Superoxide produced by activated neutrophils efficiently reduces the tetrazolium salt, WST-1 to produce a soluble formazan: a simple colorimetric assay for measuring respiratory burst activation and for screening anti-inflammatory agents

https://doi.org/10.1016/S0022-1759(00)00156-3Get rights and content

Abstract

Activation of the respiratory burst of granulocytes and macrophages by invading microorganisms is a key first line cellular defence against infection. Failure to generate this response leads to persistent life-threatening infection unless appropriate antibiotic treatment is given. The respiratory burst of neutrophils is usually measured spectrophotometrically by following ferricytochrome c reduction, and histologically by using the tetrazolium salt, nitroblue tetrazolium, which is reduced intracellularly to an insoluble formazan. In both assays, reduction is mediated by superoxide generated via NADPH oxidase. Because ferricytochrome c has a high molecular mass and high background absorbance at 550 nm, the assay lacks sensitivity and is not ideally suited to microplate measurement. We have circumvented these limitations by using the cell-impermeable, sulfonated tetrazolium salt, WST-1, which exhibits very low background absorbance and is efficiently reduced by superoxide to a stable water-soluble formazan with high molar absorptivity. This has permitted adaptation of the WST-1 assay to microplate format while retaining sensitivity. Reduction of WST-1 by activated human peripheral blood neutrophils correlated closely with ferricytochrome c reduction across a range of PMA concentrations and with time of activation by PMA and fMLP. Reduction of WST-1 was inhibited by 98% by superoxide dismutase (20 μg/ml) and by 88% by the NADPH oxidase inhibitor, diphenyleneiodinium (10 μM) but was resistant to catalase, azide and the NADH oxidase inhibitor, resiniferatoxin. WST-1 and ferricytochrome c reduction were also compared using xanthine/xanthine oxidase to generate superoxide. Under optimised assay conditions, both WST-1 and ferricytochrome c reduction were directly proportional to added xanthine. WST-1 generated approximately 2-fold greater increase in absorbance than ferricytochrome c at their respective wavelengths, and this translated into increased assay sensitivity. Addition of the intermediate electron acceptor, 1-methoxy phenazine methosulfate, increased the background of the neutrophil assay but did not affect the overall magnitude of the response. We have used the WST-1 assay to assess human neutrophil dysfunction and to compare anti-inflammatory activity.

Introduction

Activation of inflammatory cells including granulocytes and macrophages leads to chemotactic responses, phagocytosis and stimulation of the so-called ‘respiratory burst’ in which superoxide production is linked to the killing of invading microorganisms. Central to the respiratory burst is the assembly of functional membrane-associated NADPH oxidase that uses cytosolic NADPH as a substrate to transfer electrons across cell membranes to oxygen (Chanock et al., 1994). Highly reactive downstream products are generated from superoxide via hydrogen peroxide and myeloperoxidase and these are thought to be the main cytotoxic effectors of activated neutrophils both in intracellular phagosomes and extracellularly (Badwey and Karnovsky, 1980, Kettle and Winterbourn, 1997).

Superoxide production by activated neutrophils is usually quantified spectrophotometrically by measuring ferricytochrome c reduction (Babior et al., 1973). Although microplate adaptations of this assay have been described (Cerasoli et al., 1988, Chapman et al., 1991, Carey et al., 1995), these have not been widely applied because of poor signal/noise ratios which are limited by the relatively high background absorbance of ferricytochrome c. Sensitive chemiluminescent techniques have also been employed to measure superoxide production by the NADPH oxidase of phagocytic cells (Storch and Ferber, 1988, Li et al., 1998), but these methods often lack specificity for superoxide and can be confounded by redox cycling reactions which themselves generate superoxide (Fridovich, 1997). Additionally, specialised spin-trap reagents have been developed to study superoxide production by activated neutrophils (Kotake et al., 1994, Roubaud et al., 1997), but these approaches have not been widely applied because of the need for sophisticated equipment, the limited lifespan of the superoxide adducts and their limited specificity for superoxide.

The wide use of tetrazolium salts to measure cell proliferation and viability (Marshall et al., 1995) has been shown to depend on metabolic activity, in particular, cellular production of reduced pyridine nucleotides (Berridge and Tan, 1993, Berridge et al., 1996). These cofactors act as substrates for cellular oxidoreductases that reduce tetrazolium salts such as MTT, XTT and MTS and more recently, WST-1 to formazan end-products. Succinate and other potential metabolic substrates, which were at one time thought to be responsible for the cellular reduction of tetrazolium salts (Mosmann, 1983), appear to play only a minor role, most cellular reduction being associated with extra-mitochondrial membranes (Berridge and Tan, 1993). Non-proliferating cells such as granulocytes and macrophages can also be activated to a state of high metabolic activity by reagents that simulate infection, this being the basis of their histological staining by nitroblue tetrazolium (NBT) and other tetrazolium salts which produce insoluble formazan deposits intracellularly (Pruett and Loftis, 1990, Stoward and Pearse, 1991).

