Free Radical Biology & Medicine, Vol. 37, No. 8, pp. 1122–1143, 2004
`Copyright D 2004 Elsevier Inc.
`Printed in the USA. All rights reserved
`0891-5849/$-see front matter
`
`doi:10.1016/j.freeradbiomed.2004.06.013
`
`Serial Review: Mechanisms and Novel Directions in the Biological Applications
`of Nitric Oxide Donors
`Serial Review Editor: S. Bruce King
`
`NITRATES AND NO RELEASE: CONTEMPORARY ASPECTS IN
`BIOLOGICAL AND MEDICINAL CHEMISTRY
`
`GREGORY R. J. THATCHER, ADRIAN C. NICOLESCU,
`
`1 BRIAN M. BENNETT,
`
`1
`
`and VIOLETA TOADER
`
`Department of Medicinal Chemistry and Pharmacognosy, College of Pharmacy,
`University of Illinois at Chicago, 833 S. Wood Street, Chicago, IL 60612-7231, USA
`
`(Received 10 May 2004; Accepted 3 June 2004)
`Available online 6 July 2004
`
`Abstract—Nitroglycerine has been used clinically in the treatment of angina for 130 years, yet important details on the
`mechanism of action, biotransformation, and the associated phenomenon of nitrate tolerance remain unanswered. The
`biological activity of organic nitrates can be said to be nitric oxide mimetic, leading to recent, exciting progress in
`realizing the therapeutic potential of nitrates. Unequivocally, nitroglycerine and most other organic nitrates, including
`NO-NSAIDs, do not behave as NO donors in the most fundamental action: in vitro activation of sGC to produce cGMP.
`The question as to whether the biological activity of nitrates results primarily or exclusively from NO donation will not
`be satisfactorily answered until the location, the apparatus, and the mechanism of reduction of nitrates to NO are defined.
`Similarly, the therapeutic potential of nitrates will not be unlocked until this knowledge is attained. Aspects of the
`therapeutic and biological activity of nitrates are reviewed in the context of the chemistry of nitrates and the elusive
`
`reduction required to generate NO. D 2004 Elsevier Inc. All rights reserved.
`efficient 3e
`
`Keywords—Nitrate, Nitric oxide, Biotransformation, cGMP, Signaling, Nitration, Nitrosation, Free radicals
`
`Contents
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`Introduction .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`Nitrates as therapeutic agents .
`.
`.
`.
`.
`.
`.
`Nitrate biotransformation and tolerance .
`.
`.
`.
`.
`.
`.
`Nitrates as nitric oxide donors.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`Nitrates, NO, and sGC activation .
`.
`.
`.
`NO release from nitrates: models and mechanisms .
`Reactions of nitrates with thiols .
`.
`.
`.
`.
`.
`.
`.
`.
`Thionitrates and sulfenyl nitrites .
`.
`.
`.
`.
`.
`.
`.
`.
`Reactions of nitrates with metal ion assistance .
`Conclusions .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`Acknowledgment
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`References .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`1123
`1124
`1127
`1128
`1129
`1131
`1132
`1133
`1134
`1135
`1135
`1135
`
`This article is part of a series of reviews on bMechanisms and Novel Directions in the Biological Applications of Nitric Oxide Donors.Q The full list
`of papers may be found on the home page of the journal.
`Address correspondence to: Greg Thatcher, Department of Medicinal Chemistry and Pharmacognosy, College of Pharmacy, University of Illinois at
`Chicago, 833 S. Wood St., Chicago, IL 60612-t7231, USA; Fax: (312) 996 7107; E-mail: thatcher@uic.edu.
`1
`Present address: Department of Pharmacology & Toxicology, Queen’s University, Kingston K7L 3N6, Canada.
`
`1122
`
`Human Power of N Company
`EX1048
`Page 1 of 22
`
`

`

`Nitrates and NO release
`
`1123
`
`INTRODUCTION
`
`The organic nitrate nitroglycerine is well into its second
`century of use both as a medicinal agent and as an agent
`of destruction and suffering. In popular culture nitro-
`glycerine is used as a simile for anything that is volatile
`and explosive under the merest provocation. A search of
`the Web will generate equal mentions of the use of
`nitroglycerine in terrorist and sniper attacks and in new
`therapeutic applications from spider bites,
`to anal
`fissures,
`to restless leg syndrome. Despite its deep
`history, knowledge of the biological chemistry of nitro-
`glycerine and of nitrates in general is wholly incomplete.
