ANTIOXIDANT THERAPIES FOR TRAUMATIC BRAIN INJURY

Abstract

Free radical-induced oxidative damage reactions, and membrane lipid peroxidation (LP) in particular, are one of the best validated secondary injury mechanisms in preclinical traumatic brain injury models. In addition to the disruption of the membrane phospholipid architecture, LP results in the formation of cytotoxic aldehyde-containing products that bind to cellular proteins and impair their normal functions. This article reviews the progress over the past three decades in regards to the preclinical discovery and attempted clinical development of antioxidant drugs designed to inhibit free radical-induced LP and its neurotoxic consequences via different mechanisms including the O₂^•- scavenger superoxide dismutase (SOD) and the lipid peroxidation inhibitor tirilazad. In addition, various other antioxidant agents that have been shown to have efficacy in preclinical TBI models are briefly presented such as the LP inhibitors U83836E, resveratrol, curcumin, OPC-14177 and lipoic acid; the iron chelator deferoxamine and the nitroxide-containing antioxidants such as α-phenyl-tert-butyl nitrone and tempol. A relatively new antioxidant mechanistic strategy for acute TBI is aimed at the scavenging of aldehydic LP by-products that are highly neurotoxic with “carbonyl scavenging” compounds. Finally, it is proposed that the most effective approach to interrupt posttraumatic oxidative brain damage after TBI might involve the combined treatment with mechanistically-complementary antioxidants that simultaneously scavenge LP-initiating free radicals, inhibit LP propagation and lastly remove neurotoxic LP byproducts.

INTRODUCTION

At present, there are no FDA-approved pharmacological therapies for acute treatment of traumatic brain injury (TBI) patients that are conclusively proven to mitigate the often devastating neurological effects of their injuries. Nevertheless, the possibility of an effective treatment is based upon the fact that even though some of the neural injury is due to the primary mechanical events (i.e. shearing of nerve cells and blood vessels), the majority of post-traumatic neurodegeneration is due to a pathochemical and pathophysiological cascade of secondary events occurring during the first minutes, hours and days following the injury which exacerbate the damaging effects of the primary injury. Arguably, one of the most validated “secondary injury” mechanisms revealed in experimental TBI studies involves oxygen radical-induced oxidative damage to lipids, proteins and nucleic acids. This review briefly outlines the key sources of reactive oxygen species (ROS) including their derived highly reactive free radicals, the mechanisms associated with their neural damage and the past, present and future of pharmacological antioxidants that should be able to produce a clinically demonstrable neuroprotective effect, if properly applied.

OXIDATIVE DAMAGE IN TRAUMATIC BRAIN INJURY

Superoxide Radical

The first body of work showing a role of oxygen radicals in acute TBI pathophysiology was conducted by Kontos and colleagues who demonstrated an almost immediate post-injury increase in brain microvascular superoxide radical (O₂^•-) production associated with compromise of autoregulatory function in fluid percussion TBI models ^{1, 2}. These early investigators also demonstrated that scavengers of O₂^•- decrease the post-traumatic superoxide levels and protect against the loss of microvascular autoregulatory competency. Within the injured nervous system, a number of possible sources of O₂^•- may be operative during the first minutes and hours after injury including: the arachidonic acid cascade (i.e. prostaglandin synthase and 5-lipoxygenase activity), enzymatic or autoxidation of biogenic amine neurotransmitters (e.g. dopamine, norepinephrine, 5-hydroxytryptamine), “mitochondrial leak”, xanthine oxidase activity and the oxidation of extravasated hemoglobin. Activated microglia and infiltrating neutrophils and macrophages provide additional sources of O₂^•- at later timepoints.

Superoxide, which is formed by the single electron reduction of oxygen, may act as either an oxidant or reductant. While, O₂^•- itself is reactive, its direct reactivity toward biological substrates in aqueous environments is relatively weak. Moreover, once formed, O₂^•- undergoes spontaneous dismutation to form hydrogen peroxide (H₂O₂) in a reaction that is markedly accelerated by the enzyme superoxide dismutase (SOD): O₂₋^{• -} + O_2⁻ + 2H⁺ ➔ H₂O₂ + O₂ ³. In solution, O₂^•- actually exists in equilibrium with the hydroperoxyl radical (HO₂·): O₂^{• -} + H⁺ ➔ HO₂•, which is considerably more lipid soluble and a far more powerful oxidizing or reducing agent (9). Since the pKa of the O₂^•-/HO₂• is 4.8, as the pH of a solution falls (i.e. tissue acidosis), the equilibrium between O₂^{• -} and HO₂· shifts in favor of HO₂^• which is much more reactive than O₂^•-, particularly toward lipids

