Potential Targets in the Treatment of Brain Stroke

The blood supply to the forebrain is derived from the two internal carotid arteries & the basilar artery. Their branches form anastomosis at the base of brain called Circle of Wills which comprises the anterior communicating artery, two anterior Cerebral arteries, internal carotids, two posterior communicating arteries & two Posterior cerebral arteries. Brain stem & cerebellum are supplied by vertebral & Basilar arteries and their branches.

The cerebral hemispheres are drained by superficial & deep cerebral veins. Superficial cerebral veins drain cerebral cortex & empty into dural venous sinuses. The deep cerebral veins unit as great cerebral vein & empty into straight Sinus.

CNS is covered by 3 meninges, dura mater, arachnoid mater and pia mater. Dura mater is the outermost tough fibrous coat which is fused with the inner periosteum of the skull except at site of reflection which contain venous sinuses. Beneath it lies the arachnoid mater which is a thin fibrocellular layer in direct contact with the dura mater. 

A subdural space can be created by leakage of blood following a tear of cerebral vein (subdural hematoma). The innermost meninges are the pia mater that is firmly adherent to the surface of brain & spinal cord. Between arachnoid & pia mater is subarachnoid space which is in communication with the cavities (ventricles) in the brain and contain CSF.

The ventricles include:

1- 2 lateral ventricles in cerebral hemispheres.

2- The third ventricle in the diencephalons.

3- The fourth ventricle in hind brain which is continuous with central canal of spinal cord. Third &fourth ventricle are connected in midbrain by aqueduct of Sylvius.

It denotes accumulation of CSF in the ventricular system. It results from obstruction of the normal CSF circulation commonly due to obstruction of the outlets from fourth ventricle to subarachnoid space.

Stroke 

Stroke is a disease that affects the blood vessels that supply blood to the brain (1, 2).

A stroke occurs when a blood vessel that brings oxygen and nutrients to the brain either bursts (hemorrhagic stroke) or is clogged by a blood clot or some other mass (ischemic stroke). When a rupture or blockage occurs, parts of the brain don't get the blood and oxygen they need. Without oxygen, nerve cells in the affected area of the brain can't work properly, and die within minutes. And when nerve cells can't work, the part of the body they control can't work either. The devastating effects of a severe stroke are often permanent because dead brain cells aren't replaced (1, 2)

There are two main types of stroke. One (ischemic stroke) is caused by blockage of a blood vessel; the other (hemorrhagic stroke) is caused by bleeding. Bleeding strokes have a much higher death rate than strokes caused by clots.

Ischemic stroke 

Ischemic stroke is the most common type. It accounts for about 87 percent of all strokes (1) It occurs when a blood clot (thrombus) forms and blocks blood flow in an artery bringing blood to part of the brain (14, 15) Blood clots usually form in arteries damaged by fatty buildups, called atherosclerosis (3).

When the blood clot forms within an artery of the brain, it's called a thrombotic stroke. These often occur at night or first thing in the morning (15) another distinguishing feature is that very often they're preceded by a transient ischemic attack. This is also called a TIA or "warning stroke." TIAs have the same symptoms of stroke but only last a few minutes; stroke symptoms last much longer. If someones experiences a TIA, they should urgent medical care immediately (1, 2).

A wandering clot (an embolus) or some other particle that forms away from the brain, usually in the heart, may also cause an ischemic stroke (4) This is called cerebral embolism. The clot is carried by the bloodstream until it lodges in an artery leading to or in the brain, blocking the flow of blood (5, 7)

The most common cause of these emboli is blood clots that form during atrial fibrillation (AF) (6) AF is a disorder found in about 2.2 million Americans. It's responsible for 15–20 percent of all strokes. In AF, the heart's two small upper chambers (the atria) quiver like a bowl of jello instead of beating strongly and effectively. Some blood isn't pumped completely out of them when the heart beats, so it pools and clots can form (8) When a blood clot enters the circulation and lodges in a narrowed artery of the brain, a stroke occurs. This is called a cardioembolic stroke, or a stroke that occurs because of a heart problem (10).

Hemorrhagic stroke 

A subarachnoid hemorrhage occurs when a blood vessel on the brain's surface ruptures and bleeds into the space between the brain and the skull (but not into the brain itself (9).

A cerebral hemorrhage occurs when a defective artery in the brain bursts, flooding the surrounding tissue with blood. 

Hemorrhage (or bleeding) from an artery in the brain can be caused by a head injury or a burst aneurysm. Aneurysms are blood-filled pouches that balloon out from weak spots in the artery wall (12) They're often caused or made worse by high blood pressure. Aneurysms aren't always dangerous, but if one bursts in the brain, they cause a hemorrhagic stroke (5, 9).

