Biological nitric oxide signalling: chemistry and terminology

Introduction

Due to the discovery of endogenous nitric oxide (NO) generation in mammalian cells, the chemical biology of NO has been an important research topic for several decades (for an initial description see, Wink and Mitchell, 1998; for more recent reviews, see Thomas et al., 2008; Toledo and Augusto, 2012) (note: ‘nitric oxide’ is the common name for the molecule NO. The systematic name for NO is nitrogen monoxide. However, in the current literature, the use of ‘nitric oxide’ predominates and will be used exclusively herein). The biological effects of nitric oxide and derived nitrogen oxide species (this is a term that collectively refers to all nitrogen oxides including NO₂, N₂O₃, NO₂⁻, NO₃⁻, etc.) are a result of their fundamental chemical reactions with specific biological targets. The remarkably versatile and unique chemistry of NO and related nitrogen oxides (at times collectively referred to as ‘reactive nitrogen species’) is ideal for mediating numerous signalling networks. Indeed, cellular and physiological processes have exploited the novel and specific chemistry of NO with a variety of biological targets to evolve and develop signalling systems such as the regulation of vascular tone. Aberrant production of NO, often via misregulation of NO biosynthesis or problems associated with its availability, can lead to pathophysiological conditions. In such cases, problems with spatial and temporal NO generation can lead to disease states. The accurate description of NO chemistry and derived species with biological targets not only provides important insight into mechanisms of action but also provides guidance in the development of potential pharmacological agents/strategies to treat different diseases.

The fundamental chemistry of NO has been examined extensively. Importantly, these chemical studies involving the reactions of NO with inorganic and organic molecules have provided a framework for explaining the physiology and pathophysiology of NO. However, important aspects to consider when extrapolating purely chemical studies to an understanding of biological systems are potential differences in conditions, especially reactant concentrations and environment. Previous chemical studies have also provided a lexicon for describing distinct chemical processes. Therefore, it is important to be consistent and specific when referring to chemical phenomena in biological systems since a seemingly minor change in word usage can imply a significant difference in the chemistry. That is, use of proper chemical terms is vital to providing a rigorous description of chemical processes associated with NO biology. Improper use of terms/descriptors can be misleading and may misrepresent the intimate mechanisms of NO biology. Unfortunately, much of the current literature associated with nitrogen oxide biology/physiology uses a mixture of chemical terms to describe biological phenomena that is at times confusing, misleading and/or chemically incorrect. In this review, the biologically relevant chemistry of NO is described with an emphasis on the terminology and nomenclature of these processes.

Nitric oxide

A simple Lewis dot depiction of NO shows that it has one unpaired electron (Figure 1). Formal convention often requires that free radicals such as NO be depicted as ‘·R’ to clearly indicate the presence of the single unpaired electron. Herein, this convention has not been adopted for several reasons. First, a more rigorous understanding of the nature of the unpaired electron (by molecular orbital theory) indicates that it is not localized on either the nitrogen or oxygen but is delocalized over both atoms (although shared unevenly towards the nitrogen nucleus). Second, in the absence of any charges, it is clearly implied that NO has an unpaired electron thus it need not be shown explicitly. And finally, a similar molecule, O₂, actually has two unpaired electrons and few attempts to depict this in text are seen (i.e. O₂ is rarely depicted as ·O₂·).

Although there is an internationally agreed upon nomenclature (International Union of Pure and Applied Chemistry) for both organic and inorganic compounds, common names become widely used when the formal system is cumbersome. Common names are widespread in the nitrogen oxide literature as is use of NO rather than ·NO. Here, we are attempting to clarify the nomenclature rather than remake or formalize it. Regardless, that NO has an unpaired electron is one of its most important properties. Most organic molecules have all their electrons paired in either bonds or in non-bonding orbitals (e.g. lone pairs). Such molecules repel a magnetic field and thus are called diamagnetic. In contrast, unpaired electrons are attracted to magnets, leading to paramagnetic species. The rates of reactions between diamagnetic and paramagnetic molecules are often limited by high kinetic barriers due to spin forbiddenness. Thus, NO tends to diffuse readily through the medium of a cell until encountering other paramagnetic species.