`The synthesis of nitroglycerine was first reported in
`1846, and 20 years later, the taming of nitroglycerine in
`the form of dynamite and gun glycerine was the basis of
`Alfred Nobel’s fortune. Nobel’s contemporaries, Thomas
`Brunton, who was studying the effects of organic nitrites
`in angina pectoris, and William Murrell, who demon-
`strated that small doses of nitroglycerine taken sublin-
`gually provided rapid and remarkable relief from the
`intense pain of angina, are generally credited with the
`development of nitroglycerine as a therapeutic agent
`[1,2]. Nitroglycerine was renamed glyceryl
`trinitrate
`(GTN), to avoid the anxiety associated with ingesting a
`high explosive, and has been used continuously in
`treatment of angina since 1878. GTN has also been
`applied in controlled hypotension during cardiac surgery
`and in congestive heart
`failure. Decreased cardiac
`preload due to the selective venodilator response to
`organic nitrate vasodilators is the basis for their clinical
`use in treatment of angina;
`these, and other simple
`nitrates that have been used only as biological probes, we
`may class as bclassical nitratesQ (Fig. 1).
`Sinitrodil, FR 46171, and nicorandil represent nitrates
`containing functionalities, other than simple hydrocarbon
`or sugar skeletons, which influence biological activity
`(Fig. 1). Sinitrodil is proposed as an agent with enhanced
`antianginal and attenuated hypotensive activity relative
`to classical nitrates due to reduced dilation of the smaller
`coronary and resistance vessels and the resultant effects
`
`on mean arterial blood pressure and heart rate [3,4].
`Nicorandil
`is a nitrovasodilator that
`is also a KATP
`channel opener, which is in preclinical and clinical
`studies for vascular diseases,
`including myocardial
`infarction [5,6]. Nitrate esters have a significant clinical
`attribute in that a nitrate is, inherently, considerably more
`lipophilic than its parent alcohol, allowing delivery by
`routes including sublingual or transdermal, as used in
`slow-release GTN patches. Oral administration of GTN
`is ineffective because of rapid first-pass metabolism, but
`the lone criticism of GTN as a therapeutic agent is the
`onset of clinical nitrate tolerance during chronic admin-
`istration. The physiological perturbation that causes
`tolerance is not known, but most theories have a central
`role for attenuated nitrate bioactivation (also termed
`mechanism-based biotransformation). The efficient bio-
`transformation of nitrates is also central to the therapeutic
`activity of nitrate drugs, because it is almost universally
`assumed that nitrates are in fact NO prodrugs, although
`the details of this mechanism-based biotransformation of
`the nitrate moiety (RONO2) to NO remain uncertain.
`The pharmaceutical development of nitrates contain-
`ing adjunct pharmacophores is not a new pursuit. For
`example, steroid nitrates had been reported over 40 years
`ago and observed to manifest biological properties
`beyond those of the parent steroid [7,8]. However, there
`has been an explosion of activity in the area of hybrid
`nitrates over the past decade stimulated by the actions of
`Del Soldato and NicOx SA (France) in developing NO-
`NSAIDs (NO-donating nonsteroidal anti-inflammatory
`drugs), sometimes known as CINODs (cyclooxygenase
`inhibitory NO donors), and in promoting the scientific
`study of these compounds (Fig. 2) [9–12]. Interest in the
`further use of nitrates in new cardiovascular applications
`is increasing, but
`there is a growing realization that
`nitrates may represent new therapeutic agents in non-
`traditional areas, for example, in neuroprotection and
`neuropathic pain [13–17]. This represents a paradigm
`shift that has been inspired by discoveries in NO biology
`and pharmacology, examples being the application of
`nitrates in dementia [13], the association of nitrates and
`
`Fig. 1. Nitrate vasodilators, including classical nitrates: erithrityl tetranitrate (ETN), glyceryl trinitrate (GTN), pentaerithrityl tetranitrate
`(PETN), isosorbide dinitrate (ISDN), isosorbide-5-mononitrate (5ISMN).
