Synthetic Sensors for Reactive Oxygen Species Detection and Quantification: A Critical Review of Current Methods

Abstract

Significance: Redox reactions play important roles in both physiological and pathological processes, highlighting the importance of quantifying and localizing intracellular redox-active components. Most research has focused on direct investigation of reactive oxygen species (ROS). Intensity-based fluorescent methods are very sensitive and easy to use, but they lack specificity and can produce artifacts. In this article, we focus on synthetic sensors, describing experimental pitfalls associated with their use. We also present alternative methods for the detection of free radicals. Recent Advances: New approaches have been developed to overcome the main artifact of intensity-based methods: spurious changes in fluorescence intensity caused by oxidation. These new approaches are based on analytical measurements of the oxidized sensors or techniques that are not susceptible to oxidation, such as electron spin resonance and fluorescence lifetime-based methods. Regardless of the approach, the need for detection of ROS on the subcellular level, especially in the mitochondria, has motivated the development of new probes. Critical Issues: Flow cytometry systems and confocal microscopes are now available to the majority of biologists, and commercially available probes are, therefore, more widely used. The fact that these new applications are cited in thousands of publications makes these sensors even more attractive. Future Directions: The field of ROS detection by synthetic sensors continues to expand, bringing needed additional research to the development of robust techniques that are applicable both in vitro and in vivo. Antioxid. Redox Signal. 25, 520–533.

Introduction

To a chemist, a redox reaction is the transfer of an electron from one molecule (the reducer) to another (the oxidant). For a biologist, it is the beginning of life. In mitochondrial respiration, for instance, incomplete reduction of molecular oxygen leads to the production of oxygen species that act as oxidants. Although the mitochondrial respiratory chain responsible for electron transport is an important producer of reactive oxygen species (ROS) (11), it is not the only source. Another important ROS generator is NADPH (nicotinamide adenine dinucleotide phosphate, reduced form) oxidase (NOX) (62). Intracellular oxidants can not only damage macromolecules such as proteins, nucleic acids, and lipids but also regulate cell function, and some ROS participate in different signal transduction pathways. As a consequence of this duality, cells have evolved mechanisms to sense such changes and adapt their behavior accordingly.

Because numerous diseases reflect an imbalance between positive and negative ROS effects (34), researchers also need to measure this electron deregulation and understand which redox-active components are present and at what concentrations. Most research on redox sensing has focused on direct investigation and detection of oxidants that are actively involved in redox processes. Oxidation can be non-catalytic, mediated by ROS, and catalytic, performed by enzymes. However, the majority of studies have focused on ROS. It should be noted that “ROS” does not refer to a single entity; these species can be divided into non-radical oxidants, such as hydrogen peroxide (H₂O₂) and hypochlorous acid (HOCl), among others, and radical oxidants with an unpaired electron, such as superoxide anion (free radical) (O₂ ⁻) and nitric oxide (NO). The latter are also called free radicals.

Having an efficient and reliable technique to directly detect changes in intracellular ROS production is a fundamental need in many research fields. Experimental culture conditions (e.g., substrates in the medium, oxygen tension) will greatly affect the sources and magnitude of cellular ROS production. In addition, owing to the high reactivity and extremely short half-life of ROS, quantification of specific ROS has always been challenging and prone to artifacts. Several techniques are currently employed to measure ROS in cultured cells and given the massive expansion of the ROS field, rigorous recommendations have been proposed (29). The purpose of this article is to discuss synthetic sensors. The growing field of protein sensors, such as genetically encoded redox probes, though very important, is outside the scope of this article and will not be addressed.

On a cellular level, sensor-based fluorescence detection is superior because of its high sensitivity, experimental convenience, and non-invasiveness. Fluorescent sensors can be measured with spectrophotometers or microplate readers and using wide-field and confocal microscopy, and flow cytometry. Only the last two approaches allow evaluation at the single-cell level. A large number of fluorescent synthetic probes capable of being introduced into living cells and organisms have been developed. These probes exhibit a change in their fluorescence intensity in response to oxidation. In most applications, fluorescent probes are generated in situ by the reaction of an inactive form of the probe with ROS. Thus, the probes are stable in a reduced state, but they can be oxidized by the species of interest. Two types of probes are employed: chemiluminescent probes, such as luminol and lucigenin (28, 51), and fluorescent probes, such as the extensively used reagents, dihydroethidium (DHE) (107), dichlorofluorescein derivatives (65, 81), dihydrorhodamine (82), and Amplex Red (10-acetyl-3,7-dihydroxyphenoxazine [AmR]) (55, 56). However, the fact that the detection principle is based on a redox reaction will directly affect selectivity. Even if the sensor is stable in the reduced state, it must be sufficiently reactive and easily oxidized; moreover, the identity of all the molecules responsible for the oxidation is often difficult to determine. Recently, it was demonstrated that reactive sulfide species can even be detected with high sensitivity by some of these fluorescent probes (19). The sensor undergoes two electron oxidation steps, usually through formation of a radical intermediate. Consequently, these probes are associated with several common pitfalls: They have limited resistance against oxidation, the products of probe reactions with ROS may be highly unstable, and they may also be subject to redox cycling that generates additional ROS (5, 31, 42).

After a brief introduction to these common probes, we describe practical experiences that not only highlight difficulties associated with their use but also indicate how these problems can be solved. A number of alternatives have been developed to circumvent problems encountered in using these probes. Some include sensors based on a non-redox mechanism, such as deprotection of a fluorescent molecule by ROS (21, 52), that still uses changes in fluorescence intensity as the detection principle; these are reviewed in Maghzal et al. (53) and for boronate-protected sensors in Lin et al. (49) and Zielonka et al. (109). Others include analytical measurements to quantify the products of oxidation by high-performance liquid chromatography (HPLC) or mass spectroscopy (42). There is also a high interest in protein sensors, such as genetically encoded redox probes (24), that are discussed in another article of this Forum. In addition to considering these approaches, this article will discuss two other types of detection techniques: electron paramagnetic resonance (EPR) (46) and fluorescent lifetime-based methods (76). Both techniques directly detect free radicals based on their paramagnetic properties. We will then address the question of location, focusing on one subcellular compartment—the mitochondria. We will describe possibilities of targeting synthetic probes to mitochondria, and we will then evaluate contributions from different techniques to the field of ROS detection.

Intensity-Based Fluorescent Synthetic Sensors—Oxidant-Sensitive Probes

“Oxidation as a common activation route for probes” from Wardman (99).

In addition to the book by Halliwell and Gutteridge (34), there are numerous reviews on intracellular ROS detection. During the past 10 years, several reviews were published that focused on its various aspects. In 2006, Bartosz proposed an interesting description of the problems that might arise due to the limited stability and high reactivity of fluorescent sensors (5). Soh's discussion included singlet oxygen ¹O₂ (88). In 2007, Dikalov et al. presented the major assays with their strengths and weaknesses (23). Wardman completed some unexplored aspects of these earlier reviews (99). In 2010, Rhee et al. focused on H₂O₂ (72) and in 2014, Fan and Li concentrated on anion superoxide (27). Chen et al. presented unconventional sensors (14). In 2013, Woolley et al. drew up a list of ideal criteria that sensors for ROS detection must fulfill and designed a logic diagram for deciding which sensors should be used for which experimental design (100). Kalyanaraman and Zielonka continue to warn about the most used sensors (18). All authors advise the reader about the complexities associated with ROS detection and the difficulty, if not impossibility, of using well-known synthetic probes for this purpose. They recommend that these probes be considered redox indicators. However, it is understandable that biologists who need to quantify ROS will be unable to resist using commercially available probes, especially since they are cited in thousands of publications. In addition to reiterating concerns that have been well covered in the aforementioned reviews, we will also focus here on difficulties, usually not published, that can be encountered in attempting to apply these sensors in the laboratory.

