Chapter 1: Overview of Reactive Oxygen Species

The increasing concentration of molecular oxygen (O₂) in the atmosphere roughly 2.5 billion years ago,^1,2  due to oxygenic photosynthesis by cyanobacteria, allowed for the evolution of aerobic respiration, leading to the development of complex eukaryotic organisms.²  For all currently living aerobic species, molecular oxygen is a central molecule in cellular respiration. Certain derivatives of oxygen are, however, highly toxic to cells. In the 1950s, Gerschman et al. proposed that oxygen-containing free radicals were responsible for toxic effects in aerobic organisms.^3,4  Over the years, the terms ROS (reactive oxygen species), ROI (reactive oxygen intermediates) and RNS (reactive nitrogen species) have been coined to define an emerging class of endogenous, highly reactive, oxygen- (and also nitrogen-) bearing molecules. According to some definitions the term ROI describes the chemical species formed upon incomplete reduction of molecular oxygen, namely superoxide radical anion (O₂˙⁻), hydrogen peroxide (H₂O₂), and hydroxyl radicals (OH˙), while ROS includes both ROI and ozone (O₃) and singlet oxygen (¹O₂).⁵  A somewhat more encompassing definition also includes within ROS compounds such as hypochlorous (HOCl), hypobromous (HOBr), and hypoiodous acids (HOI). Incorporation of peroxyl (ROO˙), alkoxyl (RO˙), semiquinone (SQ˙⁻) and carobonate (CO₃˙⁻) radicals and organic hydroperoxides (ROOH) is also frequently encountered within the definition of ROS.^6,7  ROS may also be classified as free radicals and nonradical species.^7,8  RNS that bear oxygen atoms include nitric oxide radical (NO or NO˙), nitrogen dioxide radical (NO₂˙), nitrite (NO₂⁻), and peroxynitrite (ONOO⁻).⁵ 

Reactive oxygen species, in particular hydroxyl and peroxyl radicals, hydrogen peroxide and superoxide radical anion, have long been implicated in oxidative damage inflicted on fatty acids, DNA and proteins as well as other cellular components.⁹  ROS overproduction is associated with numerous disorders.¹⁰  Oxidative stress caused by the imbalance between excessive formation of ROS and limited antioxidant defences is connected to many pathologies including age-related disorders, cancer, cardiovascular, inflammatory, and neurodegenerative diseases such as Parkinson’s and Alzheimer’s diseases.^11–14  According to the long-held “free radical theory of aging”^15,16  advanced by Denham Harman in 1956, the noxious effects of ROS, generated during cellular respiration at the mitochondrial level are directly involved in aging processes. This hypothesis is, however, currently under revision.⁹  Mounting evidence suggests that ROS actually may have a beneficial physiological role acting as messengers in cellular signaling, a new paradigm in the rich and diverse chemistry of ROS which has attracted increased attention in the last decade (see also Figure 1.1).^{2,8,9,12,17–20}  The redox regulation typically involves controlled production of reactive oxygen and nitrogen species. They can, in turn, react with specific functional groups of target proteins (e.g. [Fe–S] clusters, cysteines, etc.) that lead to covalent protein modifications.²¹  ROS as second messengers are important for the expression of several transcription factors and other signal transduction molecules such as heat shock-inducing factor and nuclear factor. They also participate in the regulation of cell adhesion, redox-mediated amplification of immune response and programmed cell death.⁸ 

Deciphering the highly complex and diverse impact of ROS chemistry in biological environments requires a multidisciplinary approach where chemists may be actively involved providing new tools to generate and detect ROS and their byproducts with spatiotemporal control and precision. Knowledge is, however, required on the chemical properties of these species, their sources both endogenous and exogenous, their typical scavengers and the characteristic lifetimes in order to understand how, when and where they form, how far they travel, what are their targets and how we may control their activity. The following sections summarize current literature on these topics providing a glimpse of what we believe is a most intriguing, challenging and stimulating contemporary problem, that of the biological impact of ROS chemistry.