`
`Page 2 of 22
`
`

`

`1124
`
`G. R. J. Thatcher et al.
`
`Fig. 2. Selected NO-NSAIDs and hybrid nitrates. These nitrates are all described as NO-donor drugs and are in preclinical or clinical
`trials for a range of indications; NO-naproxen, AZD 3582, has completed Phase 2 clinical trials.
`
`NO with regulation of glutamate receptors [18–20], and
`the proposal of hybrid nitrate drugs containing the
`noncompetitive, open-channel NMDA receptor antago-
`nist memantine [21]. There are two observations that are
`common to many of the published studies on NO-
`NSAIDs and hybrid nitrates: (i) the biological activity of
`the nitrate is greater than that of the parent and (ii) the
`biological activity observed is not seen for the parent
`pharmacophore [22]. A further common observation is
`that neither the chemical mechanism nor the biological
`apparatus for release of NO is understood.
`Nitrate esters or organic nitrates contain the nitrooxy
`functional group (ONO2), almost all examples being
`aliphatic nitrates, owing to the presumed instability of the
`aromatic nitrate to rearrangement [23]. Contrary to some
`descriptions, nitrates are not nitroso compounds. This is
`not a point of semantics, because nitroso compounds,
`U
`NjO), nitrosamines
`including nitrosothiols (RS
`U
`U
`NjO), and nitrosoalkanes (RRVRVC
`NjO)
`(RRVN
`require only one electron reduction to yield NO.
`Conversion of the nitrooxy group of nitrates to NO is a
`three-electron reduction that must involve oxygen atom
`transfer. To date, no purified protein system has been
`demonstrated to mediate the direct reduction of nitrates
`to yield relevant quantities of NO, although it is widely
`held that the biologically active product of mechanism-
`based biotransformation of nitrates is NO [24,25]. The
`present article reviews the potential chemical mecha-
`nisms and the biochemical pathways associated with
`nitrate biotransformation. There is a vast literature on the
`biological activity of various nitrates, in contrast to a
`relative paucity of studies on the biological chemistry.
`Endogenous nitrates may be formed in vivo, for example
`in lipid peroxidation radical chain termination by nitric
`
`oxide [26–28], but nitrates are first and foremost
`exogenous therapeutic agents and it
`is the biological
`chemistry of these compounds that is of primary interest.
`
`NITRATES AS THERAPEUTIC AGENTS
`
`Physiologically, NO is produced enzymatically from
`the terminal guanidino nitrogen of l-arginine, by nitric
`oxide synthase (NOS). Endothelial NOS (eNOS) releases
`NO, which causes inhibition of platelet aggregation and
`vasodilation via activation of soluble guanylyl cyclase
`(sGC)
`in underlying vascular smooth muscle cells.
`Inducible NOS is found in many tissues and, in macro-
`phages, when induced, produces NO as part of the body’s
`immune response. NO from neuronal NOS is involved in
`neurotransmission in the central nervous system. These
`are only some of the multitude of biological roles for NO,
`which have been amply reviewed elsewhere [29–36].
`Because nitrates seem in many respects to act as
`exogenous NO sources,
`this would suggest a large
`number of potential
`therapeutic applications. Circum-
`venting nitrate tolerance in current cardiovascular thera-
`pies would be beneficial, but the newer applications of
`nitrates outside of angina and cardiovascular indications
`hold the most exciting promise. A review of the novel
`nitrate therapies under development is beyond the scope
`of
`this article, but
`it
`is worthwhile examining the
`biotransformation of nitrate therapeutics, because if
`nitrates are indeed NO-donor drugs,
`these metabolic
`pathways must involve the chemical generation of the
`NO free radical.
`In addition to GTN, several classical nitrates, and
`nicorandil, the most extensive published data on nitrate
`metabolism is for the NO-NSAIDs and related hybrid
`
`Page 3 of 22
`
`

`

`Nitrates and NO release
`
`1125
`
`
`on
`was reported to release increased amounts of NO2
`addition of esterase [38]. Unfortunately,
`the frequent
`
`as a measure of NO
`reporting of the Griess test for NO2
`release obfuscates the analysis and understanding of the
`mechanisms of action and of biotransformation, and this,
`in part, has led to some confusion in the literature as to
`the metabolic pathway that leads to NO release from
`these nitrates, with several claims that NO release simply
`requires the action of an esterase (e.g. [37]).