DHE (hydroethidine)

The usefulness of DHE, a cell-permeable compound, as an indicator of redox components has been reviewed by Zielonka and Kalyanaraman (107). Intracellularly, DHE is oxidized to the fluorescent products, ethidium and 2-OH-ethidium (2-hydroxyethidium; Fig. 1A). For several years, ethidium was considered the sole product of DHE oxidation (94). Both oxidized compounds can intercalate into nuclear DNA, which increases the quantum yield of their fluorescence. Oxidation of DHE is a superoxide-dependent process in which 2-OH-ethidium is formed by superoxide anion, although the reaction of DHE with other ROS or oxidant species and subsequent formation of ethidium may interfere. Because the fluorescence of 2-OH-ethidium overlaps with that of ethidium, it has been suggested that HPLC be performed to quantify 2-OH-ethidium and thus superoxide anion (107).

FIG. 1.

Dihydroethidium (DHE) forms fluorescent compounds after oxidation. (A) Selective oxidation of DHE by superoxide anion generates 2-hydroxyethidium, whereas nonspecific oxidation generates ethidium. (B) Fluorescence emissions after excitation at 510 nm of CCRF-CEM (black curve) and H9c2 (gray curve) cell lines. The cells were loaded with DHE (thick curve) or exposed to vehicle (thin curve). (C) Fluorescence emissions of CCRF-CEM cells treated with the drug doxorubicin after excitation at 510 nm. Experimental fluorescence spectrum (thick dark curve) is compared with a calculated curve (thick gray curve). The calculated curve was obtained by addition of the three fluorescence spectra obtained individually (drug alone, oxidized DHE alone, and intrinsic fluorescence alone). DHE, dihydroethidium.

Fluorescence-based techniques

The traditional excitation wavelength is based on the maximum emission of ethidium at 518 nm. The two forms of oxidized DHE, ethidium and 2-OH-ethidium, fluoresce as single bands at 593 and 567 nm, respectively (104). The relative intensity of the two compounds differs slightly among published reports (79, 105, 110); however, they are of the same order of magnitude. After excitation above 510 nm, the fluorescence spectrum of different DHE-loaded cell lines shows a maximum at ∼587 nm (ethidium fluorescence), but not at 568 nm (Fig. 1B). As calculated by HPLC (110), the level of ethidium detected in biological systems is nearly 10-fold higher than that of 2-OH-ethidium, making ethidium fluorescence the major contributor. Robinson et al. proposed that excitation at 396 nm would preferentially excite 2-OH-ethidium, which would thus exhibit more intense fluorescence than ethidium (79). These authors published a general protocol using two excitation wavelengths (78). It should be noted that autofluorescence of cells (Fig. 1B), and in some case fluorescence of incorporated drugs (Fig. 1C), can interfere with the fluorescence of oxidized products. An example is the interaction with doxorubicin, described by us (69). In such cases, we recommend working with the entire spectrum. The complex fluorescence spectra of cells treated with drugs and loaded with DHE can be analyzed to identify and quantify all fluorescent components that contribute to the complex experimental spectra. The method was previously described (10) and uses the spectra of all fluorescent components obtained individually: drugs in cells, ethidium in cells, and the intrinsic fluorescence of cells (Fig. 1C). A computer tool allows the reconstruction of a theoretical spectrum by combining the characteristic fluorescence spectra obtained beforehand. The theoretical and experimental spectra are compared using graphical or numerical estimators.

Stability of DHE

Because DHE is also fluorescent (excitation at 350 nm, emission at 400 nm), a ratiometric approach was initially proposed for DHE-based assays using the intensity ratio, 640/420 nm, between oxidized and reduced forms of DHE (13). Subsequently, a single-wavelength, cytometry-based application of DHE has come to be the classical method for assessing ROS (59). We and others (9, 91) have found that DHE is not stable when irradiated by ultraviolet light; after irradiation at 355 nm, the fluorescence of DHE disappears within a few minutes. Because compound stability is a prerequisite for using the ratiometric configuration of the assay, DHE intensity must be recorded shortly after preparation of the compound and using extremely low levels of irradiation. However, although we have found that application of these rather drastic conditions to ratiometric assays seems to increase the reproducibility of our data, we consider that the benefit does not justify the considerable experimental effort.

Practical recommendations

The literature seldom mentions the blue color characteristic of solutions containing reduced DHE. When working solutions (1 mM) are prepared immediately before use by diluting stock solutions into degassed medium under a nitrogen flow, the resulting solution is pale pink rather than blue owing to contamination by the oxidized form of DHE. Zielonka et al. recommend preparing stock solutions in an anaerobic chamber (110). In this case, a stock solution of DHE prepared in degassed dimethyl sulfoxide at 10 mM can be stored under N₂ at −20°C for <1 month. Longer storage might require keeping DHE at −80°C. Working solutions prepared after this point take on a dark pink coloration owing to partial or total oxidation and must be discarded. Robinson et al. proposed that if contamination does not exceed 10%, DHE can be chemically reduced or purified by HPLC (79). Because the photo-oxidation of DHE is oxygen dependent, solutions are prepared under subdued light in degassed solutions. Extracellular molecules that fluoresce at wavelengths above 530 nm, such as Phenol Red, must be removed from the medium by washing. Consequently, suspensions of cells are usually incubated with 20 μM DHE in phosphate-buffered saline (PBS) in the dark at 37°C for 15–30 min. After that, the preparation can be used with or without washing. Fluorescence values must be normalized to the number of cells in each sample. In the literature, the formation of oxidative species by DHE owing to redox cycling is described as being less important than is the case for other superoxide sensors, such as luminol and nitroblue tetrazolium.

Amplex Red

The AmR assay is based on the enzymatic oxidation of the colorless, non-fluorescent compound AmR through a radical intermediate (32). AmR reacts with H₂O₂ in the presence of horseradish peroxidase to generate the stable colored and fluorescent product, resorufin (106) (Fig. 2A). AmR is not permeable, and it thus measures extracellular H₂O₂. Because H₂O₂ diffuses through lipid membranes, this assay gives an indication of cellular production, provided cells release enough H₂O₂. Production and secretion of H₂O₂ by cells has been reported for several cell types, including hemocytes (57), activated human leukocytes (56), neutrophils (106), endothelial cells (89), synaptosomes (3), and podocytes (45). Most other cell lines do not release detectable concentrations of H₂O₂ into the extracellular medium. Kits are commercially available and easy to use. However, as is the case for all synthetic sensors, AmR presents several pitfalls (18, 103). Notably, exposure of AmR to light also results in the formation of resorufin (103); thus, experiments with AmR should be conducted under subdued lighting.

FIG. 2.

Cell-impermeable Amplex-Red behavior in presence of H₂O₂. (A) Oxidation of AmR to form the fluorescent compound resorufin. (B) Degradation of H₂O₂ after 0, 5, 10, and 20 min of contact with cell lysates. For all three concentrations, about 50% of H₂O₂ is degraded after 20 min. (C) Extracellular H₂O₂ concentration remaining after addition of 1 mM H₂O₂ in PBS. After 60 min of contact with cells, only 44% of H₂O₂ remains. (D) The intercept of standard curves increases with time (gray, 1 day; white, 3 days; black, 5 days) after AmR preparation, indicating slow oxidation, even if aliquots are kept at −20°C. H9c2 cell line was used for these experiments. AmR, 10-acetyl-3,7-dihydroxyphenoxazine; H₂O₂, hydrogen peroxide; PBS, phosphate-buffered saline.

Cells that do not release H₂O₂

Because some cell lines do not release detectable concentrations of extracellular H₂O₂, several teams have modified the technique to include a cell lysis step to release intracellular H₂O₂ into the external medium. Some preliminary adjustments for lysis conditions (time, concentration, medium) are necessary, and we perform them using the inexpensive FOX-2 (ferrous oxidation in xylenol orange, version 2) method (87), which allows for the determination of hydroperoxides by measuring absorbance. We have found that experiments cannot be performed in culture medium, because H₂O₂ is eliminated by cell culture components, possibly sodium pyruvate (30); thus, these experiments are performed after washing cells three times and resuspending them in PBS. Cell lysate components are also effective in eliminating extracellular H₂O₂, as shown in Figure 2B, which presents the decrease in H₂O₂ concentration over time caused by contact with cell lysate. Accordingly, we recommend performing cell lysis and measurements within 5 min. Alternatively, for cells that release H₂O₂ slowly, the duration of AmR loading was increased up to 60–120 min to accumulate AmR signal (23, 58). For such a long experiment, one has to be careful about photo-oxidation of the sensors. Noteworthy, protocols to generate oxidative stress often propose the addition of exogenous H₂O₂ solution at 1 mM to cells. After 1 h, ∼56% of added H₂O₂ enters into the cells (Fig. 2C). Obviously, H₂O₂ is also converted inside the cells during the 1-h treatment, as evidenced by the fact that <1 μM of H₂O₂ equivalents are released after lysis; in contrast, non-lysed cells do not release H₂O₂ (70).