When a cerebral or subarachnoid hemorrhage occurs, the loss of a constant blood supply means some brain cells no longer can work. Accumulated blood from the burst artery also may put pressure on surrounding brain tissue and interfere with how the brain works (8). Severe or mild symptoms can result, depending on the amount of pressure.

The amount of bleeding determines the severity of cerebral hemorrhages. In many cases, people with cerebral hemorrhages die of increased pressure on their brains. But those who live tend to recover much more than people who've had strokes caused by a clot (13).  That's because when a blood vessel is blocked, part of the brain dies — and the brain doesn't regenerate itself; in other words, brain cells can't be replaced. But when a blood vessel in the brain bursts, pressure from the blood compresses part of the brain. If the person survives, gradually the pressure goes away. Then the brain may regain some of its former function (15).

Risk Factors for Stroke 

Evaluating the risk for stroke is based on heredity, natural processes, and lifestyle. Many risk factors for stroke can be changed or managed, while others that relate to hereditary or natural processes cannot be changed (1).

• high blood pressure                                                                                   

The most important controllable risk factor for brain attack is controlling                                  high blood pressure.

• diabetes mellitus

Diabetes is manageable, but having it increases the risk for stroke.     However, persons with diabetes are at higher risk of having a stroke as a result of the long-term effects of diabetes (1).

• heart disease

Heart disease is the second most important risk factor for stroke, and the major cause of death among survivors of stroke.

• cigarette smoking

The use of oral contraceptives, especially when combined with cigarette smoking, greatly increases stroke risk (3).

• history of transient ischemic attacks (TIAs)

A person who has had one (or more) TIA is almost 10 times more likely to have a stroke than someone of the same age and sex who has not had a TIA (2).

• high red blood cell count

A moderate increase in the number of red blood cells thickens the blood and makes clots more likely, thus increasing the risk for stroke.

• high blood cholesterol and lipids

High blood cholesterol and lipids increase the risk for stroke.

• lack of exercise, physical inactivity

Lack of exercise and physical inactivity increases the risk for stroke.

• obesity

Excess weight increases the risk for stroke (6).

• excessive alcohol use

More than two drinks per day raises blood pressure, and binge drinking can lead to stroke (1).

• drug abuse (certain kinds)

Drug abuse carries a high risk of stroke from cerebral embolisms (blood clots). Cocaine use has been closely related to strokes, heart attacks, and a variety of other cardiovascular complications. Some of them, even among first-time cocaine users, have been fatal (2).

• abnormal heart rhythm

Various cardiac diseases have been shown to increase the risk of stroke. Atrial fibrillation is the most powerful cardiac precursor of stroke (6).

• cardiac structural abnormalities

New evidence shows that cardiac structure abnormalities including patent foramen ovale and atrial septal defect increase risk for embolic stroke (3).

Physiological factors for stroke (1):

• age

For each decade of life after age 55, the chance of having a stroke more than doubles.

• race

African-Americans have a much higher risk of death and disability from a stroke than Caucasians, in part because the African-American population has a greater incidence of high blood pressure and diabetes.

• gender

Stroke occurs more frequently in men, but more women than men die from stroke.

• history of prior stroke

The risk of stroke for someone who has already had one is many times that of a person who has not had a stroke.

• heredity/genetics The chance of stroke is greater in people who have a family history of stroke.

Indian Epidemiology 

After coronary heart disease (CHD) and cancer of all types, stroke is the third commonest cause of death worldwide. However unlike the Caucasians, Asians have a lower rate of CHD and a higher prevalence of stroke. Among the Asians, the number who died from stroke was more than three times that for CHD. In one report, the agestandardized, gender-specific stroke mortality rate was 44 to 102.6/100,000 for Asian males, compared with only 19.3 for Australian white males. In the early 1980s the prevalence rates of stroke were around 500-700 per 100,000 in the western countries6 and 900 per 100,000 in Asia. The disparity between the stroke and CHD incidence rates is usually attributed to high prevalence of hypertension and low levels of blood lipids among the Orientals. Hypertension was related to high salt intake and perhaps to genetic factors and low serum lipid was due to low levels of animal fats and protein in oriental diet.

In India, several epidemiological studies have been undertaken in different parts of the country since the eighties. Most of this population based surveys however, were cross-sectional and determined the prevalence rates of stroke in the communities.  Population based surveys on stroke in various parts of India both in urban and in rural communities during the eighties and nineties. The prevalence rates determined from the major epidemiological surveys are listed in Table 1. It can be seen that other than the Parsis, Indians had much lower prevalence of stroke when compared to Caucasians and Chinese. The age adjusted prevalence rate of stroke was between 250-350/100,000 (16, 17).