In order to understand the chemical and the associated biological impact of ROS, knowledge on their origin, type, and reactivity (specifically, their electronic configuration and redox potential) is required.

Ground-state triplet molecular oxygen is a paramagnetic biradical with two electrons occupying separate π* orbitals with parallel spins (Figure 1.2). Most nonradical organic molecules are diamagnetic, with pairs of electrons with opposite spins. A spin restriction applies for molecular oxygen to participate in redox reactions with other atoms or molecules as it has to accept, from the reductant, a pair of electrons that have the same spin (i.e. nondiamagnetic) so they can fit into the vacant spaces in the π* orbitals of oxygen. Oxygen is thus unable to efficiently oxidize biomolecules via e.g. addition (2-electron process).

The spin restriction results in oxygen preferably accepting one electron at a time during redox reactions. Thus molecular oxygen can react fast with other radicals by single-electron transfer. It may also react with other species bearing unpaired electrons e.g. transition metals such as Fe found in [Fe–S] clusters.²²  The one-electron reduction of oxygen results in the formation of superoxide radical anion (O₂˙⁻). One-electron reduction of O₂˙⁻ leads to the formation of other ROS such as hydrogen peroxide (H₂O₂) that is a closed-shell molecule (Scheme 1.1, also Figure 1.2). Reduction of hydrogen peroxide in turn yields the hydroxyl radical (OH˙) that undergoes reduction to yield water (or hydroxide OH⁻).

Ground-state or molecular oxygen O₂ can be, however, converted to more reactive oxygen containing forms. Energy transfer to O₂ leads to the formation of the more reactive molecular oxygen form, singlet oxygen (¹O₂), amply discussed in the following chapters. Singlet oxygen has paired electrons with opposite spins (Figure 1.2). Thus the spin restriction is removed, increasing the oxidizing ability of ¹O₂ compared to ground-state O₂.

Knowledge of the thermodynamics of free radical reactions is necessary towards understanding the direction of the electron transfer. Redox potentials of the various ROS intermediates involved in the reduction of molecular oxygen to water are listed in Table 1.1, see also Scheme 1.1. Table 1.1 additionally lists, organized from highly oxidizing to highly reducing, the one-electron redox potentials of various other ROS of biological importance, as well as the one electron redox potential of ROS scavengers, as originally compiled by Buettner.²³ 

The biradical nature of oxygen restricts it to accepting electrons one at a time during a redox reaction with spin-paired molecules (see above). Molecular oxygen, with a redox potential of −0.16 V (for oxygen concentration of 1 M, pH 7 as the standard state, −0.33 V for 1 bar as the standard state²⁴ ), is, however, a poor univalent electron acceptor (see Scheme 1.1).²²  Consequently, molecular oxygen itself is a poor oxidant and is fairly benign to biomolecules. The unpaired electrons of oxygen may, however, interact with unpaired electrons of transition metals and organic radicals. While one-electron reduction of molecular oxygen to superoxide radical anion is thermodynamically less favorable than its direct two-electron reduction to hydrogen peroxide (+0.30 V), the simultaneous 2e⁻ requirement of the latter is unfavorable.²⁵  Formation of superoxide radical anion but not hydrogen peroxide is thus characteristic of auto-oxidative processes.

The superoxide radical anion has limited reactivity with electron-rich centers because of its anionic charge.²⁶  Upon protonation of O₂˙⁻ the perhydroxyl radical is obtained (pK_a = 4.8²⁷ ). The new species has an increased reduction potential (+1.06 V²³ ) and is a better oxidant. The biological relevance of the perhydroxyl radical is, however, believed to be minor given its low concentration at physiological pH. Despite its high reduction potential of +0.94 V, O₂˙⁻ can oxidize very few biological compounds. One-electron reduction of O₂˙⁻ leads to the formation of hydrogen peroxide.