Practical recommendations

The proposed protocol was not only developed for cardiac cells but can also be performed on other adherent and detached cells. After washing and trypsin treatment (if necessary), cells are suspended in PBS or KRPG (AmR kit), and then 5% Triton-X (without H₂O₂) is added to a final concentration of 0.25% (v/v). Cells are lysed for 5 min (including 2 min of centrifugation, if needed) before addition of the reagent. The lysate is suspended in half the volume of the micro-wells. Identical volumes of AmR reaction mixtures are added, avoiding exposure to light, and spectra are recorded between 560 and 600 nm after excitation at 550 nm (max: 580 nm) for 1 h. Fluorescence changes can be converted to H₂O₂ equivalents (μM released) using a standard curve. H₂O₂ standard curves must be prepared daily by adding equal amounts of AmR reaction mixture to different dilutions of H₂O₂ in KRPG (AmR kit). Fluorescence readings become stable within 10–20 min. If the slope of the standard curve is stable and ranges from 9 to 11 μM ⁻¹, the intercept varies depending on the time between AmR preparation and measurement (Fig. 2D). We strongly recommend preparing blank and H₂O₂ solutions for each set of experiments. According to several authors, AmR assays are reported to be able to detect concentrations as low as 0.05 μM (106), although in our hands, with a conventional fluorimeter, this technique requires concentrations >0.1 μM. Because resorufin is subject to autoproduction of ROS and other artifacts, it is doubtful that biologically realistic H₂O₂ production can be quantified using this approach.

Chemiluminescent sensors

Lucigenin (bis-N-methyl-acridinium) and luminol (5-amino-2,3-dihydro-1.4-phthalazinedione) have frequently been used for the detection of superoxide anion (Fig. 3A). The complex chemistry of both compounds was already described in a review in 1993 (28). Proceeding through free radical formation, an electronically excited molecule is produced that emits a photon when returning to its ground state. The luminescence can be detected by a scintillation counter or a luminometer (2). The main disadvantage consists of production of O₂ ⁻ during the process (a phenomenon known as redox cycling). Because of redox cycling, strong pH dependence, and absence of specificity, luminol was abandoned for this application (44). L-012 (8-amino-5-chloro-7-phenyl-pyrido[3,4-d]pyridazine-1,4(2H,3H)dione), a brighter analog, was developed (65, 108). The other compound, lucigenin is also prone to redox cycling. The use of low concentrations of this compound minimizes this problem (98), but its use remains controversial (23, 99). Therefore, other compounds, not prone to redox cycling, were proposed. Based on the central structure of coelenterazine and cypridina luciferin (Fig. 3B), various structural analogs were developed. Among them, MCLA [2-methyl-6-(p-methoxyphenyl)-3,7-dihydroimidazo(1,2-α)pyrazin-3-one] is described as being a useful probe for O₂ ⁻ detection, even if its specificity is not entirely sure (43). New analogs are still under development that have various emitting wavelengths (41). It should be noted that the localization of these sensors is disputed: extracellular (54), intracellular, or even mitochondrial localization (68). One explanation of this discrepancy could be that localization depends on concentration and cell line. Bioluminescence was also used to detect H₂O₂ (33). PCL-1 [peroxy caged luciferin 1 ((4S)-2-[6-[(4-Boronophenyl)methoxy]-2-benzothiazolyl]-4,5-dihydro-4-thiazolecarboxylic acid)] is a boronate-protected analog of firefly luciferin peroxynitrile (Fig. 3C). After being deprotected by H₂O₂ (95) and/or peroxinitrile (85) and a reaction with luciferase, photons will be emitted.

FIG. 3.

A selection of structures of chemiluminescent sensors. (A) Lucigenin and luminol. (B) The natural coelenterazine and cypridina luciferin have the same central structure (bold) that was used to develop new probes such as MCLA. (C) Analog of firefly luciferin, PCL-1. MCLA, 2-methyl-6-(p-methoxyphenyl)-3,7-dihydroimidazo(1,2-α)pyrazin-3-one; PCL-1, peroxy caged luciferin 1 ((4S)-2-[6-[(4-Boronophenyl)methoxy]-2-benzothiazolyl]-4,5-dihydro-4-thiazolecarboxylic acid).

Alternative Techniques for Detection of Free Radicals

Two detection methods based on the paramagnetic properties of free radicals have been developed. The first, EPR (or ESR) spectroscopy allows the direct detection and characterization of molecules or a paramagnetic center having unpaired electrons (46); however, it cannot directly detect short-lived, intracellular free radicals. Instead, detection of these species is based on the formation of persistent radicals by spin trapping or spin probing (92, 97). The second method is based on fluorescence lifetime measurements (6, 83). Sensors with long fluorescence lifetimes—the time required for a population of excited sensors to return to their non-excited state by emitting light—are sensitive to collisions with paramagnetic species (20, 66). This phenomenon, called dynamic quenching, decreases the fluorescence intensity and lifetime of the sensors, with the decrease being proportional to quencher concentration.

Electron paramagnetic resonance

The fundamental physical concepts of EPR are analogous to those of nuclear magnetic resonance, except that electron spin, rather than nuclear spin, is excited. An EPR spectrum is usually obtained by varying magnetic field strength at a fixed microwave frequency. This technique has historically required expensive equipment and skilled operators; however, a number of manufacturers have worked to produce EPR spectrometers that are easy to operate. Each free radical exhibits a specific EPR spectrum, and the intensity of an EPR signal is proportional to its concentration, allowing for identification and quantitative measurement of free radicals. In principle, this technique is ideal, but it has limitations, namely that it is not sensitive enough to directly detect highly reactive free radicals in cells. Increasing the signal requires the addition of spin traps or spin probes—diamagnetic molecules that react with intracellular free radicals to form new, more stable radicals with half-lives that are long enough to accumulate and be EPR detectable (5, 23).

Spin traps form covalent bonds with the radical by an addition reaction that yields a long-lived spin adduct. The reaction creates a unique adduct for each free radical that can be specifically identified (Fig. 4A). The most common traps are nitroso, nitrone (12), and cyclic nitrone (1, 90), which yield nitroxide adducts (accurately termed aminoxyl radicals) that can be detected by EPR. Among conditions that spin traps must fulfill to be considered efficient free radical detectors, the most crucial are kinetic criteria; specifically, they must trap free radicals rapidly. Besides low spin-trapping rates, spontaneous and cell-enhanced decomposition to products not detectable by EPR and the need for high concentrations of spin traps are potential pitfalls that limit the applications of spin traps in biological systems (25).

FIG. 4.

A selection of structures of spin traps and spin probes. (A) Addition of a free radical to spin traps (nitroso, linear nitrone, and cyclic nitrone) yields specific nitroxide adducts. (B) Oxidation of cyclic hydroxylamine spin probes by the free radical yields a nonspecific nitroxide radical.