The recent observation that the membrane-impermeable tetrazolium salt, WST-1, used in conjunction with an intermediate electron acceptor, can be reduced to a soluble formazan extracellularly by plasma membrane NADH oxidase of proliferating cells, by a mechanism involving superoxide (Berridge et al., 1996, Berridge and Tan, 1998), raises the possibility that WST-1 may serve as an alternative to ferricytochrome c to measure rapid rises in metabolic activity such as those associated with the respiratory burst. In this context, MTT (Madesh and Balasubramanian, 1997), and XTT and MTS which produce soluble formazan end-products that can be monitored continuously (Sutherland and Learmonth, 1997, Ukeda et al., 1997), have been used to determine superoxide production by xanthine/xanthine oxidase, but these assays have not been applied to measuring the respiratory burst of viable phagocytic cells.

In this study we have used WST-1 (Ishiyama et al., 1993, Ishiyama et al., 1995a) to measure superoxide production by neutrophils activated by the phorbol ester, PMA and the chemotactic peptide, fMLP. We demonstrate that WST-1 has many advantages over ferricytochrome c for determining the respiratory burst in that it exhibits extremely low background and high sensitivity due to its low molecular mass and the high molar extinction coefficient of the reduced formazan. This allows the WST-1 assay to be readily adapted to microplate format. We show that WST-1 is reduced by NADPH oxidase via superoxide and describe the use of this assay for measuring neutrophil function and pro-inflammatory and anti-inflammatory responses. A preliminary report that WST-1 can be used to measure the respiratory burst of PMA-activated human peripheral blood neutrophils has been published recently (Berridge and Tan, 1998).

Section snippets

Reagents and chemicals

Polymorphprep™, density 1.113 g/ml, was obtained from Nycomed Pharma (AS, Oslo, Norway). MTT, superoxide dismutase (SOD) and the WST-1 cell proliferation reagent were from Boehringer Mannheim and the chemicals, WST-1 and 1-methoxy PMS from Dojindo Laboratories (Kumamoto, Japan). Phorbol 12-myristate 13-acetate (PMA), N-formyl-methionine-leucine-phenylalanine (fMLP), horse heart ferricytochrome c, catalase, azide, xanthine, xanthine oxidase, diphenyleneiodinium chloride (DPI), resiniferatoxin

Results

Initial investigation of the use of WST-1 for measuring superoxide produced by activated human peripheral blood neutrophils utilised a cell proliferation kit containing WST-1 and a proprietary intermediate electron acceptor. Fig. 1a compares the responsiveness of neutrophils to PMA using the WST-1 kit and ferricytochrome c at saturating concentrations. When background absorbances of 0.35 A450 and 0.53 A550, respectively, were subtracted, plateaus were observed at 100–200 ng/ml PMA which were

Discussion

We have developed a simple real-time colorimetric assay for quantifying neutrophil activation. This assay has many advantages over the ferricytochrome c assay presently in wide use for measuring the respiratory burst and associated inflammatory and anti-inflammatory responses. In this assay, the sulfonated tetrazolium salt, WST-1, is reduced extracellularly to its soluble formazan by superoxide, as shown by extensive inhibition by superoxide dismutase. Compared with ferricytochrome c, the WST-1

Acknowledgements

We thank Professor Christine Winterbourn for helpful comments on the manuscript. The mussel glycogen samples were provided by Dr George Slim, Industrial Research Ltd. This research was funded by the Health Research Council of New Zealand and the Cancer Society of New Zealand.

References (28)

  • J. Storch et al.

    Detergent-amplified chemiluminescence of lucigenin for determination of superoxide anion production by NADPH oxidase and xanthine oxidase

    Anal. Biochem.

    (1988)
  • H. Ukeda et al.

    Spectrophotometric assay for superoxide dismutase based on tetrazolium salt 3′-1-(phenylamino)-carbonyl-3, 4-tetrazolium]-bis(4-methoxy-6-nitro)benzenesulfonic acid hydrate reduction by xanthine-xanthine oxidase

    Anal. Biochem.

    (1997)
  • B.M. Babior et al.

    Biological defense mechanisms. The production by leukocytes of superoxide, a potential bactericidal agent

    J. Clin. Invest.

    (1973)
  • J.A. Badwey et al.

    Active oxygen species and the functions of phagocytic leukocytes

    Annu. Rev. Biochem.

    (1980)
  • Cited by (302)

    • Reactive oxygen species, the trident of Neptune in the hands of hecate; role in different diseases, signaling pathways, and detection methods

      2022, Archives of Biochemistry and Biophysics
      Citation Excerpt :

      So, these methods have to be able to compete with antioxidants and produce stable and quantitative products [10,11]. Classical detection methods such as spin trapping and ferricytochrome C react with the high level of free radicals [12], but some sensitive and site-specific detection methods have been developed to recognize the lower concentration of free radicals; among them we can mention probe-based techniques like Cyclic hydroxylamine [13], fluorescent protein-based redox [14], boronate fluorescent [15], and immunospin [16]. Most of these methods are based on chemical reagents that initiate and react with free radicals in extracellular, intracellular, and mitochondrial spaces.

    View all citing articles on Scopus
    View full text