There were limited data available on stroke related mortality in India. Although medical certification of the cause of death is a legal requirement, only 13.5% of all deaths in India were medically certified in 1994. Therefore ascertainment of the cause of death was grossly inadequate in India. However, it was estimated that stroke represented 1.2 % of the total deaths in the country, when all ages were included. The proportion of stroke death increased with age, and in the oldest group (> 70 years of age) stroke contributed to 2.4% of all deaths. The gender ratio of death due to stroke was One would expect a high mortality of stroke with low prevalence and median annual incidence of stroke in India (16, 17).

MOLECULAR PATHOPHYSIOLOGY OF STROKE   

Cerebral ischemia occurs when the amount of oxygen and other nutrients supplied by blood flow is insufficient to meet the metabolic demands of brain tissue. In ischemic stroke, the blood supply to the brain is disrupted by cerebrovascular disease. For decades, extensive research and clinical approaches to combat stroke have focused on the vascular aspects of cerebral ischemia. Therapeutic advances, including carotid endarterectomy, thrombolytic therapy, anticoagulation for cardiogenic stroke, antiplatelet agents, and the treatment of risk factors such as hypertension and hyperlipemia, have had significant effects on the morbidity and mortality of stroke.

The final event in cerebral ischemia is the death of neurons, resulting in irreversible loss of neurologic function. The advent of animal and tissue culture models of ischemia has led to many new insights into the mechanisms by which ischemic neurons die. If ischemia is complete and prolonged, neuronal death is inevitable. However, it has become increasingly clear that many secondary biochemical changes that exacerbate injury occur in response to the initial insult. In models of cerebral ischemia in rodents, as much as 50% or more of ischemic brain may be spared from infarction by preventing these secondary biochemical events. Understanding of the mechanisms by which neuronal cell death takes place has resulted in a number of therapeutic strategies that aim to prevent secondary biochemical changes and thus decrease the damage that results from cerebral ischemia. Ischemic neuronal death may involve the activation of enzymes and receptors that are constitutively expressed in brain. These existing receptors and enzymes do not require energy or the synthesis of new protein to exacerbate necrotic cell death. New evidence suggests that ischemic injury may also be exacerbated by the inducible proteins that mediate programmed cell death. These mechanisms are appealing targets for therapeutic intervention because they may occur hours or days after the initiation of ischemia (18).                                                                                            

NATURE OF NEURONAL DEATH IN CEREBRAL ISCHEMIA 

In ischemia, a mismatch between energy supply and demand may result in energy failure. Without adequate energy, protein synthesis cannot occur, and the genes that execute programmed cell death may not be expressed. The predominant histological feature of stroke is infarction. Infarction is synonymous with necrosis (i.e., cytoplasmic swelling, dissolution of organelles and plasma membranes, and inflammation are present). Under other circumstances, however, cerebral ischemia may produce neuronal death with many of the characteristics of programmed cell death. In models of transient ischemia, for example, energy failure is transient, and neuronal death develops more slowly than in permanent focal ischemia, with many features of apoptotic cell death. Under these circumstances, cleavage of genomic DNA into fragments of various sizes on DNA gels, characteristic of programmed cell death, occurs. However, the most convincing evidence that the production of new gene products may be important in the pathogenesis of neuronal death after transient ischemia is that protein synthesis inhibitors block delayed death of neurons. Thus, depending on the duration and severity of ischemia, stroke may produce cell death with features of necrosis or apoptosis (18).

MECHANISMS OF NECROTIC CELL DEATH 

The primary pathologic mechanism in stroke is the depletion of energy stores; however, considerable evidence indicates that excitatory amino acids (EAAs) exacerbate ischemic injury. EAAs such as glutamate are released by approximately 40% of all synapses in the central nervous system. Under physiologic conditions, EAAs participate in many neurological functions, including memory, movement, sensation, cognition, and synaptic plasticity. However, EAAs can also have a pathologic effect. EAA-mediated toxicity was first demonstrated by Olney and co-workers by peripheral administration of an EAA agonist that selectively killed neurons in the arcuate nucleus of the hypothalamus. These neurons contain high concentrations of glutamate receptors. Choi demonstrated that micromolar extracellular glutamate and other EAAs produce rapid increases in intraneuronal cytosolic Ca2+ concentrations. This increase in intracellular calcium concentration is rapidly lethal to primary neuronal cultures. 