Iron and Hydroxyl Radical

The CNS is an extremely rich source of iron and its regional distribution varies in parallel with the sensitivity of various regions to oxidative damage ⁴. Under normal circumstances, low molecular weight forms of redox-active iron are maintained at extremely low levels. In plasma, the iron transport protein transferrin tightly binds iron in the Fe⁺⁺⁺ form. Intracellularly, Fe⁺⁺⁺ is sequestered by the iron storage protein ferritin. While both ferritin and transferrin have very high affinity for iron at neutral pH and effectively maintain iron in a non-catalytic state ³, both proteins readily give up their iron at pH values of 6.0 or less which is a level of acidosis that has been shown to be reached in the injured brain. Once iron is released from ferritin or transferring, it can actively catalyze oxygen radical reactions. Therefore, within the traumatized brain, where pH in injured areas is typically lowered, conditions are favorable for the potential release of iron from storage proteins ³. In the case of ferritin, its iron can also be released by reductive mobilization by O₂^{• -}.

A second source of catalytically active iron is hemoglobin. Hemorrhage resulting from mechanical trauma provides an obvious source of hemoglobin. While hemoglobin itself has been reported to stimulate oxygen radical reactions, it is more likely that iron released from hemoglobin is responsible for hemoglobin-mediated oxidative damage ^{5, 6}. Iron is released from hemoglobin by either H₂O₂ or by lipid hydroperoxides (LOOH; see below) and this release is further enhanced as the pH falls to 6.5 or below. Therefore, hemoglobin may catalyze oxygen radical formation and LP either directly or through the release of iron by H₂O₂, LOOH and/or acidic pH.

Free iron or iron chelates participate in free radical reactions at two levels. The autoxidation of Fe⁺⁺ results in the formation of O₂^{• - 3}: Fe⁺⁺ + O₂ ➔ Fe⁺⁺⁺ + O₂^{• -}. Secondly, Fe⁺⁺ is also oxidized in the presence of H₂O₂ to form hydroxyl radical (•OH) (Fenton reaction): Fe⁺⁺ + H₂O₂ ➔ Fe⁺⁺⁺ + •OH + OH⁻. Using the salicylate trapping method, a rise in brain ·OH levels has also been documented in a mouse diffuse and rat focal TBI models by the senior author ^{7, 8} and others ^8–10. As with the work of Kontos and colleagues discussed above, the cerebral microvasculature appears to be the initial source of post-traumatic ·OH production.

Chemistry of Lipid Peroxidation and Target Mechanisms for it’s Pharmacological Inhibition

The most studied mechanism of oxidative damage in models of TBI concerns free radical-induced lipid peroxidation (LP). The process of LP is presented in Figure 1 in the context of the ●OH-induced peroxidation of the LP-susceptible arachidonic acid (AA) which is highly enriched in brain cell membranes. “Initiation” of LP occurs when a radical species such as ●OH reacts with and removes an allylic carbon (carbon surrounded by adjacent double bonds) and extracts a hydrogen and it’ single electron from AA (AA + R• ➔ AA• + RH). In the process, the initiating radical is quenched by receipt of an electron (hydrogen) from the polyunsaturated AA. This, however, converts the AA into a lipid or “alkyl” radical (AA●). This sets the stage for a series of “propagation” reactions which begins when the alkyl radical takes on a mole of oxygen creating a lipid peroxyl radical (AA-OO•; AA• + O₂ ➔ AA-OO•). The peroxyl radical then reacts with a neighboring AA within the membrane and steals its electron forming a lipid hydroperoxide (AA-OOH) and a second alkyl radical (AA•; AA-OO• + AA ➔ AA-OOH + AA•).

1.

Chemistry involved in the initiation, propagation and termination reactions of arachidonic acid during lipid peroxidation with the resulting formation of the aldehydic end-product 4-hydroxynonenal (4-HNE).