`The action of an esterase or nonenzymatic hydrolysis
`in the gut does not release NO, but generates a simple
`aliphatic nitrate, in the case of NCX 4016, hydroxyben-
`zyl nitrate (HBN), or, in the cases of NCX 4215 and
`AZD 3582, hydroxybutyl nitrate (Fig. 4). Therefore,
`focusing on the NO-ASA NCX 4016, one must query
`why a combination therapeutic of ASA and a benzyl
`nitrate does not provide similar benefits and even a
`clinical advantage in the ability to titrate the NO and
`NSAID effects independently (Fig. 4). In reviewing NO-
`NSAIDs in colon cancer chemoprevention, Rigas poses
`the same question of the effectiveness of classical nitrates
`in combination with NSAIDs in such indications [39,40].
`Ester hydrolysis of the NO-NSAID yields a primary
`aliphatic nitrate which may undergo subsequent reduc-
`
`) or nucleo-
`tive denitration to inorganic nitrite (NO2
`
`philic substitution to generate inorganic nitrate (NO3
`)
`(Fig. 4). There is ample evidence that a number of
`
`Fig. 3. General depiction of the mode of NO release from NO-NSAIDs
`and other hybrid nitrates.
`
`nitrates (Fig. 2). The latter two classes of agents are
`explicitly described in the literature as NO-donor drugs,
`but it must be emphasized that these compounds do not
`merely bresemble organic nitrates,Q they are nitrates and
`will manifest nitrate chemistry. NO-NSAIDs and related
`hybrid nitrates are proprodrugs, containing an ester
`moiety that is cleaved by nonspecific esterase action:
`the general depiction of such NO-donor drugs (using an
`NO-NSAID as an example in Fig. 3) is of an NO moiety
`(more accurately a nitrate moiety) connected via a linker
`to the parent pharmacophore. It
`is reported that
`the
`activity of NO-NSAIDs and hybrid nitrates in vitro is
`influenced by addition of esterases, for example, NCX
`1015 at high micromolar concentrations (100–500 AM)
`
`
`
`
`
`Fig. 4. (1) Biotransformation of an NO-NSAID therapeutic leading to eventual formation of inorganic nitrate, inorganic nitrite, and NO.
`A large number of enzymes, including cytochrome P450 oxidase, cytochrome P450 reductase, glutathione S-transferase, and aldehyde
`
`. Nucleophiles such as glutathione (GSH)
`dehydrogenase, have been demonstrated to mediate the reductive denitration of nitrates to NO2
`may displace inorganic nitrate via SN2 substitution. This analysis suggests that (2) a combination therapy might have similar
`effectiveness. (3) Hybrid nitrates, similarly, are designed to hydrolytically yield free thiol and simple aliphatic nitrate.
`
`Page 4 of 22
`
`

`

`1126
`
`G. R. J. Thatcher et al.
`
`
`
`reduction of
`diverse enzymes may catalyze the 2e
`
`
`nitrates to NO2
`is a good
`(vide infra), whereas NO3
`leaving group for nucleophilic substitution. In contrast,
`there is no unambiguous evidence for any enzyme
`
`reduction of nitrates to NO.
`catalyzing the direct 3e
`There is little evidence for significant biological activity
`
`,
`the levels of which in human
`associated with NO3
`plasma and rat aorta are relatively high (~25 and ~50
`AM) and fluctuate substantially with diet
`[41–43].
`However,
`there is renewed interest
`in the biological
`
`, which has been shown to be oxidized
`activity of NO2
`by peroxidase to NO2 [44,45] and to be reduced by
`ferrous hemoglobin (Hb) to yield NO-Hb or NO [43].
`
`have been
`Physiological concentrations of NO2
`reported to range from 0.5 to 25 AM in plasma and
`tissue and may be higher under pathophysiological
`
`has
`conditions [46,47]. The direct addition of NO2
`failed to elicit a biological response in several systems
`[41,48,49]; however, compartmentalization and active
`transport, combined with the demonstrated reactivity of
`
`with transition metal centers,
`leaves open a
`NO2
`
`, via its oxidation or
`its
`biological
`role for NO2
`reduction products [43,50–52]. Thus, it is possible that
`
`-donor drugs,
`nitrates act, not as NO donors, but as NO2
`
`to specific
`providing intracellular delivery of NO2
`locations.