At first glance, it appears that NO would either readily gain or lose an electron in order to have both atoms obey the octet rule, which states that for molecules containing main group elements, each nucleus should be surrounded by eight bonding and/or lone pair electrons so as to adopt an inert gas electron configuration. Reduction to NO⁻ or HNO would pair the lone electron while oxidation to NO⁺ would result in loss of this electron. In both cases, a Lewis structure can be drawn whereby both atoms possess an octet. However, NO is a stable molecule without a high propensity to be reduced or oxidized or even to dimerize in a biological system. The reasons for the stability of NO as a paramagnetic species are numerous and are covered elsewhere (e.g. see Fukuto et al., 2012). The important consideration is that although NO is a free radical, it is remarkably unreactive. This characteristic is critical for NO to function as a signalling molecule. In most cases, NO is produced in one cell for the purpose of eliciting a response in a neighbouring cell. This is possible due to the neutral charge and low reactivity of NO with diamagnetic species (as are most organic molecules). In biological systems, NO primarily reacts with other paramagnetic molecules as well as with transition metals. The impact of this chemistry is described below.

Nitrosation versus nitrosylation

Before proceeding further with a discussion of nitrogen oxide chemistry, it is worthwhile to first address some basic nomenclature associated with this chemistry. Especially important and a significant problem with the vernacular of NO chemical biology lies in the use of the terms ‘nitrosation’ and ‘nitrosylation’. These terms have very specific chemical definitions. Therefore, defining these terms (and their proper usage) provides important descriptors of detailed mechanisms that allude to important pharmacological and physiological insights.

In aqueous conditions, NO⁺ has an appreciable lifetime only under highly acidic conditions, and thus is generally not considered to be biologically relevant as an independent entity. Instead, nitrosation involves NO⁺ donors (species with significant ‘NO⁺-like’ character), formed by processes described in more detail below. Thus, the term nitrosation should be used when referring to chemical or biochemical processes that involve reactions of electrophilic NO⁺ donors with nucleophilic centres.

The related term ‘nitrosative stress’ refers to the indiscriminate nitrosation of biological nucleophiles that can lead to cell death and/or pathophysiological conditions. Nitrogen oxides such as NO₂ and ONOO⁻ are potent oxidants (vide infra) rather than nitrosating agents and toxicity associated with these species should not be referred to as nitrosative stress. It should be noted, however, that although NO₂ itself is not a nitrosating agent, since it cannot directly participate in nitrosation chemistry, it can serve as a precursor to nitrosating species (vide infra). Therefore, ‘nitrosative stress’ should not be used as a general descriptor of all NO-mediated cellular stresses but rather used when referring specifically to stresses associated with nitrosation chemistry. Recently, nitrosation events in cellular signalling have been recognized as a new mechanism in cellular regulation (e.g. Marozkina and Gaston, 2012). ‘Nitrosative signalling’ is a non-toxic event that is part of a signal cascade to elicit a specific biological response. In cancer, for example, nitrosative stress resulting from an immune response leads to tumour eradication (e.g. Stuehr and Nathan, 1989). In contrast, nitrosative signalling in cancer cells leads to increased metastasis, proliferation and chemoresistance resulting in poor outcome in patients (e.g. Glynn et al., 2010; Ridnour et al., 2012; Switzer et al., 2012). Therefore, biological nitrosation chemistry can result in either a stress or a signalling pathway that is both context and concentration dependent.

The term ‘nitrosylation’ traditionally refers to direct addition of NO to a reactant. The term originates from chemistry that describes the coordination of NO to a metal centre to form a metal nitrosyl complex (Reaction 2) (e.g. Ford et al., 2005), with nitrosyl being the common name for NO when bound as a ligand (akin to ‘carbonyl’ being used to describe metal coordination complexes involving carbon monoxide, CO).

(2)

It is worth noting that metal nitrosyls can also be formed via other chemistries. For example, they can be generated by reaction of a metal complex with acidified nitrite, although this reaction may involve a change in the oxidation state of the metal in a multistep process (e.g. Ford, 2010). The first characterized and probably the most important physiological receptor for NO in mammalian systems is the metalloenzyme soluble guanylate cyclase (sGC) (e.g. Ignarro, 1999). Coordination of NO to the ferrous heme of sGC is the quintessential example of nitrosylation in biology.

In biological systems, many post-translational modifications of macromolecules are described using the suffix ‘-ylation’ such as ‘phosphorylation’, ‘methylation’ and ‘sumoylation’. With respect to NO, descriptions of its biological actions have often used the term ‘nitrosylation’ to generally describe appending an NO group to macromolecules (akin to the terms above). From a purely chemical perspective, ‘nitrosylation’ is not nearly as general a term (vide supra). Indeed, the ubiquitous use of ‘nitrosylation’ can obscure the important chemical and biological mechanisms involved in nitrogen oxide signalling. Possibly the most prevalent, established and important macromolecular signalling phenomenon is protein phosphorylation, a process that is carefully regulated and controlled via a multitude of specific kinases and phosphatases that, in turn, are controlled in numerous ways. Therefore, the use of nitro‘syl'ation to draw analogy to, for example, phosphorylation (and other established signalling systems) should be done with caution at this time since an analogous level of regulation established for these other signalling systems has not yet been demonstrated for ‘nitrosylation’.