Once LP begins the propagation phase, iron may participate in driving the process as lipid •hydroperoxides are decomposed by reactions with either ferrous iron (Fe⁺⁺), ferric iron (Fe⁺⁺⁺). In the case of Fe⁺⁺, the reaction results in formation of a lipid alkoxyl radical (AA-O•; AA-OOH + Fe⁺⁺ ➔ AA-O• + OH⁻ + Fe⁺⁺⁺). If, however, the reaction involves Fe⁺⁺⁺, the AA-OOH is converted back into a lipid peroxyl radical (AA-OO•; AA-OOH + Fe⁺⁺⁺ ➔ AA-OO• + Fe⁺⁺). Both of the reactions of AA-OOH with iron have acidic pH optima causing them to be augmented by tissue acidosis. Either alkoxyl (AA-O^·) or peroxyl (AA-OO^·) radicals arising from AA-OOH decomposition by iron can initiate so called lipid hydroperoxide-dependent LP resulting in “chain branching” reactions: (AA-OO• + AA ➔ AA-OOH + AA• or AA-O• + AA ➔ AA-OH +AA•).

Ultimately, the LP process leads to “fragmentation” or “scission” reactions in which the peroxidized AA breaks down to give rise to the neurotoxic aldehydes 4-HNE or 2-propenal (acrolein). The 4-HNE (as well as acrolein) produces neurotoxicity by binding to basic amino acids such as lysine or histidine as well as sulfhydryl-containing cysteine residues in cellular proteins as illustrated in Figure 2. The resulting chemical modifications have been shown to inhibit the function of a variety of structural and enzymatic cellular proteins.

2.

Chemical reactions of 4-HNE with amino acids that lead to impairment of protein structure and function.

Other Forms of Oxidative Damage

The central nervous system is exquisitely sensitive to LP because of its high content of peroxidation-susceptible lipids such as AA, linoleic acid, linolenic acid and docosahexaenoic acid and the high levels of iron. While LP disrupts the normal phospholipid architecture of cellular and subcellular organellar membranes, end-products of LP, most notably 4-HNE and acrolein, can bind to proteins, modifying their structure and compromising function. However, primary radical-mediated oxidative damage can also occur in proteins. For instance, iron-catalyzed, •OH mechanisms can target certain basic amino acids (e.g. lysine, arginine, histidine) leading to the formation of “protein carbonyl” moieties. Another form of protein oxidative damage involves the oxidation of cysteine sulfhydryl groups which can lead to the formation of abnormal disulfide bridges.

Nucleic acids, both DNA and RNA, are also susceptible to oxidative medication by inorganic and organic (i.e. lipid) radicals. In addition to potentially compromising DNA replication, transcription and mRNA translation, DNA oxidative damage also triggers DNA repair mechanisms that can greatly stress cellular function and survival. One such mechanism concerns the activation of poly ADP ribose polymerase (PARP) whose action can lead to severe depletion of cellular stores of ATP. In addition, DNA-protein cross-linking can occur (e.g. thymine-tyrosine) ³. However, compared to the numerous studies that have documented post-traumatic LP and protein oxidative damage in TBI models, very little examination of nucleic acid oxidation has occurred.

Peroxynitrite

Nearly 20 years ago, Beckman and coworkers introduced the theory that the principal ROS involved in producing tissue injury in a variety of neurological disorders is the “reactive nitrogen species” peroxynitrite (PN; ONOO⁻), which is formed by the combination of NOS-generated ·NO radical and O₂^•-: O₂^•- + •NO ➔ ONOO^{− 11}. Since that time, the biochemistry of PN has been in large part defined. PN-mediated oxidative damage is actually caused by PN decomposition products that possess potent free radical characteristics. These are formed in one of two ways. The first involves the protonation of PN to form peroxynitrous acid (ONOOH) which can undergo homolytic decomposition to form the highly reactive nitrogen dioxide radical (·NO₂) and ·OH; (ONOOH ➔ ·NO₂ + ·OH). Perhaps more important physiologically, PN will react with carbon dioxide (CO₂) to form nitrosoperoxocarbonate (ONOOCO₂) which can decompose into ·NO₂ and carbonate radical (·CO₃); (ONOOCO₂ ➔ ·NO₂ + ·CO₃).

Each of the PN-derived radicals (·OH, ·NO₂ and ·CO₃) can initiate LP cellular damage by abstraction of an electron from a hydrogen atom bound to an allylic carbon in polyunsaturated fatty acids or cause protein carbonylation by reaction with susceptible amino acids (e.g. lysine, cysteine, arginine). Additionally, ·NO₂ can nitrate the 3 position of tyrosine residues in proteins; 3-NT is a specific footprint of PN-induced cellular damage. Peroxynitrite-mediated protein nitration can involve the initial oxidation of a tyrosine moiety by a lipid peroxyl or alkoxyl radical followed by nitration by •NO₂.