Over time, research has improved spin adduct stability and reduced its toxicity through structural modifications to the trap (37). Hydroxylamide, a diamagnetic product, is formed by direct reduction of nitroxide adduct, a reaction that competes with free radical quantification. However, this drawback can be usefully exploited to differentiate between normal and pathological redox states status by using the reduction of cell-permeable nitroxides to hydroxylamines (35). Cyclic hydroxylamine-based ROS sensors that are oxidized by ROS without binding have also been developed (Fig. 4B). Because they do not “trap” the radicals, the terms “spin label” or “spin probe” are preferred. A redox reaction involving one-electron oxidation by a free radical leads to the recovery of the nitroxide radical, which is detected by EPR (25); consequently, hydroxylamines yield the same nitroxide radical regardless of the free radical being detected. Among the advantages of hydroxylamine spin probes is their suitability for use at very low concentrations (sub-millimolar range) compared with spin traps; they also show better stability and faster kinetics. However, free radical identification is complicated by the fact that hydroxylamines can be potentially oxidized by other oxidants. It should be noted that the term “spin labeling” is also used for an entirely different method in which nitroxide radicals are used for probing local dynamics of protein or biological membranes, a nomenclature overlap that might lead to confusion for the unprepared reader (7, 38).

Time-resolved microfluorimetry

The change in the fluorescence lifetime of a given sensor can be used to detect and quantify free radicals. Dynamic quenching, which forms the basis of this technique, is in some ways similar to Förster resonance energy transfer (FRET), a more common process. In both cases, the excited fluorescent sensor loses its energy without emitting light. In FRET, this energy is transferred to a second molecule in close vicinity; whereas in dynamic quenching, this energy is lost via collision with quenchers. In the latter case, the fluorescence intensity and lifetime of the sensor decrease as a result of multiple processes that occur at the moment of collision. Paramagnetic species, such as oxygen and free radicals, have the property of quenching long-lived fluorophores, such as metal complexes and aromatic compounds, that have long lifetimes (hundreds of nanoseconds). Use of lifetime-based fluorescent sensors circumvents a certain number of problems associated with intensity-based sensors, since the measurement does not depend on variations in sample thickness, cellular uptake, or sensor concentrations. Several additional advantages include stability of the sensor, which does not need to react with free radicals, and the reversibility of the process, since cessation of free radical production would be immediately observed. Measured values are not attributable to the accumulation of free radicals, but they instead reflect their production at the exact time of measurement.

Successful applications

Ten years ago, we started using 1-pyrene butyric acid (PBA) in living cells, and by following its lifetime variation, we were able to quantify free radicals. Dynamic quenching plays an important role in measuring oxygen, and in the 1970s, Vaughan and Weber were the first to demonstrate that PBA could be used as an oxygen probe (96). More recently, it has been shown that fluorescence quenching of PBA can also be used to measure free radicals (16, 20, 66). The concept has been validated for PBA and other probes (Fig. 5A) in solution (66) and in loaded primary cells (57, 67, 73), cultured cells (69, 71, 84), and microalgae (8). We have shown that this sensor is sensitive to small, mobile radical molecules, but not to non-radical ROS such as H₂O₂.

FIG. 5.

A selection of sensors with long fluorscence lifetimes and associated decays in cells. (A) Structures of commercially available (^¥) sensors with lifetimes of hundreds of nanoseconds. (B) Fluorescence decay profile of a single cell loaded with PBA. The inset shows a zoomed view of the beginning of the decay. The experimental decay is resolved into three exponentials. From left to right: free NAD(P)H, bound NAD(P)H, and PBA fluorescence decay. At the bottom of the figure is shown the associated weighted residual curve. (C) Free radical levels (fold increase vs. control) in cells treated with DPI, Baker solution, or both. Data are means ± SD of three independent experiments (n = 70–140 cells/condition; *p ≤ 0.05 vs. control, Student's t-test). DPI, diphenyleneiodonium chloride; NADPH, nicotinamide adenine dinucleotide phosphate, reduced form; PBA, 1-pyrene butyric acid.

Fluorescence-based techniques

Pulsed-lifetime measurements are recorded using a previously described apparatus (74, 76). Fluorescence decay of PBA in PBA-loaded cells is recorded and analyzed for each individual cell. Application of a 404-nm filter to the emissions path after excitation with a nitrogen laser at 337 nm allows decay to be resolved into three exponential curves (Fig. 5B). Time constants (i.e., lifetimes) and amplitude values of each exponential curve in the decay are obtained using the downhill simplex method (63). The two short time constants are attributable to free and bound NAD(P)H, and they provide information about cellular activity, a property that will not be further discussed here. The relationship between the long time constant (>100 ns), which is characteristic of pyrene derivatives, and free radical concentration is described by the Stern–Volmer equation. One can, therefore, calculate intracellular free radical production in treated cells by a comparison with control cells (expressed as fold change) using the following equation: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}\left[ {{ \rm{ROS}}} \right] = ( \tau { \rm{c}} / \tau ) \times [ ( { \tau _0} - \tau ) \left] / \right[ ( { \tau _0} - \tau { \rm{c}} ) ] \times \left[ {{ \rm{ROS}}} \right] { \rm{c}} \tag{1}\end{align*} \end{document}

where τ is the fluorescence lifetime measured for treated cells, τc is the mean of the lifetime measured for non-treated cells, and τ₀ is the mean of the fluorescence lifetime in the absence of free radicals.

Practical recommendations

Cells must be rinsed once with Hanks Balanced salt solution (HBBS) to remove protein that can bind PBA and prevent loading. Then, cells must be incubated at 37°C for 5–15 min in HBSS containing 0.2–1.6 μM PBA, depending on the cell line. Removal of all extracellular PBA is a key point; cells are rinsed three times and resuspended in HBBS, after which the cell suspension is placed on the objective. A total of 12–20 cells should be observed in a single recording session, reducing the time required to complete the entire process to ∼30 min. During the recording procedure, the cells remain physiologically intact, and ROS formation is stable.

One challenge in applying the lifetime-based method is that quantification of free radical changes requires obtaining intact cells that no longer produce reactive oxygen. Generally, placing cells in paraformaldehyde (Baker solution) stops mitochondrial activity and ROS generation (70), while preventing a loss of membrane integrity and probe reorganization. However, for other cell lines such as cardiac cells (69), which are known to contain membrane-bound NOX, fixation is not sufficient to stop all free radical production, and additional treatment with the NOX inhibitor diphenyleneiodonium chloride (DPI) is necessary (Fig. 5C). DPI is extensively used as anion superoxide inhibitor. It is a flavin inhibitor that blocks not only NOX but also mitochondrial production of superoxide (50). The method is sensitive to substantial variations in intracellular oxygen. However, several experiments have shown no impact of oxygen (70, 76), probably owing to the rapid diffusion of oxygen inside cells. It is important to bear in mind that fluorescence lifetime (as intensity) is highly sensitive to changes in the environment, which must be adequately controlled.

Mitochondrial Targeting Sensors

Mitochondria play a central role in ROS biology. They are an important source of ROS within cells, are involved in cell survival and death, and are promising therapeutic targets. Different carriers and vectors have been developed to reach mitochondria. To serve these functions, these probes need to cross not only the cell membrane but also outer and inner mitochondrial membranes to access the organelle. Although some research has focused on detecting redox-active components in isolated mitochondria (4, 60), there is considerable interest in performing measurements in living cells. As is the case for intracellular ROS detection, mitochondrial ROS detection has been the subject of a number of reviews (22, 24, 102), which describe problems similar to those encountered with cellular sensors (107).

Lipophilic cationic compounds

The families of mitochondrial vectors are presented in Figure 6A. These vectors exploit the higher membrane potential maintained by mitochondria, which offers a unique chemical opportunity to selectively target the organelle. Structurally, these vectors include positive charges and lipophilic groups (40, 77, 86). The lipophilic part facilitates movement across the membrane, whereas the positive charge takes advantage of the proton gradient and subsequent electrochemical potential generated within the matrix of mitochondria. However, good ROS sensors must enter mitochondria regardless of their polarization status and should not change the potential gradient on entering the organelle. Since fluorescence signals are directly dependent on the number of mitochondria per cell, mitochondria number must be determined simultaneously. Finally, sensors should not interfere with the biological system. It is possible that loading sensors in mitochondria could disturb cells and influence both membrane potential and ROS. Designing properly working sensors for mitochondria is, therefore, not trivial.

FIG. 6.