The terms ‘nitrosation’ and ‘nitrosylation’ are often used when describing the formation of S-nitrosothiols (RS-NO, vide infra). S-nitrosothiols can be formed by translocation of an NO⁺ group from one sulfur to another (Reaction 3).

(3)

This process has been referred to as ‘transnitrosylation’ (e.g. Anand and Stamler, 2012). Since the chemistry of this transfer likely involves nucleophilic attack of a thiol at the electrophilic nitrogen atom of the S-nitrosothiol (Singh et al., 1996; Wong et al., 1998; Houk et al., 2003), this process is more appropriately termed transnitrosation (assuming the mechanism described above is being implied).

Transfer of a nitrosonium ion is presumably an important mechanism for modification of protein thiols. For example, Mitchell and Marletta (2005) have shown that the thiol redox protein thioredoxin can, with some specificity, transfer a nitrosonium species from one of its cysteines to the active site cysteine of capase-3. Thus, transnitrosation may in fact be a significant component of cellular signalling (e.g. Martinez-Ruiz and Lamas, 2007). Given the pre-existence of the terminology, their individual importance to nitrogen oxide signalling and the lack of similarity to other enzymatically controlled post-translational modifications, following the suggestion by Ford et al. (2005), we recommend that nitrosation be reserved exclusively for reactions involving transfer of NO⁺ and nitrosylation for direct addition of NO.

Formation of S-nitrosothiols

It has become increasingly clear that modification of both low molecular weight thiols and thiol-containing proteins occurs as a result of the biosynthesis of NO, and there are several possible chemical mechanisms for their generation. As alluded to in the above section, NO itself does not react with thiols or thiolates under biological conditions (or at the very least, this is a very slow process that generally precludes biological relevance) (Pryor et al., 1982). Modification of thiols by NO requires either prior oxidation of NO (to give an NO⁺-donor) or of the thiol (to give a thiyl radical, RS·).

Oxidation of NO

One-electron oxidation of NO to give free NO⁺ is highly unfavourable. In fact in aqueous systems, NO⁺ only has an appreciable lifetime in very strong acid. In purely chemical systems, NO⁺ can be formed by addition of nitrite (NO₂⁻) to ∼4 M sulfuric acid, but such conditions are not physiological. However, species with NO⁺-like reactivity can be formed under milder conditions. For example, NO and nitrogen dioxide (NO₂, another paramagnetic species) can combine to give dinitrogen trioxide (N₂O₃) (Reaction 4).

(4)

This reaction is reversible, suggesting that N₂O₃ can simplistically be described as an adduct of NO and NO₂. However, NO₂ is a reasonable one-electron oxidant, and thus N₂O₃ may possess some NO⁺NO₂⁻ character. In the presence of a nucleophile, a Nuc···NO⁺···NO₂⁻ intermediate would lead to transnitrosation from NO₂⁻ to the nucleophile, to produce for instance an S-nitrosothiol (Reaction 5).

(5)

In this way, N₂O₃ functions as an NO⁺ donor through transfer of NO⁺ rather than spontaneous release of free NO⁺.

Generation of S-nitrosothiols via reaction of thiols with N₂O₃ is a chemically established process. The formation of N₂O₃, as mentioned above, occurs when NO₂ is generated in the presence of NO (Reaction 4). One mechanism to produce NO₂ is through reaction of NO with O₂ (autoxidation of NO; Reaction 6).

(6)

The paramagnetism of the reactants leads to a low kinetic barrier [high reaction rate constant, k, of 8 × 10^6·M^−2·s⁻¹ (Wink et al., 1993a; Keshive et al., 1996)]. However, the second order dependence on NO (rate = k[NO]²[O₂]) also indicates that significant levels of N₂O₃ are formed only at high concentrations of NO. The high order kinetics of NO autoxidation indicates that the time required for an aerobic solution of NO to decay to half its original concentration varies hugely with concentration (on the order of hours for nanomolar and seconds for micromolar – note that this is not a true half life since this is not a first order decay process).

Under normal biological conditions/concentrations autoxidation of NO is generally considered to be too slow to be significant. However, higher levels of NO (and O₂) may concentrate in lipophilic compartments (e.g. membranes) due to the favourable partitioning of these non-polar species from aqueous solution (Liu et al., 1998). Furthermore, inflammation leads to increased levels of NO. Alternatively, NO₂ may be formed by a different route, such as through metal-mediated oxidation of nitrite (Thomas et al., 2008). At present, the biologically relevant mechanisms by which N₂O₃ is formed are not fully established.