The implication of PN in post-TBI pathophysiology is derived from four lines of evidence. First of all, all three NOS isoforms (endothelial, neuronal and inducible) are known to be up regulated during the first 24 hrs after TBI in rodents ^12–14. Secondly, several laboratories have shown that the acute treatment of injured mice or rats with NOS inhibitors can exert a neuroprotective effect and/or improve neurological recovery ^15–22. Thirdly, biochemical footprints of PN-mediated damage have been documented in rodent TBI paradigms including an increase in 3-NT levels ^{15, 21}, and ADP ribosylation (evidence of PARP activation). Fourthly, the notion that these markers of PN-mediated damage are pathophysiologically important is supported by the finding that the NOS inhibitor L-NAME can lessen the accumulation of 3-NT in injured brains ^{15, 21} at the same doses which improve neurological recovery ²³.

MECHANISMS FOR PHARMACOLOGICAL INHIBITION OF OXIDATIVE DAMAGE IN TRAUMATIC BRAIN INJURY

Based upon this outline of the steps involved in oxygen radical-induced oxidative damage, and LP in particular, a number of potential mechanisms for its inhibition are apparent which fall into three categories. The first category includes compounds that inhibit the initiation of LP and other forms of oxidative damage by preventing the formation of ROS or RNS species. For instance, NOS inhibitors, discussed above exert an indirect antioxidant effect by limiting •NO production and thus PN formation. However, they also have the potential to interfere with the physiological roles that •NO is responsible for including antioxidant effects which are due it’s important role as a scavenger of lipid peroxyl radicals (e.g. AAOO• + •NO ➔ AAOONO) ²⁴. Another approach to blocking posttraumatic radical formation is the inhibition of the enzymatic (e.g. cyclooxygenase, 5-lipoxygenases) AA cascade during which the formation of O₂^•- is produced as a by-product of prostanoid and leukotriene synthesis. Kontos and colleagues ^{2, 25} and Hall and coworkers ^{26, 27} have shown that cyclooxygenase inhibiting non-steroidal anti-inflammatory agents (e.g. indomethacin, ibuprofen) are vaso-and neuro-protective in TBI models.

A second indirect LP inhibitory approach involves chemically scavenging the radical species (e.g. O₂^•-, •OH, •NO₂, •CO₃) before they have a chance to steal an electron from a polyunsaturated fatty acid and thus initiate LP. The use of pharmacologically-administered SOD represents an example of this strategy. Another example concerns the use of the nitroxide antioxidant tempol which has been shown to catalytically scavenge the PN-derived free radicals •NO₂ and •CO₃ ²⁸. In either case, a general limitation to these first two approaches is that they would be expected to have a short therapeutic window and would have to be administered rapidly in order to have a chance to interfere with the initial posttraumatic “burst” of free radical production that has been documented in TBI models ^{2, 29}. While it is believed that ROS, including PN production persists several hrs after injury, the major portion is an early event that peaks in the first 60 minutes after injury making it clinically impractical to pharmacologically inhibit, unless the antioxidant compound is already “on board” when the injury occurs or available for administration immediately thereafter.

In contrast to the above indirect-acting antioxidant mechanisms, the third category involves stopping the “chain reaction” propagation of LP once it has begun. The most demonstrated way to accomplish this is by scavenging of lipid peroxyl (LOO•) or alkoxyl (LO•) radicals. The endogenous scavenger of these lipid radicals is alpha tocopherol or vitamin E (Vit E) which can donate an electron from its phenolic hydroxyl moiety to quench LOO●. However, the scavenging process is stoichiometric (1 Vit E can only quench 1 LOO•) and in the process vitamin E loses its antioxidant efficacy and becomes Vitamin E radical (LOO• + Vit E ➔ LOOH + Vit E•). Although Vit E• is relatively unreactive (i.e. harmless), it also cannot scavenge another LOO• until it is reduced back to its active form by receiving an electron from other endogenous antioxidant reducing agents such as ascorbic acid (Vitamin C) or glutathione (GSH). While this tripartite LOO• antioxidant defense system (Vit E, Vit C, GSH) works fairly effectively in the absence of a major oxidative stress, numerous studies have shown that each of these antioxidants is rapidly consumed during the early min. and hrs. after TBI. Thus, it has long been recognized that more effective pharmacological LOO• and LO• scavengers are needed. Furthermore, it is expected that compounds that could interrupt the LP process after it has begun would be able to exert a more practical neuroprotective effect (i.e. possess longer antioxidant therapeutic window).

