide. This spontaneous dismutation reaction occurs rapidly in acidic conditions, and at the rate of 8.5x10⁵ - 8.5x10⁴ M^-1s^-1 at physiological pH (pH 7 - 8). Therefore, the second-order rate constant of the reaction between O₂^- and the probe should exceed the rate constant of the dismutation reaction. In the case where the rate constants are almost the same, the concentration of the probe should be increased. There are several different types of probes: those which change their color (colorimetric probes) when they react with O₂^-, those which emit light (chemiluminescence probes) when they react with O₂^-, and those which produce specific radicals (spin trap agents) when they react with O₂^-.

3.1 Spectrophotometric Detection

The most typical SOD detection method is the one based on spectrophotometric detection. This method uses either cytochrome C or nitroblue tetrazolium (NBT) (Fig. 1). The detection of O₂^- by the cytochrome C reducing method is based on the color change to generate purple color dye from reduced cytochrome C (scheme 2). This is the most commonly utilized method since the discovery of SOD.

Cyt(Fe^III) + O₂^- -> Cyt(Fe^II) + O₂ (Scheme 2)

Since cytochrome C is easily reduced by reductases such as NADPH reductase and other reducing agents, it is necessary to consider contaminants in the samples. Moreover, this method requires continuous monitoring in 1.5-minute intervals, so it is not suitable for high-throughput detection. The NBT method is based on the generation of water-insoluble blue formazan dye (lmax: 560 nm) by a reaction with O₂^-. Because the dye is not water-soluble, a non-homogeneous suspension is created during long-term analysis that causes problems in the reproducibility of the data. In order to solubilize formazan, alternative methods such as addition of BSA have been developed. However, the addition of unnecessary proteins may make the data analyses much more complicated. Moreover, NBT is reduced by various reducing agents. This characteristic of NBT is utilized for the detection of keto-amine, which is a marker of diabetes and is the intermediate in the Maillard reaction. The most significant disadvantage of the NBT method is that 100% inhibition cannot be achieved even with the addition of excessive amount of SOD into assay solution. The direct interaction between NBT and xanthine oxidase is speculated to be the cause.

3.2 Chemiluminescence Method

The chemiluminescence probe used for O₂^- detection can also be applied for SOD assay detection. There are two types of these probes. One is a lucigenin, and the other is a luciferin derivative (MCLA). These chemiluminescence reactions are highly pH dependent. For example, lucigenin shows extremely intense chemiluminescence at pH 9.0 and higher.¹⁵⁾ Therefore, SOD detection by chemiluminescence under physiological pH conditions is not feasible. On the other hand, MCLA emits strong luminescence under physiological conditions, and is therefore used for the detection of Cu, Zn-SOD activity in the human brain. However, MCLA is not a suitable SOD assay probe since it reacts not only with O^2- but also with singlet oxygen. Moreover, MCLA reacts with dissolved oxygen to emit background luminescence, and the transitional metal ions accelerate the oxidation reaction.

3.3 Electron Spin Resonance Spectroscopy (ESR) Method

At room temperature, the ESR signal of O₂^- in solution cannot be detected directly, but can be indirectly detected by a spin trap method. The most common trapping agent is 5,5-dimethyl-1-pyrroline N-oxide (DMPO). Since O₂^- trapping DMPO indicates a particular ESR spectrum pattern, ESR detection is the most specific method for O₂^- detection. However, the second-order rate constant between DMPO and O₂^- is relatively lower than the reaction constant of the spontaneous reaction of O₂^-. Therefore, a large amount of DMPO should be added to the solution (e.g., using a final concentration of 0.45 M). Unfortunately, this large volume of DMPO increases the cost per assay. Another problem with this method is the requirement of a relatively expensive ESR instrument.

4. New SOD Assay by Use of Water-soluble Tetrazolium Salts

In order to overcome above-mentioned problems, the following aspects need to be considered for the development of SOD assay: an economical method using a simple instrument, a less pH-sensitive method, and a highly O₂^--specific method. If a general spectrophotometer is applied as a simple instrument, colorimetric probes similar to cytochrome C and NBT are preferable. The most important point for the assay is to be able to determine 100% inhibition by SOD without interference from other components. We examined new tetrazolium salts that generate water-soluble formazan by reduction in order to establish a new SOD assay.

4.1 XTT

XTT is a water-soluble tetrazolium salt first reported in 1988. Since then, it has been utilized as a substrate for the electron transfer system of bacteria cells or mammalian cells. Its structure is indicated in Fig.1. While NBT has a bis-tetrazolium structure, the XTT structure is monotetrazolium and it has two sulfonic acid groups.



Fig. 1: Structures of tetrazolium salts used in SOD assay.