However, that N₂O₃-mediated nitrosation chemistry does occur in vivo is clear and numerous studies in in vitro and in vivo systems allude to the relevance of N₂O₃ in mediating NO biology. Shinyashiki et al. (2001) provide just one example in which disruption of the actions of the thiol-containing, metal responsive yeast transcription factor ACE1 was observed under conditions consistent with N₂O₃ generation. ACE1 inhibition occurred upon addition of NO under aerobic conditions (Reactions 6 and 4), upon addition of nitrite to acidic solution (Reaction 7), or when NO and NO₂ were present simultaneously (Reaction 4).

(7)

Since all of these conditions are amenable to N₂O₃ formation, it was concluded that N₂O₃ was the species responsible for modifying and inhibiting ACE1 activity.

Oxidation of thiols

S-nitrosothiols may also be formed by one-electron oxidation of the thiol to a thiyl radical followed by direct reaction with NO (Reactions 8 and 9).

(8)

(9)

Thus oxidation of thiol to thiyl radical followed by addition of NO is termed oxidative nitrosylation. It needs to be stressed that NO does not directly react with thiols under any biological condition, therefore there is no direct nitrosylation mechanism for RSNO formation via reaction of NO with thiols (vide infra). The relevance of this nitrosylation pathway to S-nitrosothiol generation is, of course, dependent on the prevalence of thiyl radicals as well as competing reactions with thiol or O₂. Significantly, many proteins/enzymes are known to possess cysteine thiyls (among other radical species) that are crucial for enzymatic activity (e.g. Stubbe and van der Donk, 1998). Thus, enzymes/proteins with existing thiyl radicals may be expected to be reactive with NO via Reaction 9. Oxidation of thiols to thiyl radicals typically is driven by the presence of strong one-electron oxidants such as HOO· and HO·. Two-electron oxidants such as H₂O₂ can instead produce the sulfenic acid, RSOH, which is not susceptible to nitrosylation. Both thiyl radicals and sulfenic acids can undergo further reactions to give other more stable species such as disulfides. Thus, for thiol oxidation to result in S-nitrosothiol formation requires coordination with NO biosynthesis.

NO₂ is also a strong oxidant that is capable of oxidizing thiols. However, since NO and NO₂ rapidly associate (Reaction 4), their simultaneous presence is commonly assumed to produce S-nitrosothiols via the nitrosation pathway. However, there have been several reports suggesting that thiols can effectively compete for the NO₂ generated during autoxidation at low fluxes of NO such that S-nitrosothiol formation under these conditions may not be entirely due to a nitrosation event (Jourd'heuil et al., 2003; Koshiishi et al., 2007). This suggests that there may be more complex reaction mechanisms at low (physiological/pathophysiological) levels of NO. Regardless, it should be noted that the mechanism(s) by which NO-mediates S-nitrosothiol formation requires further investigation.

Nitrosylation of metal complexes and metalloproteins

Although sGC possesses basal activity, it is activated many fold by NO. Elucidation of the full mechanism of this activation process is hampered by the lack of detailed structural data for the large protein. However, it is generally accepted that nitrosylation of the regulatory ferrous heme, producing a ferrous-nitrosyl (Fe^II-NO) complex, leads to dissociation of the proximal histidine ligand (Traylor and Sharma, 1992). Although other small molecules such as O₂ and CO also can bind to ferrous hemes, cleavage of the proximal ligand (due to a strong trans directing nature) is unique to NO. In turn, movement of the helix that includes the proximal histidine leads to structural changes that facilitate conversion of GTP to cGMP at a distant catalytic site.

NO reacts with a variety of other Fe^II systems including the heme centres of haemoglobin, myoglobin, NO synthase and cytochrome P450 (Khatsenko et al., 1993; Wink et al., 1993b; Stadler et al., 1994) as well as the iron-sulfur cluster of aconitase (Drapier and Bouton, 1996). In contrast to sGC, nitrosylation of most enzymes leads to regulation of function by reducing activity. For example, nitrosylation of the Co(II)(H₂O) form of cobalamin (vitamin B₁₂ derivative) results in a diminished ability for this complex to serve as a cofactor for methionine synthase (Brouwer et al., 1996).

Metal complexes can also function as NO scavengers. Haemoglobin, and to a lesser extent myoglobin, in particular is considered to regulate NO levels. In these proteins, NO reacts with the Fe(II)O₂ complex to rapidly produce methaemoglobin (Fe^III) and nitrate (Reaction 11, shown for haemoglobin, Hb) (Eich et al., 1996).