The importance of calcium entry and excitotoxicity is supported by data demonstrating a direct correlation between extracellular calcium stimulation and neuronal death following exposure to glutamate. The increase in intraneuronal Ca2+ in response to extracellular EAAs in vitro is mediated by the opening of a receptor gated ion channel, the N methyl-D-aspartate (NMDA) channel. The NMDA channel, named after its highest affinity ligand, primarily gates calcium entry into the neuron. Treatment with antagonists that compete with glutamate and other EAAs for the receptor (competitive NMDA antagonists) or antagonists that bind to the ion channel itself (noncompetitive antagonists) can block calcium entry into neurons and prevent cell death induced by glutamate. Glycine is a co-agonist that is required in addition to glutamate to open the NMDA Ca2+ channel. Antagonists that bind to the glycine site on the NMDA receptor also block excitotoxicity in vitro. In addition to rescuing cells from EAA toxicity by blockade of the EAA receptors, it is possible to rescue neurons in culture by removal of extracellular calcium and sodium from the culture media following glutamate exposure. Conversely, inhibition of the sodium–calcium exchanger that normally facilitates extrusion of calcium results in an increase in neuronal death. Similarly, dantrolene, which attenuates decompartmentalization of intracellular stores of calcium, can reduce glutamate neurotoxicity in cortical neurons. Finally, neurons containing high concentrations of calcium binding proteins, such as calbindin or parvalbumin, are relatively resistant to excitotoxic injury. These data provide compelling evidence that EAA-induced increases in intracellular Ca2+ are toxic to neurons in culture. Compelling evidence is also available to indicate that excitotoxicity mediated by the NMDA receptor contributes to injury from cerebral ischemia in vivo. A rapid and large increase in the concentration of extracellular amino acids can be monitored by microdialysis after cerebral ischemia. Although NMDA antagonists are not effective in global ischemia models in which temperature is carefully controlled, a large number of studies have found that they decrease infarction volume in both permanent and temporary middle cerebral artery occlusion models in rodents. Blocking the translation of a gene that encodes a subunit of the NMDA receptor with intraventricular injection of antisense oligonucleotides also decreases infarction volume after middle cerebral artery occlusion in the rat. 

These data and many other studies support the hypothesis that excitotoxicity contributes to ischemic injury in vivo. Several calcium-dependent or calcium-induced enzymes mediate the toxic effects of increased intracellular calcium.. These include nitric oxide synthase, cyclooxygenase, phospholipase A2, and calpain 1. Calpain 1 is a calcium-activated protease that has been specifically linked to glutamate receptors in the rat hippocampus. Calpain 1 participates in the conversion of xanthine dehydrogenase to xanthine oxidase, which metabolizes xanthine to its reactive oxygen species, superoxide similarly, phospholipase A2 is activated by calcium and facilitates the release of arachidonic acid from injured cell membranes. Arachidonic acid is then metabolized by the enzyme cyclooxygenase into a prostaglandin, PGH2. The cyclooxygenase enzyme may produce a superoxide ion as a by-product of arachidonic acid metabolism. In addition, intracellular calcium can activate calcium-dependent isoforms of nitric oxide synthase to produce nitric oxide. The nitric oxide then combines with the superoxide produced as the byproduct of cyclooxygenase, xanthine oxidase, or other sources to form the highly reactive species peroxynitrite, which exacerbates tissue damage. Therefore, EAA-mediated elevation of intracellular calcium concentrations activates both cyclooxygenase and nitric oxide synthase, which then synergistically contribute to ischemic brain injury. Extracellular EAAs may activate other receptors besides the NMDA channel. EAA receptors can be categorized as ionotropic or metabotropic receptors. The metabotropic receptors may also increase intracellular calcium by mobilizing calcium from stores in the endoplasmic reticulum. Studies with antagonists of the metabotropic receptor show that, depending on their subunit specificity, some, but not all, drugs of this class are neuroprotective in models of focal ischemia. In addition to the direct downstream effects of enzymes that are activated by elevation of intracellular calcium, a number of complex interactions and positive feedback loops augment the contribution of EAAs to ischemic brain injury. For example, free arachidonic acid can potentiate NMDAevoked currents in neurons and inhibit reuptake of glutamate by astrocytes. In addition, platelet-activating factor, a phospholipase A2 metabolite, can stimulate the release of glutamate. Acidotic conditions favor the release of free iron, which can then participate in the metabolism of peroxide into the hydroxyl radical (Fenton reaction). In addition, glutamate can interfere with the function of the cystine transporter. Inhibition of the cystine transporter results in decreased intracellular concentrations of glutathione and diminished intracellular endogenous (18).