`The metabolism of the NO-NSAID NCX 4016, has
`been studied in detail [53–57]. In rats, the intact nitrate,
`NCX 4016, was not detected in plasma after oral
`administration, and in rat liver microsomal S9 fractions,
`metabolism was complete within 5 min. The sole
`observable nitrate in rat liver fractions was HBN and
`the glutathione adduct shown in Fig. 4. Compatible with
`these observations, after administration in rats, plasma
`
`levels were unchanged (~1 AM), whereas plasma
`NO2
`
`was elevated from ~20 to ~130 AM over 6 h.
`NO3
`Reports of detection of NO itself as a product of in vivo
`administration of NO-NSAIDs are limited: NO-Hb was
`observed in rat blood and NO-Mb (myoglobin) in
`myocardial tissue in vivo and in vitro after administration
`of NCX 4016, using the sensitive ESR spectroscopic
`detection technique. Using this method, ISMN and NCX
`4016 (1 mM) generated similar levels of NO-Hb (~30 AM
`at 1 h) in rat blood. Because ferrous-Hb is known to react
`
`to generate NO-
`directly with nitrates, nitrites, and NO2
`Hb, it is difficult conclude from these experiments that the
`NO-NSAID nitrates are directly reductively biotrans-
`formed to NO in vivo. Thus, the best studied NO-NSAID,
`NCX 4016, is claimed to be largely absorbed in the small
`intestine, subject to substantial first-pass metabolism, and
`to possess little or no systemic bioavailability. Exciting
`data on the CNS activity of the NO-flurbiprofen NO-
`NSAIDs have been assembled [15,58], and like NCX
`4016, the nitrooxybutyloxy ester of flurbiprofen under-
`
`goes rapid, extensive first-pass metabolism in rats, such
`that: (i) only flurbiprofen could be detected in brain and
`
`+
`plasma in significant amounts and (ii) levels of (NO2
`
`) increased in plasma over 8 h, but were unchanged
`NO3
`in brain [59]. The hydroxybutyl nitrate metabolite was not
`quantified in this study.
`The wealth of exciting biological data being generated
`on NO-NSAIDs and on hybrid nitrates demonstrates
`enhanced and additional biological activity over the
`parent pharmacophores, which would be very difficult
`to explain simply by enhanced bioavailability deriving
`from either the increased lipophilicity of the nitrate
`derivative or the increased local blood flow due to nitrate
`vasodilation. The published data show that NO-NSAIDs
`are rapidly metabolized to simple aliphatic nitrates, and in
`
`.
`the case of NCX 4016, the major NOx metabolite is NO3
`Classical nitrates are themselves simple aliphatic nitrates:
`GTN is rapidly metabolized to yield further simple
`
`. Therefore, metab-
`aliphatic nitrate metabolites and NO2
`olism studies confirm that nitrates, whether classical or
`
`
`.
`and NO3
`modernist, are efficiently metabolized to NO2
`In contemporary nitrate therapeutics, hybrid nitrates in
`which the nitrooxy group is ancillary to a primary
`pharmacophore have generated the majority of reports;
`however, Schwarz Pharma scientists were the first to
`describe hybrid nitrates that contain a cysteine moiety,
`designed to be metabolized to free cysteine and a free,
`simple aliphatic nitrate (Fig. 4) [60,61]. This design is
`based upon the postulates either that free thiol is needed
`for biotransformation of nitrates or that thiol overcomes
`nitrate tolerance. A natural extension of this design is a
`hybrid nitrate which will release a free thiol other than
`cysteine, for example thiosalicylic acid (TSA), which
`like cysteine is an bactive thiolQ toward nitrates (vide
`infra) (Fig. 4) [62,63].