(11)

Nitrate is then excreted [or possibly recycled back to NO (Kapil et al., 2010)], and metHb can be enzymatically reduced back to the O₂-binding ferrous haemoglobin.

NO also reacts with more oxidized metals. Nitrosylation of ferric (Fe^III) complexes is generally considered to be reversible, due to lower binding affinities compared to the corresponding ferrous analogues (see Lim et al., 2005). However, nitrosylation of ferric heme centres can also lead to enzyme inhibition, as observed for catalase (Hoshino et al., 1993; Farias-Eisner et al., 1996). Furthermore, the resulting ferric nitrosyl complex often is written as Fe^II-NO⁺ rather than Fe^III-NO, indicating that significant redistribution of electron density occurs upon binding [Reaction 12; also applicable to Mn(III)]. As with N₂O₃ and S-nitrosothiols, ferric nitrosyls can therefore be susceptible to nucleophilic attack (Reaction 13).

(12)

(13)

In this case, nitrosylation of the metal in the presence of a nucleophile leads to transnitrosation to the nucleophile and one-electron reduction of the metal centre. A second nitrosylation of the reduced metal centre then produces a stable ferrous nitrosyl complex (Reaction 14).

(14)

Together, this multistep process is called reductive nitrosylation (e.g. Gwost and Caulton, 1973; Wayland and Olson, 1973). Wade and Castro demonstrated that ferric nitrosyl complexes could nitrosate biomolecules including N-acetyl cysteine (Wade and Castro, 1990).

Nitrosation of other biological targets

Nitrosation is not limited to thiols, and the study of nitrosation chemistry is more than a century old (reviewed in Williams, 1983). Prior to the discovery that NO is biosynthesized, nitrosation of amines in the gastrointestinal tract became a concern as a potential source of carcinogens (Bartsch et al., 1990) (Reaction 15).

(15)

In fact, the detection of N-nitrosamines following activation of the immune system was critical to discovery of endogenous formation of NO (Stuehr and Marletta, 1985; Miwa et al., 1987; Hibbs, 1991). Formation of N-nitrosamines has also been proposed to play a role in carcinogenesis associated with inflammation (Marletta, 1988).

Primary amines (RNH₂) such as found in nucleobases can also be nitrosated by either N₂O₃ or ferric nitrosyl complexes (Castro and Bartnicki, 1994). However, primary N-nitrosamines are unstable to further reaction as they can tautomerize. As shown in Figure 2 for cytosine, the N-nitrosamine can tautomerize and then dehydrate to produce a diazonium ion. This intermediate hydrolyzes to release N₂ to produce an enol. Tautomerization to the keto form produces uracil. Importantly, Caulfield et al. (1998) have demonstrated that DNA structure plays an important role in this potentially mutagenic reaction.

Reaction of NO with radical species

Unpaired electrons can impart high reactivity to many radical species in that they can be electron poor and, therefore, oxidants. The resistance of NO to dimerization is highly unusual for a radical, but is critical to its function as a signalling agent. NO does react rapidly with other radicals, as shown in Reactions 4 and 9. In these reactions, two radicals combine to form a diamagnetic species (formally via nitrosylation). Since O₂ has two unpaired electrons, two NO molecules react with it (Reaction 6). Another important radical to consider with NO is superoxide (O₂⁻). Although reduction of O₂ is unfavourable, interaction of O₂ with strong reductants can result in production of O₂⁻. In particular, O₂⁻ can be produced by the respiratory chain necessitating a high concentration of superoxide dismutase in mitochondria. However, immune cells have evolved to utilize O₂⁻ in the control of invading microorganisms. Since the immune system also produces NO, the interaction of NO and O₂⁻ has received substantial attention both as a scavenger of NO and in the formation ONOO⁻ (Reaction 16).

(16)

Peroxynitrite itself is not a radical species (i.e. it does not have any unpaired electrons). However, proton dependent decomposition of ONOO⁻ has the potential to generate potent odd electron oxidants NO₂/HO· as fleeting intermediates contained in a solvent cage in a reaction pathway that gives predominantly nitrate (NO₃⁻) (Reaction 17) (e.g. Gunaydin and Houk, 2008).

(17)

The biochemical fate of ONOO⁻ is highly dependent on its environment. Under normal biological conditions, reaction of ONOO⁻ with CO₂ is likely the major reaction giving nitrosoperoxycarbonate (Reaction 18). Further decomposition of nitrosoperoxycarbonate (via possible homolytic cleavage of the O-O bond, analogous to Reaction 17 above), generates potent oxidants, NO₂/CO₃^·−, as fleeting intermediates on a reaction path that eventually forms nitrocarbonate (which will further decompose to give NO₃⁻) (Reaction 19).