Figure: 4  Schemating diagram illustrating the mechanism by which ischemia and   excitotoxicity injure neurons

antioxidant stores. In vivo, excitotoxicity may be ameliorated by additional strategies besides inhibition of the NMDA receptor.Glutamate release into synaptic cleft, where it interacts with EAA receptors, is primarily mediated by the release of glutamate from the synaptic pool. Thus, a large component of excessive neuronal excitation may be the result of synaptic release of EAAs. Neuronal depolarization of presynaptic neurons in turn depends on activation of non- NMDA receptor-gated channels and other depolarizing neurotransmitter receptors. The excitatory action of depolarizing neurotransmitter receptors is countered by hyperpolarizing receptor-gated ion channels, such as the GABA (aminobutyric acid) receptor. Propagation of the action potential induced by depolarization of the neuronal cell body requires voltage-dependent sodium channels. Finally, the release of glutamate itself depends on P- and Q-type voltagedependent calcium channels. Glutamate release into the synaptic cleft can bind to the NMDA receptor and open the calcium channel. As a result, calcium enters the cell driven by its concentration gradient. However, intraneuronal calcium may increase by other mechanisms. Postsynaptic voltage-dependent calcium channels may allow calcium entry into the neuron when cells are depolarized, and glutamate released into the extracellular cleft may activate non-NMDA receptor-gated channels and depolarize the neuron. Also, Na+ may enter the cell

Figure: 5 A simplified neuronal circuit diagram illustrating the ion channels that determine the synaptic release of glutamate and intraneuronal Ca2+ concentrations in response to ischemia. 1, N-methyl- D-aspartate (NMDA) receptor-gated ion channel; 2, aminobutyric acid (GABA) receptor-gated Clchannel; 3, voltage-dependent Na channel; 4, voltage dependent Ca2+ channel; 5, non-NMDA receptorgated ion channel.

via the NMDA receptor- gated channel and depolarize the neuron. Thus, excitotoxicity may be ameliorated at a number of sites in vivo. Many drugs that can inhibit excitotoxicity at each of these steps have been developed. GABA agonists such as clomethazole have been shown to be neuroprotective in vivo and are currently undergoing clinical trials. In rodent models of stroke, BW1003, 619 and phosphenytoin prevent prolonged opening of the voltage-dependent sodium channel, ameliorate increases in extracellular glutamate, and decrease infarction volume. Drugs that prevent prolonged opening of P- and Q-type calcium channel antagonists are also neuroprotective in animal models of stroke. In contrast to their very limited effects in primary neuronal tissue culture models, non-NMDA antagonists are very effective in both global and focal ischemia models in rodents. Indeed, such agents have a longer time window of efficacy than do NMDA antagonists when administered after the onset of ischemia. 

Likewise, voltage-dependent calcium channel antagonists are not effective in vitro; however, the voltage-dependent calcium channel antagonist nimodipine is effective in reducing infarction volume in temporary focal ischemia in rats. Blockade of excitotoxicity via all these pharmacologic strategies has proved effective in temporary focal ischemia models in rodents, the model that most closely resembles human stroke. Unfortunately, results with these agents in human trials have to date been very disappointing, for several possible reasons. First, drugs that affect neurotransmission in the brain have many undesirable side effects, which in turn have led to reductions to doses that may have been ineffective. Side effects include effects on respiration and cardiac rhythm. In addition, agents that directly antagonize the NMDA receptor may injure a circumspect population of neurons in the cingulate and retrosplenial cortex in rodents, and may induce hallucinations and psychosis in humans. 

Another obvious reason for the lack of efficacy in these drugs in clinical trials is the time interval between the onset of ischemia and the administration of drug. When given before the onset of ischemia, these treatments can spare 50% or more of ischemic rat brain tissue from eventual infarction. When given after the onset of ischemia, they are  progressively less effective; however, such agents are effective up to 2 hours after the onset of middle cerebral artery occlusion in the rat. In the clinical trials, most patients were enrolled 6 to 12 hours after the onset of ischemia, long after the time that these drugs were effectively administered in animal studies. Whatever the reason for the failure of these anti-excitotoxic drugs in human trials, it has become clear that it may be more practical to select treatment approaches that target mechanisms that are active at longer intervals after ischemia. Accordingly, efforts to understand the delayed mechanisms of neuronal injury have been increased, in particular the role of programmed cell death in ischemic neurons.

Many of the key molecular events in programmed cell death have now been determined. Just as calcium entry into the neuron is a key step in excitotoxicity, the release of cytochrome c from the mitochondria is a key event in initiating apoptosis in many cell types. Cytosolic cytochrome c complexes with APAF-1 and procaspase 9. As a result, procaspase 9 is cleaved into its active form, caspase 9. Caspase 9 then cleaves and activates other caspases, including caspase 3. Caspases are a family of proteases that play a key role in executing programmed cell death. They were first identified by their homology with CED3, the key gene that irreversibly commits neurons in C. elegans to programmed cell death. A dozen mammalian caspases have been identified that have a variety of roles in executing programmed cell death and other cellular functions. Among the caspases, caspase 3 has the closest homology with CED3 and appears to play a key role as the final committed step in programmed cell death. Caspase 3 executes programmed cell death via cleavage of many other proteins.