`Hybrid nitrates are uniformly designed to metabol-
`ically degrade to yield a simple aliphatic nitrate (and
`thence NO), and observations show that much of the
`therapeutically beneficial biological activity derives from
`the aliphatic nitrate moiety of such drugs. It is therefore
`reasonable to explore the structure–activity relationships
`of aliphatic nitrates themselves toward NO mimetic
`biological activity and toward metabolism or degradation
`to NO, by design of novel nitrates through incorporation
`of other
`functional groups into the organic nitrate
`[26,64,65]. These nitrates are predicted to show a
`different spectrum of NO mimetic activity to simple or
`classical nitrates, which has been borne out in literature
`reports. The disulfanyl S-nitrate GT 715 has been
`reported: (a) to reduce the cerebral infarct in the rat
`middle cerebral artery occlusion model of ischemic
`stroke by 60–70% when delivered 4 h after the onset
`of ischemia [66] and (b) to reverse the scopolamine-
`induced cognitive deficit in the Morris water task rat
`
`Page 5 of 22
`
`

`

`Nitrates and NO release
`
`1127
`
`behavioral model of dementia [13]. Interestingly, these
`studies showed that the S-nitrate was more effective and
`relatively more potent than GTN in activating hippo-
`campal sGC and demonstrated greatly attenuated periph-
`eral vasodilatory responses compared to GTN. A
`member of the S-nitrate family is currently in clinical
`trials for Alzheimer’s disease.
`
`NITRATE BIOTRANSFORMATION AND TOLERANCE
`
`Mechanism-based biotransformation of nitrates has
`been defined as the pathway generating a proximal
`activator of sGC in vascular smooth muscle cells, which
`is presumed to be NO, whereas the term clearance-based
`biotransformation has been coined to describe pathways
`that lead to nitrate metabolism without activation of sGC,
` [67,68]. The biotransfor-
`often presumed to be via NO2
`mation of GTN yields the dinitrate metabolites glyceryl-
`1,2-dinitrate (1,2-GDN) and glyceryl-1,3-dinitrate (1,3-
`GDN) as products. The vascular biotransformation of
`GTN is regioselective for formation of 1,2-GDN [69–
`74], and it is this isomer that is formed exclusively during
`the initial exposure of tissues to GTN [75]. In tolerant
`cultured cells and tissues, the regioselectivity is lost and
`GTN biotransformation is attenuated [76–79]. Several
`proteins have been identified that are capable of
`mediating the denitration of GTN, yielding GDN and
` as products, including Hb, Mb, xanthine oxidor-
`NO2
`eductase (XOR), old yellow enzyme, glutathione S-
`transferase (GST), cytochrome P450 oxidase (CYP), and
`cytochrome P450 reductase [73,74,78,80–87]. More
`recently, it was confirmed that ALDH2 was capable of
`mediating denitration of classical nitrates [88–90].
`Long-term administration of nitrates is characterized
`by a decrease in their vasodilator and antianginal
`effectiveness, which is termed nitrate tolerance. Clinical
`tolerance can be simulated in animal models by
`continuous delivery of GTN, but not by repetitive bolus
`administration [91]. There have been proposed two major
`hypotheses for nitrate tolerance: blood vessel-dependent
`and vessel-independent mechanisms. The blood vessel-
`dependent hypothesis accounts for
`the inability of
`vascular smooth muscle cells to adequately respond to
`nitrates, whereas the vessel-independent hypothesis
`accounts for the activation of neurohormonal counter-
`regulatory mechanisms (e.g.,
`intravascular volume
`expansion, increase in catecholamine levels, and activa-
`tion of the renin–angiotensin system) [92]. A major
`criticism of the involvement of neurohormonal mecha-
`nisms is that tolerance can be induced in isolated vessels
`in which the presence of circulating vasoconstrictor
`hormones is excluded, although the possibility exists
`that
`in vitro tolerance models do not reflect clinical
`nitrate tolerance [93].