A selection of structures of mitochondrial sensors. (A) Examples of mitochondrial delivery vehicles, including the TPP group, rhodamine and rosamine fluorescent vectors, and mitochondria-targeted peptides. (B–K) Examples of structures of vehicles and cargos. (B–F) Sensors attached to TPP. Fluorescence-based sensor (B, C); mass spectra-based sensor (D); and EPR-based sensor spin traps (E) and spin probes (F). (G–I) Rh123 and Mito Tracker series are fluorescence-based sensors for ROS detection. (J) Pyrene attached to Rh110 for time-resolved microfluorimetry. (K) Pyrene attached to a cargo peptide for time-resolved microfluorimetry. Commercially available sensors (^¥). EPR, electron paramagnetic resonance; Rh123, [6-amino-9-(2-methoxycarbonylphenyl)xanthen-3-ylidene]azanium chloride; ROS, reactive oxygen species; TPP, triphenylphosphonium.

Triphenylphosphonium-based probes

The triphenylphosphonium (TPP) group has a positive charge on the phosphonium and is surrounded by three lipophilic phenyl groups (80). There are several sensors that carry the TPP group (Fig. 6B–F). These allow ROS detection by various methods, including fluorescence intensity as well as mass spectroscopy and spin trap. Some are already commercially available, such as the widely used Mito-HE, better known as mitochondrion-targeted DHE (MitoSOX). MitoSOX (Fig. 6B) is based on DHE linked to TPP by a hexyl carbon chain. Designed for anion superoxide detection, it behaves similarly to its parent compound DHE (79, 107). Not unexpectedly, MitoSOX is also preferentially oxidized by superoxide to a hydroxylated product by the same mechanism as DHE. Among other TPP-based sensors is MitoPY1 (triphenylphosphonium conjugate with boronate-based peroxy-yellow 1; Fig. 6C), which was designed for H₂O₂ detection (21). The protecting boranate moiety reacts with H₂O₂; removal of the boronate moiety allows light emission. Notably, it also reacts with peroxynitrite. To avoid difficulties associated with fluorescence-based detection, several teams have developed mitochondria-targeted sensors that use alternative detection methods. A new sensor, MitoB [(3-boronophenyl)methyl]triphenyl-phosphonium (Fig. 6D), detected by mass spectroscopy, was recently described (15). Also in the area of EPR, several nitrone spin traps (26, 36) (Fig. 6E) and hydroxylamine spin probes (25) (Fig. 6F) have been modified by addition of a TPP group.

Lipophilic cations

Rhodamine and rosamine sensors share the property of encompassing a cationic functionality in an otherwise nonpolar framework. Their accumulation is usually strongly dependent on membrane potential. Rh123 ([6-amino-9-(2-methoxycarbonylphenyl)xanthen-3-ylidene]azanium chloride) and Mito Tracker series probes are extensively used for mitochondrial imaging and membrane potential evaluation. A number of fluorescent ROS sensors derived from these parent compounds have been developed (Fig. 6G–J). One of these, DHR [dihydrorhodamine 123 (2-(3,6-Diamino-9H-xanthene-9-yl)-benzoic acid methyl ester; Fig. 6G], is transformed by oxidation into the fluorescent Rh123, but can react with various ROS and oxidizing species (17). However, DHR cannot correctly quantify ROS in the mitochondria, because it is cationic and only localizes in mitochondria after oxidation. More recently synthesized rhodamine-like sensors include MitoHR and MitoAR (Fig. 6H) (47). Among the best-known sensors in the Mito Tracker series is Mito Tracker Red CM-H₂XRos (Fig. 6I). This reduced, non-fluorescent version of MitoTracker Red that fluoresces on oxidation was characterized in a comparative study by Kuznetsov et al. (48). Our group has used a rhodamine carrier to synthesize the pyrene-based sensors, PRE-4 [(1"-pyrene butyl)-2-rhodamine ester] (75) and PRE-6 [(1"-pyrene hexyl)-2-rhodamine ester] (Fig. 6J), which successfully reach mitochondria. However, folding of these two chromophores and subsequent energy transfer between them has so far prevented their utilization.

Peptidic delivery agents

Peptides containing both naturally occurring and non-natural amino acids designed by alternating lipophilic and positively charged moieties constitute the third group of probe types (39, 101). Shiller and Szeto peptides were designed to target mitochondria (93), they present intrinsic antioxidant activity. Subsequently, the Kelley laboratory created mitochondria-penetrating peptides (MPP) combining synthetic and natural residues that are either cationic (e.g., arginine) or hydrophobic (e.g., cyclohexylalanine). The solid-phase synthesis of such peptides is relatively straightforward, and sensors can be added through peptide bond formation. A variety of cargos have been proposed (101), but to our knowledge, no ROS sensor has been designed using this technique. Our group recently used this approach to synthesize Mito-PB [pyrene-F_XrF_XrF_Xr (F_X = cyclohexylalanine, r = darginine)], a pyrene moiety carried to mitochondria by this peptide vector (Fig. 6K). Because fluorescent lifetime is used for detection, changes in mitochondrial membrane potential or mitochondria numbers will not influence quenching by free radicals and, thus, do not affect free radical quantification. However, our initial studies showed that the presence of the sensor in mitochondria can disturb cellular activity in a concentration- and cell line-dependent manner.

Conclusions

There is a vast literature on the role of ROS in biological systems, how they cause damage, and how they are involved in cell regulatory and signaling pathways. To continue and broaden the progress in these areas, researchers need robust methods. Consequently, the challenges presented by difficulties associated with measuring redox species, especially intracellular ROS, require solutions. As summarized during the conference, “Chemistry and Biology of Reactive Oxygen Species” (61), two points warrant particular emphasis: Available knowledge must be transmitted to a wider audience, and better approaches for ROS detection need to be developed. These approaches must include the design of sensors with improved selectivity, reversibility, and compartment-targeting properties. Among the numerous approaches under development are fluorescence-based sensors that can be used with in vivo imaging technology to visualize redox-active components at the level of intact cells, whole organs or tissues, and even living organisms.

Footnotes

Abbreviations Used

References

Abbas

, Hardy

, Poulhès

, Karoui

, Tordo

, Ouari

, and Peyrot

. Detection of superoxide production in stimulated and unstimulated living cells using new cyclic nitrone spin traps. Free Radic Biol Med, 71: 281–290, 2014.

Agarwal

, Allamaneni

, and Said

. Chemiluminescence technique for measuring reactive oxygen species. Reprod Biomed Online, 9: 466–468, 2004.

Barbosa

, Capela

, Oliveira

, Silva

, Ferreira

, Siopa

, Branco

, Fernandes

, Duarte

, de Lourdes Bastos

, and Carvalho

. Pro-oxidant effects of Ecstasy and its metabolites in mouse brain synaptosomes. Br J Pharmacol, 165: 1017–1033, 2012.

Barja

. Minireview: The quantitative measurement of H₂O₂ generation in isolated mitochondria. J Bioenerg Biomembr, 34: 227–233, 2002.

Bartosz

. Use of spectroscopic probes for detection of reactive oxygen species. Clin Chim Acta, 368: 53–76, 2006.

Berezin

and Achilefu

. Fluorescence lifetime measurements and biological imaging. Chem Rev, 110: 2641–2684, 2010.

Berliner

. History of the use of nitroxides (aminoxyl radicals) in biochemistry: past, present and future of spin label and probe method. In: Nitroxides - Theory, Experiment and Applications, edited by Kokorin

. InTech, 2012. Available at www.intechopen.com/books/nitroxides-theory-experiment-and-applications/history-of-the-use-of-nitroxides-aminoxyl-radicals-in-biochemistry-past-present-and-future-of-spin-l (Accessed June 22, 2016 .).

Bijoux

and Ribou

A-C

. Time-resolved microfluorimetry: An alternative method for free radical and metabolic rate detection in microalgae. Biotechnol J, 9: 294–300, 2014.

Bindokas

, Jordan

, Lee

, and Miller

. Superoxide production in rat hippocampal neurons: Selective imaging with hydroethidine. J Neurosci, 16: 1324–1336, 1996.

10.

Birla

, Prieto

, Noguel

, Vigo

, and Ribou

A-C

. Comparative study of the interaction between BSA and three fluorescent probes: Investigation of some critical parameters. Rev Roum Chim, 52: 639–646, 2007.

