Real-Time Monitoring of Reactive Oxygen and Nitrogen Species in a Multiwell Plate Using the Diagnostic Marker Products of Specific Probes

Abstract

Developing rigorous assays for cellular detection of reactive oxygen and nitrogen species (O₂^·−, H₂O₂, ^·NO, and ONOO⁻) is an active area of research in our laboratory. Published reports suggest that diagnostic marker products are formed as a result of interaction of these species with small molecular weight fluorescent and nonfluorescent probes. In this chapter, we describe an HPLC-based methodology to detect formation of these species in biological and cellular systems. By monitoring the diagnostic marker products formed from reaction between specific chemical probes and the oxidant species, it is possible to simultaneously assay these species using a multiwell plate (e.g., 384-well plate).

1. INTRODUCTION

The purpose of this article is to demonstrate the feasibility of using a diagnostic marker-based high-throughput methodology for detecting reactive oxygen and nitrogen species (ROS/RNS) [e.g., superoxide (O₂−), hydrogen peroxide (H₂O₂), nitric oxide (^·NO), peroxynitrite (ONOO⁻)] in a multiwell (96-well, 384-well) plate format. The main methodology to be used is HPLC or UHPLC with absorption and fluorescence detection or in combination with mass spectrometry. The chemical probes to be used are hydroethidine (HE) for O₂^·−, coumarin boronate (CBA) for H₂O₂ and ONOO⁻, diaminofluorescein (DAF-2) for ^·NO, and ortho-triphenylphosphonium-substituted phenylboronate [o-Mito-PhB(OH)₂] for specific detection of ONOO⁻. The structures of the probes and diagnostic marker products derived from them as a result of interaction with specific ROS/RNS are shown in Fig. 9.1. Previously, we described a similar experimental design in which the initial screening of these species was performed using fluorescent probes (HE, CBA, DAF-2, and Amplex Red) in a 96-well plate by real-time monitoring increase in fluorescence intensity using a plate reader (Zielonka, Sikora, Hardy, Joseph, Dranka, & Kalyanaraman, 2012; Zielonka, Zielonka et al., 2012). We recommended the use of HPLC-based analyses to further confirm and rigorously identify the specific products formed from the fluorogenic probes.

Figure 9.1.

Structures of chemical probes and their diagnostic marker products formed from reaction between (A) HE and superoxide or one-electron oxidants, (B) DAF-2 and nitric oxide and oxidant, and (C) CBA/o-Mito-PhB(OH)₂ and peroxy anion or peroxynitrite.

Recently, significant advances have been made with regard to reaction mechanisms (rate constants, stoichiometry, and product characterization) of O₂^·−, H₂O₂, ONOO⁻, and^·NO with fluorogenic dyes that have infused increased confidence, rigor, and reliability in ROS/RNS assays (Michalski, Zielonka, Hardy, Joseph, & Kalyanaraman, 2013; Robinson et al., 2006; Sikora et al., 2011; Sikora, Zielonka, Lopez, Joseph, & Kalyanaraman, 2009; Zhao et al., 2005, 2003; Zielonka, Sikora, et al., 2012; Zielonka, Vasquez-Vivar, & Kalyanaraman, 2008; Zielonka, Zielonka, et al., 2012). For example, the long-held notion that O₂^·− reacts with HE to form ethidium (E⁺) has been debunked; instead, the only major product of the reaction between HE and O₂^·− was identified as 2-hydroxyethidium (2-OH-E⁺) (Zhao et al., 2003, 2005). Other oxidants (ONOO⁻-derived oxidants including OH, and higher oxidants derived from peroxidases) react with HE to form other oxidation products (E⁺, and dimeric products), all of which can be readily separated and identified by HPLC and mass spectrometry (Zhao et al., 2005; Zielonka, Srinivasan, et al., 2008; Zielonka, Vasquez-Vivar, et al., 2008). These studies have exposed and highlighted the pitfalls and limitations of fluorescence microscopy in assessing intracellular and mitochondrial O₂^·− (Kalyanaraman, Dranka, Hardy, Michalski, & Zielonka, in press; Zielonka & Kalyanaraman, 2010). The discovery of a rapid, direct, and nearly stoichiometric reaction between ONOO⁻ and boronates forming specific products has enabled the differentiation between ONOO⁻ and myeloperoxidase/H₂O₂/NO ⁻-based nitration mechanisms (Sikora et al., 2011, 2009; Zielonka, Sikora, et al., 2012; Zielonka, Zielonka, et al., 2012). Kinetic analyses of boronate reactions with H₂O₂ and ONOO⁻ have made it possible to clearly and distinctly interpret the data in enzymatic and cellular systems cogenerating these species (Sikora et al., 2011, 2009; Zielonka, Sikora, et al., 2012; Zielonka, Sikora, Joseph, & Kalyanaraman, 2010; Zielonka, Zielonka, et al., 2012).