A second approach to inhibiting the propagation of LP reactions is to chelate free iron, either ferrous (Fe⁺⁺) or ferric (Fe⁺⁺⁺), which potently catalyzes the breakdown of lipid hydroperoxides (LOOH), an essential event in the continuation of LP chain reactions in cellular membranes. The prototypical iron-chelating drug which chelates Fe⁺⁺⁺, is the bacterially (streptomyces pilosus)-derived tri-hydroxamic acid compound deferoxamine.

NEUROPROTECTIVE EFFECTS OF PHARMACOLOGICAL ANTIOXIDANTS

TBI Clinical Trial Results with PEG-SOD, Tirilazad and Dexanabinol

During the past 25 years, there has been an intense effort to discover and develop pharmacological agents for acute treatment of TBI. This has included multiple compounds that possess free radical scavenging/antioxidant properties including polyethylene glycol-conjugated superoxide dismutase (PEG-SOD), the LP inhibitor tirilazad ^30–32 and more recently the mixed glutamate antagonist/antioxidant compound dexanabinol ³³. However, each of these trials was a therapeutic failure in that no overall benefit has been documented in moderate and severe TBI patient populations which was the primary goal in each case. These failures can be attributed to several factors. Perhaps most importantly, the preclinical assessment of compounds destined for acute TBI trials has often been woefully inadequate in regards to the definition of neuroprotective dose-response relationships, pharmacokinetic-pharmacodynamic correlations, therapeutic window and optimum dosing regimen and treatment duration. However, a number of other issues related to design of the clinical trials are also believed to be involved ³². The following sections briefly review the TBI histories of PEG-SOD and tirilazad. Dexanabinol (HU211) is not discussed further since it is a mixed glutamate antagonist/antioxidant compound that was studied very little in preclinical TBI paradigms prior to being the subject of clinical development for that indication.

PEG-SOD

As mentioned earlier, the earliest studies of free radical scavenging compounds in TBI models were carried out with Cu/Zn SOD based upon the work of Kontos and colleagues who showed that post-traumatic microvascular dysfunction was initiated by O₂^{• 2212} generated as a by-product of the arachidonic acid cascade which is massively activated during the first minutes and hours after TBI ^{1, 2, 25}. Their work showed that administration of SOD prevented the post-traumatic microvascular dysfunction. This lead to clinical trials in which the more metabolically stable polyethylene glycol (PEG)-conjugated SOD was examined in moderate and severe TBI patients when administered within the first 8 hrs after injury. Although an initial small phase II study showed a positive trend, subsequent multi-center phase III studies failed to show a significant benefit in terms of increased survival or improved neurological outcomes ³⁴. Although many explanations for these negative results may be postulated, one reason may be that a large protein like SOD is unlikely to have much brain penetrability and therefore its radical scavenging effects may be limited to the microvasculature. A second reason may be that attempting to scavenge the short-lived inorganic radical O₂^{• -} may be associated with a very short therapeutic window, as suggested above. Indeed, the time course of measurable post-traumatic •OH formation in the injured rodent brain has been shown to largely run its course by the end of the first hour after TBI ^{9, 29}. A more rational strategy would be to inhibit the LP that is triggered by the initial burst of inorganic radicals. A comparison of the time course of LP with that of post-traumatic •OH shows that LP reactions continue to build beyond the first post-traumatic hr ⁹ and may continue for 3–4 days ³⁵. Despite the failure of PEG-SOD in human TBI, experimental studies have shown that transgenic mice that over-express Cu/Zn SOD are significantly protected against post-TBI pathophysiology and neurodegeneration ^36–40. This fully supports the importance of post-traumatic O₂^{• -} in post-traumatic secondary injury, despite the fact that targeting this primordial radical is probably not the best antioxidant strategy for acute CNS injury compared to trying to stop the downstream LP process that is initiated by the early increases in O₂^{• -}, •OH, •NO₂ and •CO₃.