`
`S
`
`S
`
`S
`
`tolerance mecha-
`Several blood vessel-dependent
`nisms have been suggested: (i) attenuation of nitrate
`bioactivation [67,68,94,95],
`(ii) generation of
`free
`radicals during nitrate biotransformation [69,92,95–97],
`(iii) upregulation of phosphodiesterase activity with the
`subsequent reduction in intracellular levels of cGMP
`[98], (iv) loss of sGC responsiveness to NO [99–101],
`and (v)
`intracellular depletion of
`thiol equivalents
`[102,103]. The observation that spontaneous NO donors
`(e.g., nitroprusside, nitrosothiols, and diazeniumdiolates)
`are not subject to tolerance or to cross-tolerance toward
`nitrates has been argued to support
`the attenuated
`mechanism-based biotransformation of nitrates in toler-
`ant vessels as the mechanism underlying tolerance. The
`thiol depletion hypothesis of Needleman and Johnson
`proposed that tolerance might result from depletion of
`critical thiols which are involved in the metabolism of
`nitrates [102]. However, a causal role in clinical nitrate
`tolerance due to reduced bioavailability of sulfhydryl
`groups was not confirmed by more recent investigations
`[104,105].
`A proposed free radical hypothesis of nitrate toler-
`ance is based on the observation that administration of
`,
`GTN increases the endothelial production of O2
`although this observation does not seem to hold for
`can
`nitrates in general [72,106]. The elevation of O2
`limit
`the bioavailability of both endogenous and
`exogenous NO through the near diffusion-controlled
`with NO yielding peroxynitrite
`reaction of O2
`
`), which has been reported to be formed during
`(ONOO
`the uninterrupted administration of nitrates in both
`animals and humans [95,106,107]. Peroxynitrite is an
`oxidizing and nitrating cytotoxin that has been assigned
`roles in nitrate tolerance through interference with the
`normal function of NOS and cGMP-dependent protein
`kinase I [94,95].
`The loss of regioselectivity for GTN denitration in
`tolerant tissue is dramatic and might be taken to suggest
`that the mechanism-based biotransformation pathway is
`regioselective. However, it is quite reasonable to expect
`clearance-based metabolism also to show regioselectivity
`and for both pathways to be attenuated in tolerant tissue.
`For example, the selective flavoprotein inhibitor diphe-
`nyleneiodonium sulfate (DPI) was shown to attenuate the
`regioselectivity for formation of 1,2-GDN,
`to inhibit
`cGMP accumulation in blood vessels, and not to inhibit
`tissue relaxation induced by spontaneous NO donors, but
`DPI inhibited GTN-induced relaxation to the same extent
`in aortae from naRve and from GTN-tolerant animals
`[71,78]. These observations force the conclusion that
`flavoproteins are important for nitrate metabolism and
`may participate in biotransformation leading to vaso-
`dilation, but their alteration cannot be the basis for nitrate
`tolerance. Regioselectivity is not therefore, by itself, a
`
`Page 6 of 22
`
`

`

`1128
`
`G. R. J. Thatcher et al.
`
`sufficient criterion for identification of mechanism-based
`biotransformation. ALDH2 activity in hepatic mitochon-
`dria also shows significant regioselectivity toward 1,2-
`GDN formation from GTN, which is lost in mitochondria
`from tolerant tissue [88,108]. Owing to the perceived
`importance of ALDH2 in GTN metabolism, the data on
`ALDH2 need further discussion [109].
`In rabbit aorta in which in vitro tolerance was induced
`by soaking of tissues in high GTN concentration, ALDH
`activity was reduced, and in addition, compounds that
`inhibit ALDH activity (cyanamide, aldehydes and their
`hydrates, at 1 mM) inhibited GTN biotransformation,
`relaxation, and GTN-induced cGMP accumulation in
`isolated blood vessels [88]. One might reasonably take
`these data to suggest a role for ALDH2 in nitrate
`bioactivation. In vivo studies are confounded because the
`inhibitors currently used are metabolically and redox
`labile (including aldehydes/hydrates;
`isoflavones, e.g.,
`daidzein; and thiocarbamate inhibitors, e.g., disulfiram,
`benomyl): many require biotransformation themselves
`and are not specific for ALDH2, having been shown to
`inhibit other sulfhydryl-dependent enzymes and metal-
`loenzymes including CYPs [110–115]. The lack of
`specificity of enzyme inhibitors has been commented
`on as a general problem, not
`limited to studies on
`ALDH2, in the identification of the enzyme responsible
`for nitrate biotransformation and tolerance [68].