Whereas the initial screening in a fluorescence plate reader followed by a rigorous HPLC characterization may be a useful approach in a high-throughput screening of several thousands of compounds from chemical libraries, the HPLC/MS detection and identification of diagnostic marker products of ROS/RNS-specific probes are most critical for elucidating the chemical, biological, and signaling pathways in cells. In this chapter, we describe the use of four different chemical probes specific for monitoring O₂^·−, H₂O₂, ^·NO, and ONOO⁻ formation in cell-free and cell systems.

2. EXPERIMENTAL METHODS

2.1. Assay plate layout

Prior to using the global profiling approach in cellular systems, the uninitiated investigators should become familiar with the assay using xanthine (X) or hypoxanthine (HX)/xanthine oxidase (XO) as a source of superoxide flux and the ^·NO donor, DPTA-NONOate, that decomposes to slowly release ^·NO (Fig. 9.2A). Defined fluxes of O₂^·−and ^·NO are simultaneously generated from X (or HX)/XO and DPTA-NONOate in a 96-well plate (Fig. 9.2B). The proposed experimental approach for HPLC (or UHPLC)-based monitoring of ROS/RNS using the four different probes is shown in Fig. 9.3. HPLC conditions for fast analyses of the probes and their products have been described in earlier publications (Sikora et al., in press; Zielonka, Sikora, et al., 2012; Zielonka, Zielonka, et al., 2012).

Figure 9.2.

(A) Peroxynitrite generation from cogenerated O₂^·− and ^·NO. Schemes showing formation of O₂^·− from incubations containing hypoxanthine (100 μM) and xanthine oxidase (0–0.4 mU/ml) in a phosphate buffer (pH 7.4; 50 mM) containing dtpa (100 μM), ^·NO generation from a thermal decomposition of DPTA-NONOate (0–50 μM), and ONOO⁻ formed from cogenerated O₂^·−from HX/XO and ^·NO from DPTA-NONOate. (B) Diagrammatic representation of global profiling experiments. Varying fluxes of cogenerated ^·NO and O₂^·− in a 96-well plate. The top right corner well represents conditions generating maximal (1:1) amount of ^·NO and O₂^·− and bottom right corner well represents conditions generating maximal O₂^·− and H₂O₂ alone. (C) Reactive oxygen and nitrogen species formation from RAW 264.7 macrophages activated with LPS, IFN-γ, and PMA.

Figure 9.3.

Schematic representation of the proposed experimental approach for HPLC (or UHPLC)-based specific monitoring of superoxide (HE oxidation to 2-OH-E⁺), nitric oxide (nitrosation of DAF-2 to DAF-2T), and peroxynitrite (oxidation of CBA to COH and transformation of o-Mito-PhB(OH)₂ to o-Mito-PhOH and o-Mito-PhNO₂). The identity of the species detected is confirmed by using specific inhibitors and/or scavengers. Rapid separation and detection with UHPLC systems enable analysis of samples in 96- or 384-well plate format.

3. RESULTS AND DISCUSSION

3.1. HPLC analyses of HE-derived products

Figure 9.2B shows the “cell-free” assay setup in a 96-well plate containing HE in the presence of varying fluxes of ^·NO and O₂^·− generated from HX/XO and DPTA-NONOate. Representative HPLC chromatograms (absorption and fluorescence) of HE metabolites are shown in Fig. 9.4A. Under conditions in which only O₂^·− and H₂O₂ were generated (right bottom corner well in Fig. 9.2B), the HPLC profile corresponds mainly to 2-OH-E⁺, the diagnostic marker product of HE/O₂^·−reaction and very little E⁺ or dimers. Under conditions in which both O₂^·− and ^·NO are maximally produced at a 1:1 ratio (top right corner well in Fig. 9.2B), 2-OH-E⁺ was dramatically inhibited; instead, there was increased formation of E⁺ and dimers (HE-E⁺, E⁺-E⁺). Under other conditions, different levels of these products were formed. Indeed, a complete profile of 2-OH-E⁺ formation under different fluxes of ^·NO and O₂^·− has been reported (Zielonka, Zielonka, et al., 2012).