We used XTT for the SOD assay. As indicated in Fig. 2, 100% inhibition was observed in proportion to the increase in the SOD concentration. This 100% inhibition could not be observed by NBT method before. At an optimized condition, 100% inhibition was still observed at various pH. The observation of 100% inhibition by XTT means that XTT is specifically reduced by O₂^-, and XTT overcomes the problem associated with NBT.



Fig. 2: SOD inhibition curves using NBT (black circle) and XTT (white circle).
The reaction mixture contained 2.5 ml of 50 mM carbonate buffer (B, pH 9.4; C, 10.2) or 50 mM phosphate buffer (A, pH 8.0) and 0.1 ml each of 3 mM EDTA, 3 mM xanthine, 56.1 mU/ml xanthine oxidase, 0.75 mM XTT or NBT, and sample solution containing SOD at the concentration shown at abscissa. In the NBT method, 0.1 ml of BSA was also added.

It is also reported that NBT directly interacts with glucose oxidase except for xanthine oxidase.²⁰⁾ We added NBT or XTT in a solution of glucose oxidation reaction with glucose oxidase, and measured the production of its formazan (Fig. 3). Even in the glucose-glucose oxidase reaction, which does not generate O₂^-, NBT was reduced to generate its formazan in a time dependent manner, but no O.D. increase was observed in the case of XTT. Initial O.D. increased with an increase in the XTT concentration. This is due to the background O.D. from XTT tetrazolium. From these results, XTT does not seem to have any direct interaction with the reduced form of some enzymes that are generated during the oxidase reaction process, and XTT was shown to be a suitable probe for SOD assay.



Fig. 3: Time course of the reduction of NBT and XTT during the oxidation of glucose by glucose oxidase.
The assay mixture (2.8 ml) contained the following components as the final concentration: 50 mM glycine-NaOH (pH 9.5), 0.1 mM glucose, and 0.1 mM NBT (black circle), 0.1 mM XTT (triangle) or 0.2 mM XTT (white circle). The reaction was initiated by the addition of glucose oxidase solution. The absorbance at 470 (XTT) or 560 nm (NBT) was monitored at 25^oC.

In order to test the applicability of XTT for practical samples, we determined SOD activity of rabbit erythrocytes and compared with the data obtained by NBT method (Fig. 4: the assay was carried out at pH 10.2).



Fig. 4: Relationship between SOD activity in rabbit erythrocytes obtained by XTT and NBT methods.

A crude extract of SOD from blood erythrocytes was separated according to the conventional method.²¹⁾ The XTT method data showed a high correlation with the NBT method data with a correlation coefficient of 0.954. The value obtained by XTT method is almost double the NBT value, reflecting the difference in the sensitivity of these assays. Next, we applied this method to SOSA detection of food samples.²²⁾ As we know, food samples are very complicated multi-component mixtures. Samples such as red wine, green tea, coffee and cocoa were chosen to determine SOSA by XTT because these samples are already known to have SOSA. As expected, the color from food samples with no dilution interfered with the assay, and the materials in the samples reduced XTT directly. To achieve 10% or less absorbance change by these materials, tea, red wine, instant coffee or cocoa need to be diluted 100 times, 10 times, 50 times or 10 times, respectively at pH 8. Each diluted sample solution had over 50% inhibition activity at each dilution rate, thus it was shown that XTT assay was useful for the determination of SOSA in food samples. These results were consistent with the data obtained by ESR method. From these results, it was shown that XTT method is applicable to the use with biological samples and food samples.

XTT overcomes the problems associated with NBT and it seems that XTT is an ideal reagent for SOD assay. However, we encountered several new problems during the experiment using XTT. One of the problems is the pH dependent sensitivity change (Fig. 2). The data obtained by NBT was stable in the range between pH 8 and 10.2, however the sensitivity in the case of XTT decreased as the pH was lowered. Therefore, the sensitivity of XTT at pH 8.0 was lower than that of NBT. Another problem is the water-solubility of XTT. The water-solubility of XTT is around 2 mM and a heating process was necessary to prepare an optimal XTT solution (0.75 mM). Moreover, the high background O.D indicated in Fig. 3 is another concerning issue. Thus, we tried to further improve the assay by using different types of tetrazolium salts.

4.2 WST-1 and WST-8

From 1993 to 1998, Dr. Ishiyama and his group developed several types of water-soluble tetrazolium salts. Since the water-solubilities of these compounds are from several 10 mM to several 100 mM, we applied these water-soluble tetrazolium salts for SOD assay in place of XTT. For this assay, we used WST-1 and WST-8 (the chemical structures shown in Fig. 1). Both WST-1 and WST-8 are mono-tetrazolium salts that have sulfonate group(s) in their structure. At first, we optimized assay conditions and prepared inhibition curves by using standard SOD at pH 8.0, 9.4 and 10.2 (Fig. 5).