Although hydrogen peroxide has a positive one electron reduction potential (+0.32 V²³  to +0.38 V^22,25  based on the source), and an even more favorable two-electron reduction potential (+1.35 V²⁵ ), it is relatively stable under physiological conditions (slow reaction). In stark contrast, the hydroxyl radical, with a one-electron reduction potential of +2.33 V, is a most powerful oxidant reacting at diffusion control rates with organic matter.²⁸ 

ROS can either be generated exogenously or intracellularly from numerous sources. They are produced in a wide range of biochemical and physiological processes (Figure 1.1). Several different enzymes have been implicated in the generation of ROS.

Cytosolic enzyme systems contributing to the generation of ROS, among others, are the seven isoforms of the expanding family of transmembrane NADPH oxidases (NOXs), a superoxide-generating system.^29,30  The cytosolic domains of NOX transfer an electron from NADPH to a FAD cofactor. From there, the electron is passed to a haem group, which donates it to O₂ on the extracellular side of the membrane, generating O₂˙⁻.³⁰  Depending on the specific NADPH oxidase expressed in different cells, they can trigger different cellular transformations with widely differing biological outcomes. The NADPH oxidase family of enzymes illustrates the specificity in ROS generation and its impact on normal cellular signaling and homeostasis.³⁰ 

Mitochondria represent another major source for intracellular ROS production. The production of mitochondrial superoxide radicals occurs primarily at two discrete points in the electron-transport chain, namely at Complex I (NADH dehydrogenase) and at Complex III (ubiquinone–cytochrome c reductase) upon one electron transfer to oxygen.^31,32  In vitro, these two sites in mitochondria convert 1–2% of the consumed oxygen molecules into superoxide anions both under normobaric or hyperbaric conditions.^33–35  These initial estimates were made on isolated mitochondria and it may be concluded that the in vivo rate of mitochondrial superoxide production is considerably less.¹⁰  Although one-electron reactions predominate, two-electron reactions that allow the direct reduction of molecular oxygen to hydrogen peroxide do exist within the mitochondria.³⁶  Superoxide produced at Complex I is thought to form only within the matrix, whereas at Complex III superoxide is released both into the matrix and the inner mitochondrial space (IMS). A nonenzymatic source of ROS in mitochondria is the formation of the free radical semiquinone anion species (Q˙⁻) that occurs as an intermediate in the redox cycling of coenzyme Q₁₀.¹⁰  Once formed, Q˙⁻ can readily transfer electrons to molecular oxygen with the subsequent generation of a superoxide radical. The generation of ROS therefore becomes predominantly a function of metabolic rate.

In addition to the mitochondria and NADPH oxidases, other cellular sources of ROS production include a number of intracellular enzymes such as the flavoenzyme ERO1 in the endoplasmic reticulum, xanthine oxidase, cyclo-oxygenases, cytochrome p450 enzymes, lipoxygenases, flavin-dependent demethylase, oxidases for polyamines and amino acids, and nitric oxide synthases that produce oxidants as part of their normal enzymatic function.^30,32  Free copper ions or iron ions that are released from iron–sulfur clusters, haem groups or metal-storage proteins can convert O₂˙⁻ and/or H₂O₂ to OH˙ in what is known as the Fenton reaction (Scheme 1.2).^37–39  A similar reaction but involving lipid hydroperoxides accounts for the formation of lipid alkoxyl (LO˙) and peroxyl radicals (LOO˙) in the lipid membrane.⁷ 

We next briefly mention biological sources of singlet oxygen, discussed in the following chapters. A possible pathway of cellular singlet oxygen formation is from oxygen in areas of inflammation through the action of Phox (NOX of phagocytes mainly in neutrophils and macrophages) and the oxidation of halide ions by the phagocyte enzyme myeloperoxidase (MPO).²⁹  Additionally, superoxide and NO are readily converted by enzymes or nonenzymic chemical reactions into reactive nonradical species amongst which are singlet oxygen, hydrogen peroxide, or peroxynitrite (ONOO₂).¹² 