Figure 9.4.

Oxidation products formed from HE. (A) HPLC profiles of HE oxidation in cell- free incubations containing DPTA-NONOate alone, hypoxanthine and xanthine oxidase, and both DPTA-NONOate and hypoxanthine and xanthine oxidase, and (B) HPLC profiles obtained from cell lysates of RAW 264.7 macrophages stimulated to produce ^·NO, O₂^·−, and both ^·NO and O₂^·−.

From these results, we conclude the following: (1) 2-OH-E⁺, the specific marker product of HE reaction with O₂^·−, increased with increasing O₂^·− flux and ^·NO inhibited 2-OH-E⁺ formation in a concentration-dependent fashion. (2) With increasing ^·NO and O₂^·−, the levels of E⁺ and dimeric (e.g., E⁺-E⁺) products increased, suggesting the formation of one-electron oxidants (^·NO and O₂ and ONOO⁻-derived oxidants). (3) Although the HPLC data show a marked decrease in 2-OH-E⁺, the fluorescence intensity remained high due to increased levels ofE⁺. Clearly, the measurements of HE oxidation-derived fluorescence cannot be equated to O₂^·− formation. The mechanism of formation of 2-OH-E⁺ and E⁺ is very different.

Once the investigators become conversant with the HPLC analyses in the well-defined “cell-free” model system, they will be able to perform studies in cellular systems. To investigate oxidation products of HE by endogenously generated O₂^·−, ^·NO, and ONOO⁻, we used a system consisting of RAW 264.7 macrophages stimulated with a mixture of LPS, IFN-γ, and PMA (Fig. 9.2C). Oxidation of HE was monitored in a 96-well plate and products formed in cells and in media were analyzed by HPLC. Macrophages were seeded in 96-well plates, and PMA, LPS, or IFN-γ were added as indicated in the presence of HE. The corresponding HPLC profiles are shown in Fig. 9.4B. As shown, 2-OH-E⁺ formation increased in the presence of macrophages activated by PMA. Similar to the cell-free system, costimulation of ^·NO and O₂^·− production during the combined treatment with IFN-γ, LPS, and PMA decreased the yield of 2-OH-E⁺ with a concomitant increase in HE-derived dimeric products.

3.2. HPLC analyses of CBA-derived products

The “cell-free” assay setup is similar to conditions described in Fig. 9.2B but using coumarin boronic acid (CBA) probe (Fig. 9.1). Oxidation products of CBA were monitored under varying fluxes of ^·NO and O₂^·−. Representative HPLC chromatograms are shown in Fig. 9.5A. As can be seen, there is very little oxidation of CBA to 7-hydroxycoumarin (COH) in the presence ^·NO or O₂^·−alone. However, there is a dramatic increase in the formation of COH in the presence of both ^·NO and O₂^·−. This is consistent with the very rapid reaction between ONOO⁻ and CBA. CBA reacts with H₂O₂ stoichiometrically. However, because this reaction is rather slow, much higher concentrations (millimolar range) of CBA are needed to detect and quantitate H₂O₂. Thus, in the presence of ONOO⁻, it is essential to measure catalase-inhibitable amount of COH in order to assess H₂O₂ formation. The reaction between CBA and H₂O₂ to form COH is catalase sensitive, whereas the ONOO⁻/CBA reaction is catalase insensitive (Zielonka, Sikora, et al., 2012; Zielonka et al., 2010; Zielonka, Zielonka, et al., 2012). The oxidation of CBA by activated macrophages was monitored in a 96-well plate. Macrophages were stimulated with LPS, IFN-γ, and PMA, as described previously. Representative HPLC profiles are shown in Fig. 9.5B. Substantial oxidation of CBA to COH was observed in the presence of IFN-γ, LPS, and PMA. l-NAME significantly inhibited COH formation, as did SOD. These results suggest that CBA can be used to selectively detect ONOO− formed from macrophages activated to cogenerate ^·NO and O₂^·−.