These proteolytic targets of caspase 3 include cytoskeletal protein(s), DNA repair proteins such as PARP, and other proteins. Caspase 3 also cleaves ICAD, an inhibitor of CAD, an endonuclease that cleaves DNA between histosomes. The result is cleavage of DNA between histosomes, a hallmark of programmed cell death. The egress of cytochrome c from the mitochondria into the cytosol is controlled by several mechanisms. Genes of the bcl-2 family play an important role in controlling cytochrome c egress. Anti-apoptotic bcl-2 family members, such as bcl-2 itself and bcl-x-long, inhibit the egress of cytochrome c. Pro-apoptotic members of the bcl-2 family, such as bcl-x-short and bax, can form dimers with themselves or with anti-apoptotic bcl-2 family members. The balance between the pro-apoptotic and anti-apoptotic bcl-2 family proteins in the mitochondrial membrane determines whether permeability of the membrane will increase to allow egress of cytochrome c into the cytoplasm.Under some circumstances, cytochrome c exits the mitochondria via the mitochondrial permeability transition pore. This pore can open in response to prolonged depolarization, produced by such stimuli as an increase in intracellular calcium. Furthermore, pro-apoptotic bcl-2 family members such as bax may also interact with this pore. However, bcl-2 family members themselves may form pores in membranes, and some evidence indicates that bax induces egress of cytochrome c from the mitochondria independently of the mitochondrial permeability transition pore. Initiation of the mitochondrial apoptosis is also controlled by expression and translocation of other numerous bcl-2 family members. For example, translocation of bax from the cytosol to the mitochondria initiates programmed cell death. Bad is phosphorylated before being translocated to the mitochondria. More than 20 additional proteins are found in the bcl-2 family, including many that are also involved in mitochondrial homeostasis. 

Thus, a key event in apoptosis, egress of cytochrome c from the mitochondria, is controlled by bcl-2 family proteins. The molecular mechanisms by which programmed cell death is initiated are numerous and complex. Programmed cell death may be activated via cell surface receptors, including the Fas receptor and tumor necrosis factor (TNF). Activation of these receptors triggers activation of caspase 8, which in turn cleaves the bcl-2 family protein bid. The cleaved bid then translocates from the cytoplasm to the mitochondria, where it initiates cytochrome c egress. Other mechanisms by which the initiation of programmed cell death is controlled include the ERK (externally regulated kinase) and JNK protein kinase cascades. Finally, DNA base oxidation and other DNA damage may initiate programmed cell death via expression of the p53 transcription factor. These and other mechanisms may be involved in the initiation of programmed cell death in ischemic neurons (18).

PROGRAMMED CELL DEATH AFTER CEREBRAL ISCHEMIA 

Evidence indicates that many of the mechanisms that initiate programmed cell death are activated in ischemic neurons under certain conditions. TNF mRNA transcription is induced as an early response after cerebral ischemia. Expression of the TNF receptor is also increased after cerebral ischemia TNF-binding protein, a protein that binds and inhibits TNF, reduced infarction volume after middle cerebral artery occlusion in rats. However, ischemic injury was exacerbated in TNF receptor null mice, which suggests that TNF signaling pathways may instead have beneficial effects in ischemic injury under some circumstances. Caspase 8, which is activated by both the Fas and TNF receptors, is expressed and activated after cerebral ischemia. Changes in expression and activity of both the ERK and JNK kinase pathways occur following cerebral ischemia. The M-terminal kinases of c-Jun are activated after ischemia and phosphorylate c-Jun. The increased expression of ERK after focal ischemia and inhibitor 

Figure: 6  A schematic diagram illustrating the molecular mechanisms that control programmed cell death.

of NEK-1, another kinase in the ERK pathway, protect the brain against focal cerebral ischemia. Single-stranded DNA damage is an early event in cerebral ischemia reperfusion injury and may trigger expression of p53. Expression of p53 is induced after cerebral ischemia. A number of studies in cerebral ischemia support a role for bcl-2 family genes in controlling ischemic neuronal death. In rodent models of ischemia, anti-apoptotic members of the bcl-2 family, including bcl-2 and bcl-x long, are expressed in neurons that are ischemic yet survive. Abundant in vivo evidence also suggests that caspase activity exacerbates ischemic injury. Transgenic mice that express a dominant

Negative mutation of caspase 1 had smaller infarctions than did their wild-type litter mates. Furthermore, intraventricular infusion of peptide inhibitors of caspases decreased infarction volume in rats subjected to temporary middle cerebral artery occlusion. These peptide inhibitors of caspases also blocked damage in response to injection of excitotoxins. Caspase 3 mRNA and protein expression is induced in CA1 neurons after global ischemia. Caspase 3 is activated, and treatment with a specific peptide inhibitor of caspase 3 ameliorated neuronal death in the global ischemia model. These and other studies support a role for caspases in ischemic neuronal injury (18).