Reactive oxygen species can be produced by a host of exogenous processes. Environmental sources include ultraviolet light, ionizing radiation, and pollutants. Amongst the pollutants are chemicals (e.g. paraquat, also named methyl viologen) that react to form either peroxides or ozone; chemicals that promote the formation of superoxide such as quinones, nitroaromatics, and bipyrimidiulium herbicides (related to paraquat); chemicals that are metabolized to radicals, e.g., polyhalogenated alkanes, phenols, aminophenols; or chemicals that release iron and copper that could promote the formation of hydroxyl radicals.^8–10 

ROS could be generated rapidly through radiolysis of water molecules upon ionizing radiation (X-rays, γ-rays) or UV-light irradiation of H₂O₂.⁴⁰  Secondary ROS products generated through this method can potentially amplify the initial ionization event. However, theoretical calculations show that hydrogen peroxide or superoxide anion are generated in very low concentrations by the primary ionization event.⁴⁰  In the presence of a sensitizer UV radiation could additionally lead to the formation of singlet oxygen, amply discussed in this book.

Environmental agents including non-DNA reactive carcinogens can generate ROS in cells by metabolism to primary radical intermediates or by activating endogenous sources of reactive oxygen species.⁴¹  The induction of oxidative stress and damage has been observed following exposure to xenobiotics of varied structures and activities. Chlorinated compounds, radiation, metal ions, barbiturates, phorbol esters, and some peroxisome-proliferating compounds are among the classes of compounds that induce oxidative stress and damage in vitro and in vivo.⁴¹ 

The mechanism of action of many chemotherapeutic cancer drugs involves ROS-mediated apoptosis. For example, the classic antitumor drugs cisplatin and adriamycin appear to produce ROS at excessive levels, resulting in DNA damage and cell death.⁴²  Some classes of antibiotics rely on a similar mechanism for their bactericidal activity. For example, it was recently shown that bactericidal antibiotics, regardless of drug–target interaction, induce a breakdown in iron regulatory dynamics, stimulating the production of highly deleterious hydroxyl radicals through Fenton reaction in gram-negative and gram-positive bacteria, which ultimately contribute to cell death.⁴³ 

ROS are oxidant species that can operate via one-electron oxidation (radical ROS species) or two-electron oxidation (nonradical ROS species).⁷  In the former case reactivity is strongly linked to thermodynamics as activation barriers for the one-electron reaction of radicals are expected to be low.⁸  One may then utilize Table 1.1 in estimating the reactivity for different ROS. Radical ROS species are typically initiators or chain propagators in chain reactions; a notable example is the free radical-mediated auto-oxidation of polyunsaturated fatty acids (PUFA) that relies on lipid peroxyl radicals as chain propagators.^44–50 

The reactivity of nonradical ROS species will on the other hand be strongly dependent on the activation barrier to the reaction of interest. Based on a scale of reactions with glutathione (GSH) one may observe that HOCl is more reactive than H₂O₂ (rate constants of 3 × 10⁷ M⁻¹ s⁻¹ and 0.9 M⁻¹ s⁻¹, respectively) albeit the redox potential for the 2-electron reduction is larger for the latter in forming water, than for the former in forming chloride.⁸ 

The high toxicity associated to ROS would imply that they are indiscriminate in choosing biological targets to react with, yet a closer look to their now accepted signaling role in cells would rather point to a well-orchestrated target choice. This paradox may be reconciled if one realizes the broad range of reactivities for the various species contained within the ROS family. Highly reactive ROS (e.g. OH˙) will not be selective and will have a broad range of nonspecific targets. They will further have extremely short lifetimes in solution. ROS characterized by a relatively low reactivity, such as H₂O₂ or O₂˙⁻, will in turn be relatively selective. Their chemical activity towards different substrates in competing reactions will be dictated by the interplay of substrate relative availability and relative rate constants of reaction.