(18)

(19)

Thus peroxynitrite is not itself a nitrosating agent. However, the intermediates in H⁺- or CO₂-mediated decomposition can produce powerful oxidants (NO₂, HO· or CO₃^·−). Yet, in excess NO, these oxidants can be converted to a variety of species including the nitrosating agent N₂O₃ (Reactions 20, 21, 22). This balance between oxidative chemistry and nitrosative chemistry may provide an important interface between these two chemical conditions.

(20)

(21)

(22)

The biochemical consequences of ONOO⁻ formation and degradation will be discussed further later.

In addition to small radicals such as NO₂ or O₂⁻, NO can react with carbon-centred radicals (R·) to give the corresponding C-nitroso compound (Reaction 23).

(23)

Most alkyl nitroso compounds are unstable with respect to tautomerization to the corresponding oxime (Reaction 24).

(24)

Such reactivity (i.e. tautomerization to an oxime) is not available to tertiary or phenyl nitroso compounds.

In aerobic systems, carbon-centred radicals can be converted into oxy- and peroxy-radicals. This can be particularly deleterious since these radicals can initiate and/or propagate chain reactions. That is, a single radical initiating event can lead to the destruction or modification of numerous molecules via propagation by chain-carrying and O₂-derived reactive intermediates. One of the most important examples of this is lipid peroxidation. Production of a one-electron oxidant (e.g. HO·, NO₂, HOO·, etc.) in a membrane lipid environment can lead to the generation of a carbon-centred radical via one-electron oxidation of an unsaturated fatty acid (typically by hydrogen-atom abstraction or addition to an unsaturation). In the presence of polyunsaturated fatty acids (PUFA), membranes are more susceptible to lipid peroxidation due to the fact that PUFAs are more easily oxidized compared to saturated or monounsaturated fatty acids. In an aerobic environment, the fatty acid radical will react with O₂, but only one of the two unpaired electrons of O₂ will be paired by this process. The resulting alkyl peroxyl radical (LOO·) is a reasonable one-electron oxidant that can propagate the chain reaction by oxidizing another fatty acid. This process is schematically depicted in Figure 3 using linoleic acid as an example.

Lipid peroxidation can be initiated during the inflammatory process and can ultimately lead to cell death by disrupting cell membranes (e.g. Higdon et al., 2012). There are multiple radical intermediates in the lipid peroxidation process (HO·, lipid radical, lipid peroxyl radical), all of which can react with NO (e.g. Reaction 25).

(25)

This quenching by NO of the radical nature of intermediates involved in the initiation or propagation of lipid peroxidation leads to termination of the process (e.g. Padmaja and Huie, 1993; Rubbo et al., 1995; Wink et al., 1995). This protective function of NO has been observed in both endothelial (Struck et al., 1995) and macrophage cells (Hogg et al., 1995). Clearly, the utility of NO in this regard will be highly dependent on having NO generated in proximity to these radical events and having it at high enough concentrations. Importantly, NO will partition favourably into lipid environments (Liu et al., 1998).

The build up of cholesterol in arteries produces atherosclerotic plaques, which are the cause of most coronary artery disease and strokes. Such plaques are initiated by accumulation of oxidized lipoproteins produced by the inflammatory response. Donors of NO have been found to prevent the macrophage-dependent oxidation of low-density lipoproteins by termination of lipid peroxidation (Hogg et al., 1993). This study indicates that NO may provide endogenous protection against plaque development. In fact, NO has been shown to be a more potent inhibitor of lipid peroxidation than the antioxidant α-tocopherol (vitamin E) (O'Donnell et al., 1997).

NO has also been designated as a pro-oxidant, capable of initiating radical chain chemistry. Such effects are not likely due to NO itself since it is such a poor oxidant (one-electron reduction of NO is highly unfavourable). Rather, NO₂ and other strong oxidants, derived from the reaction of NO for instance with O₂ and O₂⁻, are likely responsible. Again, association of NO with such oxidants will lead to radical quenching and thus will limit lipid oxidation.

Lipid peroxidation can also be initiated indirectly by oxidation of metal centres. In biological systems, Fe^II and Fe^III are the most common oxidation states. However, more oxidized species can form transiently, for example, due to reaction of H₂O₂ (Reaction 26).

(26)

Such highly reactive species [i.e. Fe(IV)=O or HO·, Reaction 26] can lead to cellular damage if uncontrolled (Puppo and Halliwell, 1988). NO can inhibit this chemistry by reducing these oxidized centres to a normal valence state (Reaction 27) (Kanner et al., 1991; Wink et al., 1994), providing a protective mechanism against peroxide-mediated cytotoxicity (Wink et al., 1993c).