11.

Brand

. The sites and topology of mitochondrial superoxide production. Exp Gerontol, 45: 466–472, 2010.

12.

Buettner

. Spin trapping: ESR parameters of spin adducts 1474 1528V. Free Radic Biol Med, 3: 259–303, 1987.

13.

Chénais

, Andriollo

, Guiraud

, Belhoussine

, and Jeannesson

. Oxidative stress involvement in chemically induced differentiation of K562 cells. Free Radic Biol Med, 28: 18–27, 2000.

14.

Chen

, Tian

, Shin

, and Yoon

. Fluorescent and luminescent probes for detection of reactive oxygen and nitrogen species. Chem Soc Rev, 40: 4783–4804, 2011.

15.

Cochemé

, Logan

, Prime

, Abakumova

, Quin

, McQuaker

, Patel

, Fearnley

, James

, Porteous

, Smith

RAJ

, Hartley

, Partridge

, and Murphy

. Using the mitochondria-targeted ratiometric mass spectrometry probe MitoB to measure H₂O₂ in living Drosophila. Nat Protoc, 7: 946–958, 2012.

16.

Cramb

and Beck

. Fluorescence quenching mechanisms in micelles: The effect of high quencher concentration. J Photochem Photobiol Chem, 134: 87–95, 2000.

17.

Crow

. Dichlorodihydrofluorescein and dihydrorhodamine 123 are sensitive indicators of peroxynitrite in vitro: Implications for intracellular measurement of reactive nitrogen and oxygen species. Nitric Oxide, 1: 145–157, 1997.

18.

Debowska

, Debski

, Hardy

, Jakubowska

, Kalyanaraman

, Marcinek

, Michalski

, Michalowski

, Ouari

, Sikora

, Smulik

, and Zielonka

. Toward selective detection of reactive oxygen and nitrogen species with the use of fluorogenic probes—Limitations, progress, and perspectives. Pharmacol Rep, 67: 756–764, 2015.

19.

DeLeon

, Gao

, Huang

, Arif

, Arora

, Divietro

, Patel

, and Olson

. A case of mistaken identity: Are reactive oxygen species actually reactive sulfide species?. Am. J. Physiol Regul Integr Comp Physiol, 310: R549–R560, 2016.

20.

Denicola

, Souza

, Radi

, and Lissi

. Nitric oxide diffusion in membranes determined by fluorescence quenching. Arch Biochem Biophys, 328: 208–212, 1996.

21.

Dickinson

and Chang

. A targetable fluorescent probe for imaging hydrogen peroxide in the mitochondria of living cells. J Am Chem Soc, 130: 9638–9639, 2008.

22.

Dickinson

, Srikun

, and Chang

. Mitochondrial-targeted fluorescent probes for reactive oxygen species. Curr Opin Chem Biol, 14: 50–56, 2010.

23.

Dikalov

, Griendling

, and Harrison

. Measurement of reactive oxygen species in cardiovascular studies. Hypertension, 49: 717–727, 2007.

24.

Dikalov

and Harrison

. Methods for detection of mitochondrial and cellular reactive oxygen species. Antioxid Redox Signal, 20: 372–382, 2014.

25.

Dikalov

, Kirilyuk

, Voinov

, and Grigor'ev

. EPR detection of cellular and mitochondrial superoxide using cyclic hydroxylamines. Free Radic Res, 45: 417–430, 2011.

26.

El Fangour

, Marini

, Good

, McQuaker

, Shiels

, and Hartley

. Nitrones for understanding and ameliorating the oxidative stress associated with aging. Age (Dordr), 31: 269–276, 2009.

27.

Fan

and Li

J-M

. Evaluation of methods of detecting cell reactive oxygen species production for drug screening and cell cycle studies. J Pharmacol Toxicol Methods, 70: 40–47, 2014.

28.

Faulkner

and Fridovich

. Luminol and lucigenin as detectors for O2.-. Free Radic Biol Med, 15: 447–451, 1993.

29.

Forman

, Augusto

, Brigelius-Flohe

, Dennery

, Kalyanaraman

, Ischiropoulos

, Mann

, Radi

, Roberts

II , Vina

, and Davies

KJA

. Even free radicals should follow some rules: A guide to free radical research terminology and methodology. Free Radic Biol Med, 78: 233–235, 2015.

30.

Giandomenico

, Cerniglia

, Biaglow

, Stevens

, and Koch

. The importance of sodium pyruvate in assessing damage produced by hydrogen peroxide. Free Radic Biol Med, 23: 426–434, 1997.

31.

Gomes

, Fernandes

, and Lima

. Fluorescence probes used for detection of reactive oxygen species. J Biochem Biophys Methods, 65: 45–80, 2005.

32.

Gorris

and Walt

. Mechanistic aspects of horseradish peroxidase elucidated through single-molecule studies. J Am Chem Soc, 131: 6277–6282, 2009.

33.

Guo

, Aleyasin

, Dickinson

, Haskew-Layton

, and Ratan

. Recent advances in hydrogen peroxide imaging for biological applications. Cell Biosci, 4: 64, 2014.

34.

Halliwell

and Gutteridge

. Free Radical in Biology and Medicine, 4th ed. New York: Oxford Press, 2007.

35.

Han

, Hong

, and Oh

K-W

. In vivo electron spin resonance: An effective new tool for reactive oxygen species/reactive nitrogen species measurement. Arch Pharm Res, 33: 1293–1299, 2010.

36.

Hardy

, Poulhes

, Rizzato

, Rockenbauer

, Banaszak

, Karoui

, Lopez

, Zielonka

, Vasquez-Vivar

, Sethumadhavan

, Kalyanaraman

, Tordo

, and Ouari

. Mitochondria-targeted spin traps: Synthesis, superoxide spin trapping, and mitochondrial uptake. Chem Res Toxicol, 27: 1155–1165, 2014.

37.

Hawkins

and Davies

. Detection and characterisation of radicals in biological materials using EPR methodology. Biochim Biophys Acta, 1840: 708–721, 2014.

38.

, Liu

, Wamer

, and Yin

J-J

. Electron spin resonance spectroscopy for the study of nanomaterial-mediated generation of reactive oxygen species. J Food Drug Anal, 22: 49–63, 2014.

39.

Horton

, Stewart

, Fonseca

, Guo

, and Kelley

. Mitochondria-penetrating peptides. Chem Biol, 15: 375–382, 2008.

40.

Hoye

, Davoren

, Wipf

, Fink

, and Kagan

. Targeting mitochondria. Acc Chem Res, 41: 87–97, 2008.

41.

Ishii

, Hayashi

, Suzuki

, and Hirano

. Chemiluminescent 2,6-diphenylimidazo[1,2-a]pyrazin-3(7H)-ones: a new entry to Cypridina luciferin analogues. Photochem Photobiol Sci, 13: 182–189, 2014.

42.

Kalyanaraman

, Darley-Usmar

, Davies

KJA

, Dennery

, Forman

, Grisham

, Mann

, Moore

, Roberts

, and Ischiropoulos

. Measuring reactive oxygen and nitrogen species with fluorescent probes: Challenges and limitations. Free Radic Biol Med, 52: 1–6, 2012.

43.

Kambayashi

and Ogino

. Reestimation of Cypridina luciferin analogs (MCLA) as a chemiluminescence probe to detect active oxygen species—Cautionary note for use of MCLA. J Toxicol Sci, 28: 139–148, 2003.

44.

Khan

, Idrees

, Moxley

, Corbett

, Ahmad

, von Figura

, Sly

, Waheed

, and Hassan

. Luminol-based chemiluminescent signals: Clinical and non-clinical application and future uses. Appl Biochem Biotechnol, 173: 333–355, 2014.

45.

Kim

, Anderson

, and Dryer

. Insulin increases surface expression of TRPC6 channels in podocytes: Role of NADPH oxidases and reactive oxygen species. Am J Physiol Renal Physiol, 302: F298–F307, 2012.

46.