Signaling by – low-reactivity – ROS involves reaction with only a few atomic elements within target macromolecules, and frequently with only a subset of these atoms within a given macromolecule, leading to covalent protein modification.²⁶  Accordingly, we may first discuss the atomic targets of ROS to then address molecular targets.²¹ 

Atomic targets: reaction of ROS with sulfur, found in methionine and cysteine is typically favored; also selenium found in seleno-cysteine is a commonly encountered ROS atomic target.³⁰  In both cases and given their redox potentials, reactions may be reversible. Another common atomic target is carbon either in nucleosides, or in aminoacids such as arginine, lysine, proline and threonine,⁵  as well as carbon in polyunsaturated fatty acids.⁴⁶  An additional element targeted by ROS is iron, where ROS may react with [Fe–S] clusters²²  and with iron within haem.

Hydrogen peroxide typically reacts with thiols in their anionic form, thus thiols with low pK_a are found to be more reactive. The solution pH may further tune the reactivity towards this ROS.^26,51–54  Within a specific protein, and given the range of pK_a different thios may have, as a result of close proximity to other functional groups, only a subset of the thiols may be involved in reaction with H₂O₂. This leads to very specific response to exposure to this ROS.⁵⁵  Rate constants of reactions may range from 2 × 10¹ M⁻¹ s⁻¹ for free Cys to 1 × 10⁶ M⁻¹ s⁻¹ for specific Cys residues within a protein.²⁶  Rate constant of reactions of H₂O₂ with [Fe–S] clusters in turn are relatively small, ca. 1 × 10² to 1 × 10³ M⁻¹ s⁻¹.²⁶ 

Superoxide radical anion in turn favors reaction with [Fe–S] clusters (it can achieve diffusion-controlled rates)²⁶  where the vulnerability of these groups is partly due to the favorable electrostatic interactions with O₂˙⁻.²²  The relative reactivities of H₂O₂ and O₂˙⁻ lead to preferred biological targets, exemplified by transcription factors SoxR ([Fe–S] cluster) and OxyR (Cys residue) in Escherichia Coli.^56,57  These factors are activated by O₂˙⁻ and H₂O₂, respectively, and they are involved in the expression of antioxidant enzymes. Here, SoxR regulates responses to O₂˙⁻,⁵⁸  and OxyR regulate responses to H₂O₂.^59,60 

Molecular targets, proteins: a large number of proteins are affected by ROS, where following ROS attack conformational changes take place that regulate protein activity. This is best exemplified by the ever-increasing⁶¹  list of proteins where Cys residues act as redox switches. Here, disulfide bond formation following oxidation of Cys residues may result in structural and associated activity changes.^59,62  Cys residue oxidation in phosphatases is an important target in biological systems as it affects protein phosphorylation and thus has a broad impact in the cell proteome. Oxidative stress in proteins leads to formation of carbonyl derivatives along their backbone, used as markers of general oxidative stress.^10,63 

Molecular targets, DNA: mitochondrial DNA is a major target of ROS given that mitochondria are the prevalent source of ROS within cells. This leads to compromised mitochondrial function. ROS reactions with DNA itself, rather than proteins, may serve to promote transcription.³⁰  Even if reactions with DNA may be a negligible part of the ROS reactions within cells, their impact is far reaching.⁸  Aging cells have an increased level of ROS-damaged nuclear DNA.¹⁰ 