Tirilazad

Consistent with that rationale, the 21-aminosteroid LP inhibitor tirilazad (aka. U74006F) was discovered which inhibits free radical-induced LP by a combination of LOO• scavenging and a membrane-stabilizing action that limits the propagation of LP reactions between an LOO• and an adjacent polyunsaturated fatty acid. The protective efficacy of tirilazad has been demonstrated in multiple animal models of acute TBI in mice ⁴¹, rats ⁴² and cats ⁴³. While the compound is largely localized in the microvascular endothelium, the post-traumatic disruption of the BBB is known to allow the successful penetration of tirilazad into the brain parenchyma as noted earlier ⁴⁴. Other mechanistic data derived from the rat controlled cortical impact and the mouse diffuse concussive head injury models have definitively shown that a major effect of tirilazad is to lessen post-traumatic BBB opening ^{9, 44}.

Tirilazad was taken into clinical development in the early 1990s and following a small phase II dose-escalation study that demonstrated the drug’s safety in TBI patients was evaluated in two phase III multi-center clinical trials for it’s ability to improve neurological recovery in moderately and severely injured closed TBI patients. One trial was conducted in North America and the other in Europe. In both trials, TBI patients were treated within 4 hrs after injury with either vehicle or tirilazad (2.5 mg/kg i.v. q6h for 5 days). The North American trial was never published due to a major confounding imbalance in the randomization of the patients to placebo or tirilazad in regards to injury severity and pre-treatment neurological status. In contrast, the European trial had much better randomization balance and has been published ³⁰. The results failed to show a significant beneficial effect of tirilazad in either moderate (GCS = 9–12) or severe (GCS = 4–8) patient categories. However, a post hoc analysis showed that moderately-injured male TBI patients with traumatic SAH has significantly less mortality after treatment with tirilazad (6%) compared to placebo (24%, p<0.026). In severely injured males with tSAH (tSAH) tirilazad also lessened mortality from 43% in placebo-treated to 34% (p<0.071) ^{30, 45}. This result is consistent with the fact that this compound is also highly effective in animal models of SAH ⁷. Nevertheless, additional trials would have been required in order to establish the neuroprotective utility of tirilazad in certain human TBI subgroups and to gain FDA approval in the U.S. However, the sponsoring company Pharmacia & Upjohn opted not to continue the compound’s development for TBI although it was successfully approved and marketed for use in aneurysmal SAH in several western European and Australasian countries based upon its demonstrated efficacy in phase III SAH trials ^{46, 47}.

Effects of Direct and Indirect-Acting Lipid Peroxidation Inhibitors

In addition to tirilazad, several other LP inhibitors have been reported to be effective neuroprotectants in TBI models. These include the lipid peroxyl radical (LOO•) scavenging 2-methylaminochromans U-78517F and U-83836E ⁴⁸, the pyrrolopyrimidine U-101033E ^{48, 49}, OPC-14117 ⁵⁰ and the naturally-occurring LOO• scavengers curcumin ^{51, 52} and resveratrol ^{53, 54}, the indoleamine melatonin ^{21, 55–58} and lastly, the endogenous antioxidant lipoic acid ⁵⁹ (see Figure 3). In the case of curcumin and resveratrol, these are potent LOO• scavengers due to their possession of multiple phenolic hydroxyl groups that can donate electrons to LOO• radicals. Melatonin also has LOO• scavenging capability ⁶⁰, but in addition appears to react with PN ^{61, 62}. Lipoic acid may also have LOO• scavenging effects, but these are more likely to be indirect via the regeneration (i.e. re-reduction) of other endogenous electron-donating antioxidants including vitamin E, glutathione and vitamin C.

3.

Chemical structures of lipid peroxyl radical scavenging and nitroxide-containing antioxidants shown to be neuroprotective in TBI models.

Among these LP inhibitors, the most potent and effective LOO• scavenging LP inhibitors yet discovered is the 2-methylaminochroman compound U-83836E which combines the LOO• scavenging antioxidant chroman ring structure of vitamin E with the bis-pyrrolopyrimidine moiety of tirilazad. The phenolic chroman antioxidant moiety can be re-reduced by endogenous ascorbic acid (vitamin C) or glutathione (GSH) after it has donated its phenolic electron to an initial LOO• making it able to quench a second and then a third LOO•, etc. The bis-pyrrolopyrimidine moiety, on the other hand, can also scavenge multiple moles of LOO•, but by a true catalytic mechanism ^{49, 63}. Thus, U-83836E, is a dual functionality LOO• scavenger that is understandably more effective than either vitamin E, tirilazad ⁶³ and possibly the other naturally-occurring LOO• scavengers such as curcumin, resveratrol, melatonin and lipoic acid. Furthermore, U-83836E possesses a high degree of lipophilicity endowing it with a high affinity for membrane phospholipids where LP takes place. Recent studies from the authors’ laboratory in the mouse CCI-TBI model have shown that U-83836E is able to reduce post-traumatic LP and protein nitration and preserve mitochondrial respiratory function in injured cortical tissue and mitochondria ⁶⁴.