(27)

Oxygen-based radicals are also produced during normal metabolism, most notably the tyrosyl radical. Both ribonucleotide reductase and cyclooxygenase/prostaglandin H-synthase contain a catalytically important tyrosyl radical. Reaction with NO results in enzyme inhibition (for a review, see Guittet et al., 1999). The resulting suppression of DNA or prostaglandin synthesis is a possible deleterious function of NO. However, the mammalian immune system utilizes this inhibitory effect to combat pathogens.

As discussed immediately above, the anti- and pro-oxidative properties of NO are highly dependent on the concentration of NO as well as the biochemical environment. Thus, it is often difficult to predict the effects NO will have on a system without an intimate understanding of these issues.

Nitration via NO-derived species

In a formal sense, the term nitration is analogous to that of nitrosation, referring specifically to chemical reactions involving the addition of a nitronium ion (NO₂⁺) to a nucleophilic group. However, nitration has been used extensively, particularly by organic chemists, as a general term to denote formation of a nitro compound without implying any specific mechanism of X-NO₂ formation (X representing an atom to which an NO₂ group has been added). As with NO⁺, the highly electrophilic NO₂⁺ is only generated under highly acidic conditions (e.g. nitric acid in sulfuric acid). Such conditions have been used extensively to nitrate organic compounds but are not relevant to biological conditions. In contrast to nitrosation, a physiologically relevant NO₂⁺ transfer agent has not been established. Thus, the likely and primary mechanism for nitration in cellular systems involves association of NO₂ to existing radicals (comparable to nitrosylation). However, a unique term has not been defined for this process.

As already described, NO₂ can be formed from the autoxidation of NO (Reaction 6) and from the reaction of NO with O₂⁻ (Reaction 16). In the first case, as discussed earlier, the second order dependence of NO on autoxidation to give NO₂ limits its prevalence in the aqueous environment of cells but to a lesser extent in lipophilic regions. Also, NO₂ generated from NO autoxidation will be kept at low levels due to further reaction with NO leading to N₂O₃ formation (thus not allowing accumulation NO₂) (Reaction 4). Significantly, NO₂ can also be formed in the absence of NO from oxidation of NO₂⁻ by H₂O₂ in the presence of metals (e.g. Burner et al., 2000; Monzani et al., 2004) (Reaction 28).

(28)

Again, the metal centres function as peroxidases as shown here for Fe^III systems. In the first step, H₂O₂ leads to production of a more oxidized species (compound I; Reaction 29; please note that the metal oxidation states are shown primarily as a means of electron accounting and may not accurately represent the actual oxidizing species).

(29)

This oxidized species can participate in two sequential one-electron oxidations of NO₂⁻, ultimately regenerating the Fe^III starting complex (Reactions 30 and 31).

(30)

(31)

There has been significant speculation that biological nitration primarily follows this pathway. Nitrite is an end-product of NO biosynthesis, and thus can lead to NO₂ formation by Reactions 30 and 31. Importantly, the possible temporal separation of NO biosynthesis from metal-mediated NO₂⁻ oxidation reduces the likelihood of reaction of NO₂ with NO, which would lead to nitrosation rather than nitration (vide supra). Nitrite can also be ingested in the diet, which completely eliminates the dependence on NO biosynthesis.

In biological systems, typically an oxygen, carbon or other main group element is nitrated. It is well established that nitrated adducts, including those of tyrosine, tryptophan, fatty acids and guanosine, are prevalent biological species under conditions of NO-generation (e.g. Akuta et al., 2006; Souza et al., 2008; Nuriel et al., 2011; Schopfer et al., 2011) (Figure 4). Unlike many of the chemical processes involving nitrogen oxides, as described above, nitration is currently considered to be a biologically irreversible modification. That is, biologically mediated conversion of a nitrated species back to the original molecule is not reported to happen (although nitro group reduction has been reported).

The term ‘nitrative stress’ is often used to indicate conditions associated with nitration events on, for example, proteins or nucleotides leading to pathologies (e.g. Zaki et al., 2005). It is important to note the distinction between the terms ‘nitrosative stress’ and ‘nitrative stress’. As discussed earlier, nitrosative stress results from the nitrosation of biological molecules (e.g. protein cysteines) while nitrative stress is associated with nitration of biological molecules (e.g. protein tyrosines). ‘Nitrative stress’ has been characterized primarily via the detection of nitrated proteins. For example, nitration of tyrosines (3-nitrotyrosine formation) detected by antibodies is often used to characterize/indicate a nitrative stress. However, several important questions regarding this phenomenon remain. For example, is nitration merely a dosimeter of a chemical/biochemical condition? Are there any significant functional consequences of nitration? To date, answers to these questions are not well established. A further complication is that nitration of tyrosine may result from oxidative or nitrosative conditions (vide infra) indicating that nitration does not represent or result from a unique biochemical condition.