Molecular targets, lipids: a significant body of work, both in model membrane systems and in live cells, has examined the role lipid peroxyl radicals play in damaging the cell lipid milieu. Autoxidation of polyunsaturated fatty acid residues is initiated by a free radical such as the hydroxyl radical, which upon reaction with fatty acids generate lipid carbon centered radicals (eqn (1.1), Scheme 1.3).^{44,47,64–66}  Lipid carbon centered radicals in turn readily trap molecular oxygen under physiological conditions to form lipid peroxyl radicals,^67,68  effective chain carriers in the lipid chain auto-oxidation (eqn (1.2) and (1.3), Scheme 1.3). In the oxidation process, fatty acyl chains mostly in their cis configuration are either converted to the trans configuration,^44,69–71  or form corresponding hydroperoxides and alcohols^44,72  or may fragment into electrophilic αβ-unsaturated aldehydes,^72,73  among others. Peroxidation and destruction of the cis double bonds may in turn lead to a reduction in the membrane fluidity⁷⁴  and appearance of liquid-order domains.⁷⁵  Auto-oxidation of polyunsaturated fatty acid residues ultimately generates a variety of secondary cytotoxic products that account for pathological effects, e.g., neurodegenerative diseases,¹³  atherosclerosis,⁴⁹  and cell apoptosis.⁷⁶  Importantly, polyunsaturated fatty acids within the inner mitochondrial membrane are particularly vulnerable to ROS elicited oxidative damage.¹³ 

Oxidative signaling pathways arise from the formation of electrophilic αβ-unsaturated aldehydes that may undergo reaction with nucleotides (indirect signaling).²⁰  Additional oxidative signaling pathways have been reported that involve cardiolipin peroxidation and release of proapoptotic factors from mitochondria,⁷⁹  as well as phosphatidyl serine (PS) oxidation in the plasma membrane leading to externalization and recognition of PS on the cell surface by phagocytes.⁷⁶ 

Enzymatic and nonenzymatic antioxidant systems in cells regulate the concentration of ROS. Notably, McCord and Fridovich in the late 1960s discovered a variety of enzymes that were found to be responsible for detoxification of oxygen in aerobes but were absent in anaerobes (leading to oxygen-induced damage in these organisms).^80,81  The finding of superoxide dismutase (SOD) was a landmark discovery in the field of free-radical biology.⁸²  The presence of such enzymes suggested that if the ROS were not scavenged, they would critically injure cells. SOD ensures that the level of O₂˙⁻ remains below 0.1 nM in E. coli.^39,83  Over the years, new examples of enzymatic antioxidant systems have emerged. They include catalase, and glutathione and NADH peroxidase. The latter ensure that the steady-state concentration of H₂O₂ within E. coli does not exceed 20 nM.³⁹ 

Examples of nonenzymatic antioxidants include glutathione, vitamin C (both water soluble) and α-tocopherol (lipid soluble, see also Scheme 1.3).^2,10,84  All three antioxidants scavenge free radicals. Ascorbic acid yields ascorbyl radicals that readily disproportionate so no secondary free-radical byproducts are formed. In the case of glutathione, the thyil radical formed may be a concern as it may react with lipids and proteins yet it readily forms a disulfide bond with the thiolate of a second glutathione molecule. The disulfide radical anion is next scavenged by oxygen yielding an inert disulfide.⁸ 

A member of the vitamin E family of compounds, α-tocopherol (TOH) has long been recognized as the most active naturally occurring lipid soluble antioxidant (Scheme 1.1 and Figure 1.1).^77,85  The paradigm of TOH antioxidant activity in auto-oxidation reactions has been laid out in a number of studies conducted over the past 30 years in homogeneous solution and in the presence of initiators (see Scheme 1.3, reactions 1.4 and 1.5). In a first elementary step TOH reacts with a peroxyl radical (ROO˙ or LOO˙) via H-atom transfer to yield a tocopheroxyl radical (TO˙, eqn (1.5)) and a hydroperoxide ROOH/LOOH. Following coupling of TO˙ to a second peroxyl radical a second chain-termination reaction occurs. TOH effectively terminates two chain reactions.⁷⁶  The tocopheroxyl radical may also be scavenged by ascorbate at the lipid water interface where ascorbate acts as the ultimate ROS sink.⁷⁶ 