Figure 9.5.

Oxidation products formed from CBA. (A) HPLC profiles of CBA oxidation in cell-free incubations containing DPTA-NONOate alone, hypoxanthine and xanthine oxidase, and both DPTA-NONOate and hypoxanthine and xanthine oxidase, and (B) HPLC profiles obtained from cell media of RAW 264.7 macrophages stimulated to produce ^·NO, O₂^·−, and both ^·NO and O₂⁻.

3.3. HPLC analyses of DAF-2-derived products

The formation of a triazole, DAF-2T, formed from nitrosation of DAF-2 has been used to monitor intracellular ^·NO formation (Thomas, Kotamraju, Zielonka, Harder, & Kalyanaraman, 2007). The “cell-free” assay setup is similar to experimental conditions described in Fig. 9.2B. Nitrosation of DAF-2 was monitored in a 96-well plate in the presence of cogenerated ^·NO and O₂^·−. ^·NO alone induces formation of DAF-2T in the presence of oxygen or superoxide. Oxidation of DAF-2 may “activate” the probe to react with ^·NO and thus increase the yield of DAF-2T. The representative HPLC traces are shown in Fig. 9.6. It is clear that ^·NO is required to produce DAF-2T, both in cell-free and cellular systems. However, other factors may also affect the yield of the product, as exemplified by the amplifying effect of superoxide on the yield of DAF-2T in cell-free system (Fig. 9.6A). As shown previously, bell-shaped response of DAF-2 nitrosation to different superoxide fluxes indicates the possibility of both enhancement and inhibition of DAF-2T formation, depending on the actual ratio of superoxide and nitric oxide fluxes. Under the conditions of excess of superoxide, virtually all ^·NO is scavenged and the yield of DAF-2T decreased. In this case, the decrease of DAF-2T formation may be therefore interpreted as a diminished bioavailability of “free” ^·NO, as exemplified in Fig. 9.6B.

Figure 9.6.

HPLC profiles of DAF-2-derived products. (A) Nitrosation of DAF-2 (5 μM) was monitored in a “cell-free” system. As shown, ^·NO is required to form the specific nitrosated product, DAF-2T. (B) Nitrosation of DAF-2 (5 μM) in media of activated RAW 264.7 macrophages.

3.4. HPLC analyses of ortho-Mito-PhB(OH)₂-derived products

The oxidative transformation of o-Mito-PhB(OH)₂ probe in the presence of peroxynitrite is shown in Fig. 9.7A. In contrast to other arylboronates, with o-Mito-PhB(OH)₂, the relative yield of the minor nitration product (o-Mito-PhNO₂) is independent of the presence of biologically relevant reductants (Sikora et al., in press). The ratio between o-Mito-PhNO₂ and o-Mito-PhOH during ONOO⁻-dependent oxidation of o-Mito-PhB (OH)₂ remained constant. The actual yields of both products can be used to determine the contribution of different oxidants (ONOO⁻ or H₂O₂) to oxidation of the probe. As o-Mito-PhNO₂ is not formed by H₂O₂ or HOCl or $\text{MPO}/{\text{H}}_2{\text{O}}_2/\text{N}{\text{O}}_{\text{2}}^{-}$ system, it can be used as a specific diagnostic marker of ONOO⁻ (Sikora et al., in press). As shown in Fig. 9.7B, HPLC separation of different products can be achieved within 5 min, and the analysis time is expected to be significantly shortened with the use of UHPLC systems.

Figure 9.7.

HPLC profiles of o-Mito-PhB(OH)₂-derived products. (A) Scheme showing the major (o-Mito-PhOH) and minor (o-Mito-PhNO₂) products of the reaction of o-Mito-PhB(OH)₂ with ONOO⁻. (B) HPLC chromatogram of the standards of o-Mito- PhB(OH)₂, o-Mito-PhOH and o-Mito-PhNO₂.

4. SUMMARY

The HPLC assays using ROS/RNS-specific chemical probes reported here with well-defined redox chemistry are simple yet rigorous and can readily be implemented in cell-based high-throughput studies (e.g., in 384- and 1536-well plates) using automated liquid-handling systems. Increased understanding of the free radical and non-free radical chemistry of fluorescent probes has exposed and highlighted the limitations of fluorescence microscopy-based ROS/RNS assays.