Nitroxide Antioxidants and Peroxynitrite Scavengers

In addition to the lipid peroxyl (LOO•) radical scavengers, the neuroprotective effects of a family of nitroxide-containing antioxidants have also been examined in experimental TBI models. These are sometimes referred to as “spin-trapping agents” and include α-phenyl-tert-butyl nitrone (PBN) and its thiol analog NXY-059 and tempol (see bottom of Figure 3). Both PBN and tempol have been shown to be protective in rodent TBI paradigms ^{65, 66}. As mentioned earlier, tempol has been shown by the author and colleagues to catalytically scavenge PN-derived ·NO₂ and ·CO₃ ^{28, 67}, and to reduce post-traumatic oxidative damage (both LP and protein nitration), preserve mitochondrial function, decrease calcium-activated, calpain-mediated cytoskeletal damage and reduce neurodegeneration in mice subjected to a severe controlled cortical impact-induced focal TBI ⁶⁸. Another laboratory has reported that tempol can reduce post-traumatic brain edema and improve neurological recovery in rat contusion injury model ^{69, 70}. However, the neuroprotective effect of tempol, administered alone, is associated with a therapeutic window of an hr or less in the mouse controlled cortical impact TBI (CCI-TBI) model. Moreover, tempol is not effective at directly inhibiting LP in the latter model ⁶⁸.

Effects of the Iron Chelator Deferoxamine

The prototype iron chelator deferoxamine which binds ferric (Fe⁺⁺⁺) iron and thereby would lessen the catalytic effects of iron on LP, has also been reported to have beneficial actions in preclinical TBI or TBI-related models ^{71, 72}. However, deferoxamine is hindered by its lack of brain penetration and rapid plasma elimination rate. To partially counter the latter limitation, a dextran-coupled deferoxamine has been synthesized that has been reported to significantly improve early neurological recovery in a mouse diffuse TBI model ⁷³. Much of this activity, however, is probably due to microvascular antioxidant protection because of limited brain penetrability. Another caveat to the iron-chelation antioxidant neuroprotective approach that is at least relevant to the ferric iron chelators such as deferoxamine is that they can cause a pro-oxidant effect in that their binding of Fe⁺⁺⁺ can actually drive the oxidation of ferrous to ferric iron which can increase superoxide radical formation in the process (Fe⁺⁺ + O₂ ➔ Fe⁺⁺⁺ + O₂^•-).

Carbonyl Scavenging as an Approach to Inhibit 4-HNE and Acrolein Binding to Proteins

As pointed out earlier (Fig. 2), the LP-derived aldehydic (carbonyl-containing) breakdown products 4-HNE and acrolein have high affinity for binding to selected protein amino acid residues including histidine, lysine, arginine and cysteine. These modifications have been shown to inhibit the activities of a variety of enzymatic proteins ³. Several compounds have been recently identified that are able to antagonize this “carbonyl stress” by covalently binding to reactive LP-derived aldehydes. For instance, D-penicillamine has been demonstrated to form an irreversible bond to primary aldehydes. We have previously demonstrated that penicillamine is able to scavenge PN ⁷⁴ and to protect brain mitochondria from PN-induced respiratory dysfunction in isolated rat brain mitochondria ⁷⁵. This latter action was observed along with an attenuation of 4-HNE-modified mitochondrial proteins ⁷⁵. The PN scavenging action of penicillamine along with its carbonyl scavenging capability may jointly explain our previous findings that acutely administered penicllamine can improve early neurological recovery of mice subjected to moderately severe concussive TBI ⁷⁶.