An interesting discussion is how far a given ROS will diffuse on average before decay through a unimolecular process or upon scavenging by, e.g. an antioxidant or a target molecule. Given their low concentration in biological tissue, second order reactions involving encounter of 2 identical ROS such as lipid peroxyl radicals are rare. Mostly ROS decay via first-order or pseudofirst-order reactions (e.g. upon scavenging by ascorbate, glutathione, or an enzyme). The average lifetime τ (inverse of the experimental decay rate constant k_exp) may thus be obtained for a given ROS given the rate constant of unimolecular reaction (k⁰), the rate constant of reaction with scavenger (k_qS), and abundance of scavengers ([S]) (eqn (1.1)). In turn, one may next assume a freely diffusing ROS molecule (generally applicable for noncharged ROS) with an average diffusion coefficient D of 1 × 10⁻⁵ cm² (average value for a small molecule in water) and estimate via eqn (1.2) the mean square displacement <r²>^1/2 utilizing the average lifetime. Pryor⁸⁶  and Winterbourn⁸  provide relevant numbers to estimate the mean square displacement for a number of ROS. While OH˙ is scavenged within a few angstroms of its generation site, O₂˙⁻ and H₂O₂ may diffuse a few tens of micrometers (albeit O₂˙⁻ is charged and may not readily cross bilayers) thus exerting a long-range effect.

Valuable methods to study the generation and evolution of ROS and associated chemical processes in vitro and in vivo include HPLC, mass spectrometry, EPR (when dealing with free radicals) and other analytical procedures that provide information on the biological production of ROS by detecting specific products generated from the oxidation of protein, DNA, lipids, or other biomolecules.^46,87,88  These methods have the drawback of being generally destructive, some further lack the necessary sensitivity, and they may be limited to providing information on the products of reactions of ROS and not on the specific rate or location of ROS production.

Advances in fluorescence microscopy have allowed for the development of noninvasive tools that provide high specificity to a ROS species and sensitivity combined with spatial and temporal resolution for imaging ROS evolution in live cells. The new probes enable monitoring ROS in biological systems and correlating their sites of production to important physiological processes.^89–98  Specificity to a particular type of ROS is of high importance for the design of successful probes. The ideal molecular probe for ROS would also be highly reactive at low concentrations; sensitive; nontoxic; easy to load into organelles, cells, or tissues without subsequent leakage or unwanted diffusion, excretion, or metabolism.⁹⁹  It should further be able to identify the site of production of the oxidant, quantify the produced amount of ROS as well as provide information that will enable the mechanistic understanding of the disturbance in cellular redox state. Fluorescent probes developed in recent years cover the majority of these specifications, however quantitative evaluation of the ROS production both in vitro and in vivo still remains a challenge.¹⁰⁰ 

Most of the fluorescent probes developed to date for in vitro and in vivo imaging are designed to activate in the presence of the analyte of interest. The activation results most often in an increase in fluorescence (off/on fluorogenic probes) or a shift in the emission wavelength (ratiometric probes). Most of the widely used probes that are commercially available are prefluorescent aromatic molecules that undergo oxidation in the presence of ROS to a fluorescent product (Figure 1.3). Many of the newer probes developed in recent years are compounds containing a masked fluorophore that is released by attack of the oxidant on the masking group (Figure 1.4). Probes that fall into the second category generally rely on photo-induced electron transfer (PeT) as the molecular switch of fluorescence. Deactivation of PeT upon oxidation of the trap segment restores the emission of the reporter segment. For more detailed information we would recommend several reviews published in recent years.^{93,99,101,102} 

The complex biology of ROS is dictated not only by the chemical properties of each type of oxygen metabolite but also their production sites and further trafficking within the cell.¹¹  This provides a motivation for developing tools to study the chemistry and biology of ROS in specific organelles in the cell. Fluorescence imaging with fluorogenic probes that can target specific organelles emerges as a valuable method for site-specific sensing of the different types of ROS exploring their complex contributions to physiological processes in living organisms. Most advances in this field have been made by developing fluorogenic probes that preferentially target mitochondria and that react specifically with H₂O₂,^106,107  superoxide,¹⁰⁸  lipid-based ROS,^98,109  singlet oxygen,¹¹⁰  hypochlorous acid,¹¹¹  and highly reactive ROS.^112,113