ROLE OF PLA2 IN STROKE 

One such critical metabolic event is the activation of phospholipase A2 (PLA2), resulting in hydrolysis of membrane phospholipids and release of free fatty acids including arachidonic acid, a metabolic precursor for important cell-signaling eicosanoids. PLA2 enzymes have been classified as calcium-dependent cytosolic (cPLA2) and secretory (sPLA2) and calcium-independent (iPLA2) forms. Cardiolipin hydrolysis by mitochondrial sPLA2 disrupts the mitochondrial respiratory chain and increases production of reactive oxygen species (ROS) which degrade to reactive aldehyde products (malondialdehyde, 4-hydroxynonenal, and acrolein) that covalently bind to proteins/nucleic acids, altering their function and causing cellular damage. Activation of PLA2 in cerebral ischemia has been shown while other studies have separately demonstrated increased lipid peroxidation. Dissecting the contribution of PLA2 to lipid peroxidation in cerebral ischemia is challenging due to multiple forms of PLA2, cardiolipin hydrolysis, diverse sources of ROS arising from arachidonic acid metabolism, catecholamine autoxidation, xanthine oxidase activity, mitochondrial dysfunction, activated neutrophils coupled with NADPH oxidase activity, and lack of specific inhibitors. Although increased activity and expression of various PLA2 isoforms have been demonstrated in stroke, more studies are needed to clarify the cellular origin and localization of these isoforms in the brain, their responses in cerebral ischemic injury, and their role in oxidative stress (45).

Non-steroidal anti-inflammatory drugs (NSAIDs) are among the most used drugs worldwide. Despite the similarities between COX-1 and COX-2 in terms of catalytic mechanisms and kinetics, there are two major structural differences that have important consequences from the pharmacological and biological viewpoint. First, COX-2 has a larger and more accommodating cyclooxygenase active site as compared with COX-1 (46). This structural difference in the AA pocket size and shape has allowed the development of highly selective COX-2 inhibitors. Second, COX 1, as opposed to COX-2, exhibits negative allosterism at low AA concentrations. Thus, when both isoforms are expressed in the same cell, COX-2 competes more effectively for newly released AA (47).

COX-1 is widely expressed in neurons and microglia in several animal species (48), and its immunoreactivity is enriched in midbrain, pons and medulla (46). COX-1 is intensely expressed in microglia in both rat (47) and human (46), and relative high levels are also found in hippocampal and cortical neurons in control human brain (50). The brain is one of the few tissues that constitutively express COX-2. The immunoreactivity for COX-2 is primarily localized to neurons, and in the rat brain it prevails in the hippocampus with the densest staining in the dentate gyrus, piriform cortex, amygdala, and in layers II and III of the neocortex. COX-2 expression colocalizes selectively with glutamatergic neurons, and seems to be coupled to excitatory neuronal activity. It has been demonstrated that activation of the glutamatergic NMDA receptors results in increased neuronal COX-2 expression. COX-2 plays an important role in mediating the changes in neocortical blood flow evoked by neural activity. There is a modulatory role of arachidonic acid metabolites produced by COX-2 in the control of focal cerebral blood flow following local increases in neuronal activity, suggesting the involvement of COX 2 in cerebrovascular coupling (49). This is a very important homeostatic mechanism matching the delivery of nutrients with the energy needs of the active brain.

Dramatic increases in COX-2 mRNA and protein levels occur following in vivo excitotoxic injury. Increased upregulation of COX-2 mRNA and protein has also been found in neuronal cultures exposed to excitotoxic insults in vitro. On the contrary, COX-1 expression levels are not changed by excitotoxic damage. The increased COX-2 expression following kainate excitotoxicity can be prevented by blockade of glutamate receptors (48), and by an antagonist of platelet-activating factor (PAF) receptor (47). Substantial COX-2 overexpression was found in vulnerable neurons following kainic acid injection.

It has long been known that COX-2 expression is significantly and persistently increased following cerebral ischemia in different models. In this section, we will describe the most important observations of the effect of brain ischemia on COX-2 expression in both global and focal cerebral ischemia. Expression of COX-2 following global cerebral ischemia Global cerebral ischemia results from the transient cessation of blood flow to the brain, leading to a characteristic pattern of cell death in specific neuronal populations. The hippocampal CA1 region is considered the most susceptible to a global ischemic event. Clinically, global brain ischemia occurs after cardiopulmonary bypass surgery or cardiac arrest with resuscitation, and is associated with problems with cognition and memory, sensorimotor deficits, seizures, and death (47).