More recently, it has been demonstrated that a variety of hydrazine (-NH-NH₂)-containing compounds such as the anti-hypertensive agent hydralazine, the anti-depressant phenelzine and the anti-tubercular agent iproniazid can react with the carbonyl (CHO) moieties of 4-HNE or acrolein which prevents the latter from binding to susceptible amino acids in proteins ⁷⁷. Consistent with this effect being neuroprotective, others have shown that hydralazine inhibits either compression of acrolein-mediated injuries to ex vivo spinal cord ⁷⁸. Hydralazine, which is a potent vasodilator would be difficult to administer in vivo after either spinal cord injury or TBI in which hypotension is already a common pathophysiological problem. However, other hydrazine-containing compounds such as phenelzine and iproniazid do not compromise blood pressure as readily as hydralazine and have a long history of clinical use although never having been examined in acute neurotrauma models. Most impressive is the fact that the application of hydrazines can rescue cultured cells from 4-HNE toxicity even when administered after the 4-HNE has already covalently bound to cellular proteins ⁷⁷. Such an effect could be associated with a favorable neuroprotective therapeutic window.

RATIONALE FOR COMBINATION ANTIOXIDANT TREATMENT OF TBI

Antioxidant neuroprotective therapeutic discovery directed at acute TBI has consistently been focused upon attempting to inhibit the secondary injury cascade by pharmacological targeting of a single oxidative damage mechanism. As presented above, these efforts have included either enzymatic scavenging of superoxide radicals with SOD ³⁴ or inhibition of LP with tirilazad ³⁰. While each of these strategies has shown protective efficacy in animal models of TBI, phase III clinical trials with either compound failed to demonstrate a statistically significant positive effect although post hoc subgroup analysis suggests that the microvascularly localized tirilazad may have efficacy in moderate and severe TBI patients with tSAH ³⁰. While many reasons have been identified as possible contributors to the failure, one logical explanation has to with the possible need to interfere at multiple points in the oxidative damage portion of the secondary injury cascade either simultaneously or in a phased manner in order to achieve a clinically demonstrable level of neuroprotection.

In addition to the antioxidant strategy of scavenging the initiating radicals and stopping the propagation of LP reactions in the injured brain tissue, recent work has shown that carbonyl (CHO) scavenging compounds can also act to protect cellular proteins from the binding of neurotoxic LP-derived aldehydes. Thus, we are presently exploring the neuroprotective efficacy of three of the prototypes of this new class of compound alone and in combination with the PN-radical scavenging tempol and/or the LP-inhibiting U-83836E. Figure 4 summarizes the overall rationale for a multi-mechanistic antioxidant therapy for TBI. It is anticipated that the combination of two or three antioxidant mechanistic strategies may improve the extent of neuroprotective efficacy, lessen the variability of the effect and possibly provide a longer therapeutic window of opportunity compared to the window for the individual strategies. If these theoretical combinatorial benefits are confirmed in preclinical TBI models this should greatly enhance the chance of neuroprotective success in future clinical trials in contrast to previous failures with single antioxidant agents.

4.

Rationale for combination antioxidant therapy for TBI. Injury triggers an increase in cytoplasmic Ca⁺⁺ via voltage dependent and glutamate receptor-operated channels. The increase in intracellular Ca⁺⁺ initiates activation of cytoplasmic calpain. Mitochondrial Ca⁺⁺ uptake (buffering) stresses the mitochondria and contributes to mitochondrial dysfunction Specifically, Ca⁺⁺ uptake by the mitochondria leads O₂^•- leakage from the electron transport chain and activation of Ca⁺⁺-activated mitochondrial nitric oxide synthase (NOS). The O₂^•- and NO● combine to form the potent reactive nitrogen species PN which is able to which in turn gives rise to the highly reactive nitrogen dioxide (•NO₂), hydroxyl (•OH) and carbonate (•CO₃) radicals which cause oxidative damage to the mitochondria as well as other cellular structures due to PN’s large diffusion radius. When this becomes severe, there is a decrease in the mitochondrial ATP production and membrane potential (ΔΨ). This leads to catastrophic mitochondrial failure (mitochondrial permeability transition, MPT) and the dumping of mitochondrial Ca⁺⁺ into the cytoplasm where it exacerbates cytoplasmic calpain activation and proteolysis of a cytoskeletal proteins and other substrates. The combination of the antioxidant tempol which catalytically reacts with PN-derived radicals with a chain-breaking LP inhibitor such as U-83836E or a carbonyl (CHO) scavenging compound should produce a better neuroprotective effect than any of these compounds alone.