One mechanism of 3-nitrotyrosine formation from interaction with ONOO⁻ and derived species has been determined to involve first oxidation of tyrosine by the carbonate radical anion (CO₃^·−) generated from the nitrosoperoxycarbonate intermediate (vide supra), and then reaction of the tyrosyl radical with NO₂ (Reactions 17, 18, 19, 32, 33) (e.g. Gunaydin and Houk, 2009).

(32)

(33)

It should be noted that the yield of 3-nitrotyrosine by this chemistry is highly dependent on the relative concentrations and fluxes of NO and O₂⁻ (the precursors to ONOO⁻) with the highest yield occurring at equal fluxes. Moreover, even under ideal conditions a less than 40% yield is realized under purely chemical conditions (Goldstein et al., 2000).

Since 3-nitrotyrosine is formed by the reaction of the tyrosyl radical with NO₂ (Reaction 33), any oxidant capable of removing an electron from tyrosine in the presence of NO₂ can lead to 3-nitrotyrosine production. Importantly, NO₂ itself is capable of oxidizing tyrosine to Tyr-O·, and thus any system that generates NO₂ can lead to the formation of 3-nitrotyrosine.

The reaction of the tyrosyl radical with NO is also facile. Interestingly, the product 3-nitrosotyrosine has been suggested to further oxidize to give 3-nitrotyrosine via an intermediate iminoxyl radical (Gunther et al., 2002). In this pathway, an existing tyrosyl radical reacts with NO to give a ‘weak complex’ that tautomerizes to 3-nitrosotyrosine. Subsequent tautomerization to an oxime can be followed by one-electron oxidation generating an iminoxyl radical intermediate that is further oxidized to give an electrophilic species capable of reacting with water to give 3-nitrotyrosine (Figure 5).

Conclusion and summary of terms

The biological chemistry of NO is dominated by its ability to coordinate to metals and to react with other radical species. The subsequent chemistry that is initiated by the formation of metal nitrosyls and NO-radical adducts is important to the ultimate fate and chemical biology of NO. Importantly, much of the lexicon of nitrogen oxide chemistry was established prior to the relatively recent explosion of biological interest/application. Unfortunately, a lack of adherence to the established chemical terminology can lead to problems or misunderstandings regarding the intimate chemical mechanisms and processes being described. Herein described are some of the biologically relevant chemistry of NO (and derived species) along with what we believe to be the proper terminology.

To be sure, not all biological/biochemical terms are as chemically rigorous or descriptive as would be preferred. For example (and particularly relevant to this discussion and mentioned above) is the term ‘phosphorylation’. In biochemistry textbooks, the term phosphorylation is used to describe the transfer of a phosphate from the atom of one biomolecule to another (typically from one oxygen atom to another). ‘Phosphoryl’ groups are derivatives of phosphoric acid (H₃PO₄). Therefore, ‘phosphorylation’, as used in biochemistry, is a term that describes the products of the reaction but not the mechanism. However, it needs to be noted that biochemical phosphorylation occurs via attack of a nucleophilic atom of a biomolecule at an electrophilic phosphorous atom of the phosporyl donor (to our knowledge, there are no other known reaction pathways) thus precluding a need for numerous mechanistically distinct terms.

Since formation of an X-NO bond can occur by two distinct mechanisms, (described above), it is preferable to use a mechanistically defined term when possible. The question then arises as to the proper nomenclature when the mechanism is neither known nor implied. Both nitrosation and nitrosylation are often used in the literature to denote the general (mechanistically ill-defined) formation of X-NO compounds. However, given that both terms inherently describe a mechanism of formation, neither term is appropriate in cases where the mechanism is unknown or ambiguous. Rather, a more generic description such as thiol modification or S-nitrosothiol production is recommended (in the case of thiol targets). Currently, there is the tendency for many to view ‘nitrosation’ as a chemical term that specifically denotes NO⁺-mediated modification of a nucleophilic species (such as a thiol) and ‘nitrosylation’ as a biological term that denotes the mere formation of an X-NO species (X = S, N, etc.) without regard to mechanism. However, it needs to be stressed that nitrosylation is also a chemical term with mechanistic implications (vide supra) and use of nitrosylation to merely draw analogy to, for example, phosphorylation seems unwarranted at this time.

Conflict of interest

There are no potential conflicts of interest.