Kleschyov

and Terekhov

. Electron paramagnetic resonance in a biomedical laboratory. Bioanalysis, 5: 2233–2237, 2013.

47.

Koide

, Urano

, Kenmoku

, Kojima

, and Nagano

. Design and synthesis of fluorescent probes for selective detection of highly reactive oxygen species in mitochondria of living cells. J Am Chem Soc, 129: 10324–10325, 2007.

48.

Kuznetsov

, Kehrer

, Kozlov

, Haller

, Redl

, Hermann

, Grimm

, and Troppmair

. Mitochondrial ROS production under cellular stress: Comparison of different detection methods. Anal Bioanal Chem, 400: 2383–2390, 2011.

49.

Lin

, Dickinson

, and Chang

. Boronate-based fluorescent probes: Imaging hydrogen peroxide in living systems. Methods Enzymol, 526: 19–43, 2013.

50.

and Trush

. Diphenyleneiodonium, an NAD(P)H oxidase inhibitor, also potently inhibits mitochondrial reactive oxygen species production. Biochem Biophys Res Commun, 253: 295–299, 1998.

51.

, Zhu

, and Trush

. Detection of mitochondria-derived reactive oxygen species production by the chemilumigenic probes lucigenin and luminol. Biochim Biophys Acta, 1428: 1–12, 1999.

52.

Maeda

. Which are you watching, an individual reactive oxygen species or total oxidative stress?. Ann N Y Acad Sci, 1130: 149–156, 2008.

53.

Maghzal

, Krause

K-H

, Stocker

, and Jaquet

. Detection of reactive oxygen species derived from the family of NOX NADPH oxidases. Free Radic Biol Med, 53: 1903–1918, 2012.

54.

Mahfouz

, Sharma

, Lackner

, Aziz

, and Agarwal

. Evaluation of chemiluminescence and flow cytometry as tools in assessing production of hydrogen peroxide and superoxide anion in human spermatozoa. Fertil Steril, 92: 819–827, 2009.

55.

Mishin

, Gray

, Heck

, Laskin

, and Laskin

. Application of the Amplex red/horseradish peroxidase assay to measure hydrogen peroxide generation by recombinant microsomal enzymes. Free Radic Biol Med, 48: 1485–1491, 2010.

56.

Mohanty

, Jaffe

, Schulman

, and Raible

. A highly sensitive fluorescent micro-assay of H₂O₂ release from activated human leukocytes using a dihydroxyphenoxazine derivative. J Immunol Methods, 202: 133–141, 1997.

57.

Mone

, Ribou

A-C

, Cosseau

, Duval

, Theron

, Mitta

, and Gourbal

. An example of molecular co-evolution: Reactive oxygen species (ROS) and ROS scavenger levels in Schistosoma mansoni/Biomphalaria glabrata interactions. Int J Parasitol, 41: 721–730, 2011.

58.

Mouche

, Mkaddem

, Wang

, Katic

, Tseng

Y-H

, Carnesecchi

, Steger

, Foti

, Meier

, Muzzin

, Kahn

, Ogier-Denis

, and Szanto

. Reduced expression of the NADPH oxidase NOX4 is a hallmark of adipocyte differentiation. Biochim Biophys Acta, 1773: 1015–1027, 2007.

59.

Mukhopadhyay

, Rajesh

, Haskó

, Hawkins

, Madesh

, and Pacher

. Simultaneous detection of apoptosis and mitochondrial superoxide production in live cells by flow cytometry and confocal microscopy. Nat Protoc, 2: 2295–2301, 2007.

60.

Murphy

. How mitochondria produce reactive oxygen species. Biochem J, 417: 1–13, 2009.

61.

Murphy

, Holmgren

, Larsson

N-G

, Halliwell

, Chang

, Kalyanaraman

, Rhee

, Thornalley

, Partridge

, Gems

, Nyström

, Belousov

, Schumacker

, and Winterbourn

. Unravelling the biological roles of reactive oxygen species. Cell Metab, 13: 361–366, 2011.

62.

Nauseef

. Detection of superoxide anion and hydrogen peroxide production by cellular NADPH oxidases. Biochim Biophys Acta, 1840: 757–767, 2014.

63.

Nelder

and Mead

. A simplex method for function minimization. Comput J, 7: 308–313, 1965.

64.

Nishinaka

, Aramaki

, Yoshida

, Masuya

, Sugawara

, and Ichimori

. A new sensitive chemiluminescence probe, L-012, for measuring the production of superoxide anion by cells. Biochem Biophys Res Commun, 193: 554–559, 1993.

65.

Ohashi

, Mizutani

, Murakami

, Kojo

, Ishii

, and Taketani

. Rapid oxidation of dichlorodihydrofluorescin with heme and hemoproteins: Formation of the fluorescein is independent of the generation of reactive oxygen species. FEBS Lett, 511: 21–27, 2002.

66.

Oter

and Ribou

. Quenching of long lifetime emitting fluorophores with paramagnetic molecules. J Fluoresc, 19: 389–397, 2009.

67.

Reinhardt

and Ribou

. Females become infertile as the stored sperm's oxygen radicals increase. Sci Rep. 3, 2013.

68.

Rembish

and Trush

. Further evidence that lucigenin-derived chemiluminescence monitors mitochondrial superoxide generation in rat alveolar macrophages. Free Radic Biol Med, 17: 117–126, 1994.

69.

Rharass

, Gbankoto

, Canal

, Kurşunluoğlu

, Bijoux

, Panáková

, and Ribou

A-C

. Oxidative stress does not play a primary role in the toxicity induced with clinical doses of doxorubicin in myocardial H9c2 cells. Mol Cell Biochem, 413: 199–215, 2016.

70.

Rharass

, Vigo

, Salmon

, and Ribou

. Variation of 1-pyrenebutyric acid fluorescence lifetime in single living cells treated with molecules increasing or decreasing reactive oxygen species levels. Anal Biochem, 357: 1–8, 2006.

71.

Rharass

, Vigo

, Salmon

, and Ribou

. New method for the detection of reactive oxygen species in anti-tumoural activity of adriamycin: A comparison between hypoxic and normoxic cells. Free Radic Res, 42: 124–134, 2008.

72.

Rhee

, Chang

T-S

, Jeong

, and Kang

. Methods for detection and measurement of hydrogen peroxide inside and outside of cells. Mol Cells, 29: 539–549, 2010.

73.

Ribou

and Reinhardt

. Reduced metabolic rate and oxygen radicals production in stored insect sperm. Proc Biol Sci, 279: 2196–2203, 2012.

74.

Ribou

, Rharass

, and Vigo

. Looking through the microscope: What can we learn from fluorescence lifetime measurements about free radicals in living cells? In: Handbook of Free Radicals: Formation, Types and Effects, edited by Kozyrev

, and Slutsky

. New York: Nova Science Publishers, 2010, pp. 505–521.

75.

Ribou

A-C

, Vigo

, and Salmon

. Synthesis and characterization of (1’-pyrene butyl)-2-rhodamine ester: a new probe for oxygen measurement in the mitochondria of living cells. J Photochem Photobiol Chem, 151: 49–55, 2002.

76.

Ribou

, Vigo

, and Salmon

. Lifetime of fluorescent pyrene butyric acid probe in single living cells for measurement of oxygen fluctuation. Photochem Photobiol, 80: 274–280, 2004.

77.

Rin Jean

, Tulumello

, Wisnovsky

, Lei

, Pereira

, and Kelley

. Molecular vehicles for mitochondrial chemical biology and drug delivery. ACS Chem Biol, 9: 323–333, 2014.

78.

Robinson

, Janes

, and Beckman

. The selective detection of mitochondrial superoxide by live cell imaging. Nat Protoc, 3: 941–947, 2008.

79.

Robinson

, Janes

, Pehar

, Monette

, Ross

, Hagen

, Murphy

, and Beckman

. Selective fluorescent imaging of superoxide in vivo using ethidium-based probes. Proc Natl Acad Sci U S A, 103: 15038–15043, 2006.

80.