`Continuous delivery of GTN via a transdermal patch
`in rats provides an in vivo model of clinical nitrate
`tolerance, validated by the markedly reduced blood
`pressure response to GTN observed in the intact animal
`[116]. Both propionaldehyde and cyanamide attenuated
`the regioselectivity of GTN biotransformation in hepatic
`mitochondria, and both compounds inhibited tissue
`relaxation induced by GTN in naRve aortae; importantly,
`however, both compounds also inhibited relaxation in
`tolerant aortae to the same extent as in naRve aortae (Fig.
`5) [108]. These and other data are compatible with a role
`for ALDH2 in mechanism- and clearance-based nitrate
`biotransformation, but are incompatible with a dominant
`role for ALDH2 in nitrate tolerance [108]. The actions of
`cyanamide and aldehydes on blood pressure and tissue
`relaxation were taken to reflect a nonspecific inhibitory
`action on GTN-induced relaxation. Furthermore,
`the
`numerous biochemical pathways that potentially contrib-
`ute to nitrate tolerance would suggest that this is likely a
`multifactorial phenomenon, which includes attenuation
`of one or more mechanism-based biotransformation
`pathways.
`The concentration–response curves shown in Fig. 5
`represent
`typical data on the effect of in vivo GTN
`tolerance on GTN-induced tissue relaxation. In GTN-
`tolerant tissue, potency is reduced by 5- to 10-fold, but
`the potency of GTN still exceeds that of other nitrate
`
`Fig. 5. Effect of in vivo GTN tolerance, 1.0 mM propionaldehyde
`(PRO), or both on GTN-induced relaxation of isolated rat aorta. The
`EC50 values for GTN-induced relaxation were Control, 19.1 F 8.3 nM;
`Tolerant, 81.4 F 57 nM; Control + PRO, 0.42 F 0.06 AM; Tolerant +
`PRO, 2.2 F 1.5 AM. Each value represents the mean F SEM (n = 3 or
`4). All EC50 values were significantly different from each other ( p b
`.001, one-way ANOVA). Arrows represent fold-shifts on treatment with
`propionaldehyde in naive and tolerant tissues.
`
`vasodilators, such as ISDN, in nontolerant tissue. It is
`therefore tempting to envision multiple biotransformation
`pathways, with GTN utilizing a high-potency or high-
`affinity pathway, sensitive to tolerance, in addition to the
`low-affinity pathway open to all other nitrates. Biphasic
`concentration–response curves have been reported for
`GTN-induced relaxation and these have supported a
`dual-mechanism hypothesis for GTN, with high-affinity
`and low-affinity components [117–119]. It is perplexing
`that erithrityl tetranitrate, which incorporates the GTN
`moiety in its structure, is reported to be 10-fold less
`potent, to manifest no biphasic activity; and to abolish
`the putative high-affinity pathway for GTN [120].
`If (a) attenuated mechanism-based biotransformation
`is responsible for nitrate tolerance and (b) nitrate activity
`is mediated through NO release, the enzyme or agent
`responsible for this biotransformation must be able to
`mediate chemical reduction of nitrates to NO.
`
`NITRATES AS NITRIC OXIDE DONORS
`
`Detection and quantification of NO in complex
`biological systems and under physiological conditions
`remain challenging. A variety of methods are used for
`detecting NO, for example: quantification of cGMP;
`chemiluminescence detection;
`fluorescence detection
`using 4,5-diaminofluorescein; conversion of Fe(II)-
`oxyHb to metHb; spin trap/ESR detection (using, e.g.,
`Fe-diethylthiocarbamate (Fe(DETC)2), Fe(II)Hb,
`Fe(II)Mb, 2-phenyl-4,4,5,5-tetramethylimidazoline-1-
`oxyl-3-oxide); electrochemical detection (amperometric
`NO-specific and porphyrinic electrodes); Griess assay of
`
`Page 7 of 22
`
`

`

`Nitrates and NO release
`
`1129
`
`); and immunoassays usingNO2, NO3, or (NO2 + NO3
`
`
`
`
`anti-nitrotyrosine antibodies. Many of these methods are
`indirect and have been queried, for example, because of
`lack of specificity for NO and interference from other
`components of biological systems [121–125].
`The release of NO from nitrates in complex biological
`systems can often be observed using a variety of different
`methods, for example, exhaled NO levels were observed
`to be increased after high-dose i.v. bolus administration of
`GTN to coronary artery bypass patients [126]. In human
`plasma, NO release was observ