Fig. 5: SOD inhibition curves using the WST-1 (black circle) and WST-8 (white circle) systems.
The reaction mixture contained 2.5 ml of a phosphate buffer (A, pH 8.0) or 50 mM carbonate buffer (B, pH 9.4; C, pH 10.2) and 0.1 ml of 3 mM EDTA, 3 mM xanthine, 58 mU/ml of xanthine oxidase, 0.75 mM WST and the sample solution containing SOD at the concentration shown on the abscissa.

Both WST-1 and WST-8 showed 100% inhibition as seen in XTT assay at high concentrations of SOD. The most remarkable point was to be able to obtain the almost same IC₅₀ in different pH solutions. This result indicated that WSTs can overcome several shortcomings associated with XTT and NBT, making them ideal reagents for SOD assay. Since 100% inhibition was observed, we used WST-1 and WST-8 to determine whether the formazan is produced by the glucose-glucose oxidase reaction as in the XTT experiment (Fig. 6).



Fig. 6: Time course of the reduction of NBT and WST during the oxidation of glucose by glucose oxidase.
The assay conditions were same as those in Fig. 3. The assay mixture contained 0.1 mM NBT (black square), 0.2 mM WST-1 (black circle) or 0.2 mM WST-8 (white circle). The absorbance at 438 (WST-1), 460 (WST-8) or 560 nm (NBT) was monitored at 25^oC.

As expected, no formazan production was observed by a glucose-glucose oxidase reaction. Moreover, no background increase was observed with the increase in the concentration of WST. Therefore, WSTs enable us to determine the SOD activity at lower background conditions. In order to compare the sensitivity of WST, we compared IC₅₀ data obtained using NBT assay and XTT assay with that of WST assay (Table 1).

Table 1: Comparison of IC50 obtained by varioustetrazolium salts at pH 10.2
tetrazolium	IC50(ug/ml)
WST-1	0.22
WST-8	0.75
XTT	0.26
NBT	0.55

These IC₅₀ values are determined at pH 10.2, and it was indicated that WST-1 could determine SOD activity with the highest sensitivity among several other tetrazolium salts. Since WST-1 seemed to show high sensitivity, we used it to determine SOD activity of rat erythrocytes and compared the data with the ones obtained by using XTT method. As a result, a good linear correlation between WST-1 method and XTT method was observed with the correlation coefficient of 0.968 (n=7).

Recently, Dr. Winterbourn and his group recognized the usefulness of WST-1 in SOD activity detection and developed a microplate assay for SOD activity detection. They used a microplate assay to determine the SOD activity of human erythrocytes and rat liver and heart homogenates, and reported that the result obtained by this assay was consistent with the data reported previously. Thus, it seems that WST-1 has a wide range of applicability for biological samples. Currently, we are investigating the applications for other samples such as plant tissues and food samples.
In order to establish an automatic analytical system based on WST-1 assay, flow injection analysis (FIA) method was utilized. A diagram of our FIA system is shown in Fig. 7.



Fig. 7: FIA manifold for SOD assay.
P, pump; IV, injection valve; D, detector

In this system, we used a xanthine oxidase-immobilized reactor to avoid the exogenous addition of the enzyme in each experiment, making this analytical system a rapid and economical method. After series of experiments, we immobilized catalase in the reactor at the same time and used hypoxanthine as a substrate in order to increase the stability of xanthine oxidase. Prior to the assay, we prepared hypoxanthine and WST-1 mixed solution and combined it with the sample solution at 1:9 volume ratio, then injected 20 ul of the mixture into the reactor. When no SOD is contained in the sample solution, a maximum amount of WST-1 formazan is produced and maximum peak height is observed. If the sample solution contains SOD, the maximum peak is decreased in proportion to the SOD activity (Fig.8).



Fig. 8: Typical response curve of SOD preparation obtained under the optimum conditions.
A, 0; B, 1; C, 2.5; D, 5; E, 10; F, 25; G, 50; H, 100; I, 250; J, 500 ug/ml

The SOD inhibition curve was prepared by the inhibition rate determined from the decreasing ratio of the peaks. From the inhibition curve, IC₅₀ was determined to be 2.7 ug/ml and the absolute value was 50 ng. This value is larger compared with the IC₅₀ value of 20 ng obtained by a batch method. However, this FIA method enables us to determine over 30 samples per hour, thus the detection speed is extremely fast. We determined the SOD activity of rat erythrocytes using this method, and compared it with the data obtained by NBT method (Fig. 9).



Fig. 9: Relationship between SOD activity in rabbit erythrocytes obtained by NBT method and WST-1-FIA method.