Ross

, Kelso

, Blaikie

, James

, Cochemé

, Filipovska

, Da Ros

, Hurd

, Smith

, and Murphy

. Lipophilic triphenylphosphonium cations as tools in mitochondrial bioenergetics and free radical biology. Biochemistry (Mosc), 70: 222–230, 2005.

81.

Rota

, Fann

, and Mason

. Phenoxyl free radical formation during the oxidation of the fluorescent dye 2’,7’-dichlorofluorescein by horseradish peroxidase. Possible consequences for oxidative stress measurements. J Biol Chem, 274: 28161–28168, 1999.

82.

Royall

and Ischiropoulos

. Evaluation of 2′,7′-dichlorofluorescin and dihydrorhodamine 123 as fluorescent probes for intracellular H₂O₂ in cultured endothelial cells. Arch Biochem Biophys, 302: 348–355, 1993.

83.

Sarder

, Maji

, and Achilefu

. Molecular probes for fluorescence lifetime imaging. Bioconjug Chem, 26: 963–974, 2015.

84.

Savatier

, Rharass

, Canal

, Gbankoto

, Vigo

, Salmon

, and Ribou

. Adriamycin dose and time effects on cell cycle, cell death, and reactive oxygen species generation in leukaemia cells. Leuk Res, 36: 791–798, 2012.

85.

Sieracki

, Gantner

, Mao

, Horner

, Ye

, Malik

, Newcomb

, and Bonini

. Bioluminescent detection of peroxynitrite with a boronic acid-caged luciferin. Free Radic Biol Med, 61: 40–50, 2013.

86.

Smith

RAJ

, Hartley

, and Murphy

. Mitochondria-targeted small molecule therapeutics and probes. Antioxid Redox Signal, 15: 3021–3038, 2011.

87.

Södergren

, Nourooz-Zadeh

, Berglund

, and Vessby

. Re-evaluation of the ferrous oxidation in xylenol orange assay for the measurement of plasma lipid hydroperoxides. J Biochem Biophys Methods, 37: 137–146, 1998.

88.

Soh

. Recent advances in fluorescent probes for the detection of reactive oxygen species. Anal Bioanal Chem, 386: 532–543, 2006.

89.

Song

, Al-Mehdi

, and Fisher

. An immediate endothelial cell signaling response to lung ischemia. Am J Physiol Lung Cell Mol Physiol, 281: L993–L1000, 2001.

90.

Stolze

, Rohr-Udilova

, Rosenau

, Stadtmüller

, and Nohl

. Very stable superoxide radical adducts of 5-ethoxycarbonyl-3,5-dimethyl-pyrroline N-oxide (3,5-EDPO) and its derivatives. Biochem Pharmacol, 69: 1351–1361, 2005.

91.

Swannell

RPJ

, Caplin

, Nedwell

, and Williamson

. An investigation of hydroethidine as a fluorescent vital stain for prokaryotes. FEMS Microbiol Lett, 101: 173–182, 1992.

92.

Swartz

, Khan

, and Khramtsov

. Use of electron paramagnetic resonance spectroscopy to evaluate the redox state in vivo. Antioxid Redox Signal, 9: 1757–1771, 2007.

93.

Szeto

. Cell-permeable, mitochondrial-targeted, peptide antioxidants. AAPS J, 8: E277–E283, 2006.

94.

Tarpey

, Wink

, and Grisham

. Methods for detection of reactive metabolites of oxygen and nitrogen: in vitro and in vivo considerations. Am J Physiol Regul Integr Comp Physiol, 286: R431–R444, 2004.

95.

Van de Bittner

GCV de

, Dubikovskaya

, Bertozzi

, and Chang

. In vivo imaging of hydrogen peroxide production in a murine tumor model with a chemoselective bioluminescent reporter. Proc Natl Acad Sci U S A, 107: 21316–21321, 2010.

96.

Vaughan

and Weber

. Oxygen quenching of pyrenebutyric acid fluorescence in water. A dynamic probe of the microenvironment. Biochemistry (Mosc), 9: 464–473, 1970.

97.

Villamena

and Zweier

. Detection of reactive oxygen and nitrogen species by EPR spin trapping. Antioxid Redox Signal, 6: 619–629, 2004.

98.

Vladimirov

and Proskurnina

. Free radicals and cell chemiluminescence. Biochemistry (Mosc), 74: 1545–1566, 2009.

99.

Wardman

. Fluorescent and luminescent probes for measurement of oxidative and nitrosative species in cells and tissues: progress, pitfalls, and prospects. Free Radic Biol Med, 43: 995–1022, 2007.

100.

Woolley

, Stanicka

, and Cotter

. Recent advances in reactive oxygen species measurement in biological systems. Trends Biochem Sci, 38: 556–565, 2013.

101.

Yousif

, Stewart

, Horton

, and Kelley

. Mitochondria-penetrating peptides: Sequence effects and model cargo transport. Chembiochem, 10: 2081–2088, 2009.

102.

Zhang

and Gao

. Imaging mitochondrial reactive oxygen species with fluorescent probes: Current applications and challenges. Free Radic Res, 49: 374–382, 2015.

103.

Zhao

, Summers

, and Mason

. Photooxidation of Amplex red to resorufin: Implications of exposing the Amplex red assay to light. Free Radic Biol Med, 53: 1080–1087, 2012.

104.

Zhao

, Kalivendi

, Zhang

, Joseph

, Nithipatikom

, Vásquez-Vivar

, and Kalyanaraman

. Superoxide reacts with hydroethidine but forms a fluorescent product that is distinctly different from ethidium: Potential implications in intracellular fluorescence detection of superoxide. Free Radic Biol Med, 34: 1359–1368, 2003.

105.

Zhao

, Joseph

, Fales

, Sokoloski

, Levine

, Vasquez-Vivar

, and Kalyanaraman

. Detection and characterization of the product of hydroethidine and intracellular superoxide by HPLC and limitations of fluorescence. Proc Natl Acad Sci U S A, 102: 5727–5732, 2005.

106.

Zhou

, Diwu

, Panchuk-Voloshina

, and Haugland

. A stable nonfluorescent derivative of resorufin for the fluorometric determination of trace hydrogen peroxide: Applications in detecting the activity of phagocyte NADPH oxidase and other oxidases. Anal Biochem, 253: 162–168, 1997.

107.

Zielonka

and Kalyanaraman

. Hydroethidine- and MitoSOX-derived red fluorescence is not a reliable indicator of intracellular superoxide formation: Another inconvenient truth. Free Radic Biol Med, 48: 983–1001, 2010.

108.

Zielonka

, Lambeth

, and Kalyanaraman

. On the use of L-012, a luminol-based chemiluminescent probe, for detecting superoxide and identifying inhibitors of NADPH oxidase: A re-evaluation. Free Radic Biol Med, 65: 1310–1314, 2013.

109.

Zielonka

, Sikora

, Hardy

, Joseph

, Dranka

, and Kalyanaraman

. Boronate probes as diagnostic tools for real time monitoring of peroxynitrite and hydroperoxides. Chem Res Toxicol, 25: 1793–1799, 2012.

110.

Zielonka

, Vasquez-Vivar

, and Kalyanaraman

. Detection of 2-hydroxyethidium in cellular systems: A unique marker product of superoxide and hydroethidine. Nat Protoc, 3: 8–21, 2008.

Synthetic Sensors for Reactive Oxygen Species Detection and Quantification: A Critical Review of Current Methods

Abstract

Introduction

Intensity-Based Fluorescent Synthetic Sensors—Oxidant-Sensitive Probes

DHE (hydroethidine)

Fluorescence-based techniques

Stability of DHE

Practical recommendations

Amplex Red

Cells that do not release H2O2

Practical recommendations

Chemiluminescent sensors

Alternative Techniques for Detection of Free Radicals

Electron paramagnetic resonance

Time-resolved microfluorimetry

Successful applications

Fluorescence-based techniques

Practical recommendations

Mitochondrial Targeting Sensors

Lipophilic cationic compounds

Triphenylphosphonium-based probes

Lipophilic cations

Peptidic delivery agents

Conclusions

Footnotes

Abbreviations Used

References

Cells that do not release H₂O₂