There was a very good correlation in the data between the FIA method and NBT method. Even though this assay is a rapid assay system, the results obtained by this method were sufficiently consistent with the data obtained by conventional methods.

5. Conclusion

We described the importance of SOD assay and several detection methods. Since the discovery of SOD, various assay systems have been proposed every year. This fact suggests that currently there is no single satisfactory method available. The WST-1 based SOD assay we developed seems to be the most attractive method that overcomes many problems associated with conventional methods. In order to render this method as a standard method for SOD assay, we plan to investigate the specificity of WST-1 as an O₂^- probe.




Author:
Hiroyuki UKEDA, Ph.D.
Department of Agriculture
Khochi University
200 Mononobe Otsu
Nangoku-city, Khochi 783-8502
Japan

Click here for SOD Assay Kit-WST information

Click here for WST-1 information

References:
1) J.M. Tolmasoff, T. Ono and R.G. Cutler, Proc. Natl. Acad. Sci. USA, 77, 2777 (1980).
2) J.M. McCord and I. Fridovich, J. Biol. Chem., 244, 6049(1969).
3) T. Yoshikawa and T. Tanigawa, Kagaku to Seibutsu, 37, 475 (1999).
4) H.R. Warner, Free Radic. Biol. Med., 17, 249 (1994).
5) N. Taniguchi, Adv. Clin. Chem., 29, 1 (1992).
6) H. Ukeda, Y. Hasegawa, T. Ishii and M. Sawamura, Biosci. Biotech. Biochem., 61, 2039 (1997).
7) L.S. Monk, K.V. Fagerstedt and R.M.M. Crawford, Physiol. Plant., 76, 456 (1989).
8) F. Khanom, H. Kayahara and K. Tadasa, Biosci. Biotech. Biochem., 64, 837 (2000).
9) K. Suetsuna and H. Ukeda, Nissui-shi, 65, 1096 (1999).
10) K. Suetsuna, H. Ukeda and H. Ochi, J. Nutr. Biochem., 11, 128 (2000).
11) C. Beauchamp and I. Fridovich, Anal. Biochem., 44, 276 (1971).
12) R.N. Johnson, P.A. Metcalf and J.R. Baker, Clin. Chim. Acta, 127, 87 (1982).
13) A. Nishida, Y. Misaki, H. Kurata and S. Takashima, Brain Dev., 16, 40 (1994).
14) P. Corbisier, A. Houbion and J. Remacle, Anal. Biochem., 164, 240 (1980).
15) M.M.J. Oosthuizen, M.E. Engelbrecht, H. Lambrechts, D. Greyling and R.D. Levy, J. Biolumin.Chemilumin., 12, 277 (1997).
16) Y. Noda, K. Anzai, A. Mori, M. Kohno, M. Shinmei and L. Packer, Biochem. Mol. Biol. Int., 42, 35 (1997).
17) K.D. Paull, R.H. Shoemaker, M.R. Boyd, J.L. Parsons, P.A. Risbood, W.A. Barbera, M.N. Sharma, D.C. Baker, E. Hand, D.A. Scudiero, A. Monks, M.C. Alley and M. Grote, J. Heterocyclic Chem., 25, 911 (1988).
18) N.W. Roehm, G.H. Rodgers, S.M. Hatfield and A.L. Glasebrook, J. Immunol. Methods, 142, 275 (1991).
19) H. Ukeda, S. Maeda, T. Ishii and M. Sawamura, Anal. Biochem., 251, 206 (1997).
20) S.I. Liochev and I. Fridovich, Arch. Biochem. Biophys., 318, 408 (1995).
21) J. Maral, K. Puget and A.M. Michelson, Biochem. Biophys. Res. Commun., 77, 1525 (1977).
22) T. Shimamura, S. Maeda, H. Ukeda and M. Sawamura, Nou-ka, 72 , 1181 (1998).
23) M. Ishiyama, M. Shiga, K. Sasamoto, M. Mizoguchi and P. He, Chem. Pharm. Bull., 41, 118 (1993).
24) M. Ishiyama, Y. Miyazono, M. Shiga, K. Sasamoto, Y. Ohkura and K. Ueno, Anal. Sci., 12, 515 (1996).
25) M. Ishiyama, Y. Miyazono, K. Sasamoto, Y. Ohkura and K. Ueno, Talanta, 44, 1299 (1997).
26) H. Ukeda, D. Kawana, S. Maeda and M. Sawamura, Biosci. Biotech. Biochem., 63, 485 (1999).
27) A.V. Peskin and C.C. Winterbourn, Clin. Chim. Acta, 293, 157 (2000).
28) H. Ukeda, A.K. Sarker, D. Kawana and M. Sawamura, Anal. Sci., 15, 353 (1999).