Ionotropic Glutamate Receptors 

Glutamate is the principle mediator of ischemic neuronal damage and, as such, has been the focus of much attention in efforts to understand the pathophysiology of cerebral ischemia. Glutamate acts on both ionotropic and metabotropic receptors. The ionotropic receptors are ligand-gated ion channels and the metabotropic receptors are linked via G-proteins to the cAMP and IP3 second messenger systems. The ionotropic receptors are further subdivided into 3 types based on their pharmacology: NMDA, a-amino-3-hydroxy-5- methyl-4-isoxazole propionate (AMPA) and kainite (KA) (67). Of these types there exist several more subtypes based on structural/functional classifications. Glutamate receptors are activated during ischemia when glutamate levels rise several- fold. Despite the fact that glutamate levels return to normal shortly after reperfusion is initiated, glutamate receptor activation is an important mediator of ischemic neuronal death. NMDA receptors may be the major source of the lethal postischemic Ca2+ influx after ischemia. NMDA receptors are composed of NR1 and NR2A-D subunits. Their functional characteristics are determined by the receptors specific combination of NR1 and NR2 subunits (Table 1). There are also modulatory sites on the NMDA receptor for glycine, Mg21 and polyamines (66). There is a plethora of NMDA receptor antagonists that act at these sites. 

The role of glutamate receptors in cerebral ischemia is further supported by numerous studies that describe ischemia-induced changes in glutamate receptor expression. In the hippocampus there is a selective decrease in NMDA-R1 mRNA in the CA1 region91 and a decrease in mRNA and protein expression of both NMDA NR2A and NR2B subunits in CA1 and dentate gyrus regions early after ischemia but a recovery of NR2A and NR2B mRNA in dentate at later times. Although still controversial, there may be a postischemic increase in the expression of Ca2+-permeable AMPA receptors resulting from a decrease in the expression of the GluR2 subunit. This could lead to the lethal postischemic Ca2+ influx that occurs in vulnerable neurons. This observation is consistent with the neuroprotection obtained by Buchan et al98 administering NBQX as late as 12 hours after ischemia. Ischemia induced changes in glutamate receptor subunits have led to a search for subunit selective antagonists that would show greater specificity than currently available drugs (66, 67).

GABA Receptors 

As with glutamate, ischemia results in a 2-fold increase in extracellular GABA. The increase is not as large as that of glutamate and does not persist as long. GABA has the opposite effect of glutamate on neuronal excitation. GABA receptors are both ionotropic, GABAA, and metabotropic, GABAB the ionotropic GABA receptors are composed of 17 different a, b, d, and g subunits. Agonists for both types of receptors are neuroprotective against ischemia. The majority of GABAergic neurons in the hippocampus are interneurons, (66). which are spared after ischemia and seem to provide protection to neighboring cells. Any protection of these GABAergic neurons would be anticipated to aid the survival of the more vulnerable pyramidal neurons. GABA uptake inhibitors are neuroprotective, GABAB receptor activation appears to block the release of excitatory neurotransmitters. Because of the number of different possible molecular subunits that can combine to form a heteromeric GABA channel complex, recent work has only scratched the surface in terms of looking at altered subunit expression after ischemia. There is an early decrease in the expression of the predominant adult subunits a1 and b2 in all regions of the hippocampus, but expression recovers in the resistant CA3 and dentate gyrus regions. At the same time, there is a decrease in the functionality of both GABAA and GABAB receptors. It remains to be seen whether these changes in message are translated into changes in protein and function and whether or not they can be exploited to improve neurological outcome after ischemia (67).

Calpains are calcium-dependent proteases thought to play an important role in cytoskeletal reorganization and muscle protein degradation. Calpains exist as heterodimers comprised of a small regulatory subunit and one of the 3 large catalytic subunits: calpain-1, -2, or -3. Calpains and caspases often synergize in the apoptotic process, especially in neuronal calpain cleave Bid independently of caspase activation and thus trigger a consecutive apoptosis. Grp94 was shown to protect human neuroblastoma cells from hypoxia/reoxygenation-induced apoptosis involving calpains . Grp94 was also shown to be cleaved by calpain in etoposide-induced apoptosis. It is interesting to note that cisplatin, which may interact with Grp94, induces the activation of calpain in the apoptotic process (68, 69).

HSP70 AS POTENTIAL TARGET IN BRAIN STROKE 

Heat shock proteins and phosphatidyl-inositol-3-kinase/Akt pathway The major antiapoptotic pathway, the phosphatidylinositol- 3-kinase (PI-3-kinase)/Akt pathway, mainly responds to growth factor withdrawal. The Akt kinase is

Activated via the PI-3-kinase, and the activation stalls

Figure:9  Interactions of Hsp with the signaling network regulating cell survival and apoptosis. On the figure, a comprehensive summary is given about the network of Hsp interactions with various apoptosis-related signaling pathways, such as the activation of stress kinases, the Akt survival pathway, as well as various interactions with Hsp synthesis itself.

Apoptosis. Once activated, Akt phosphorylates the proapoptotic factors, like Bcl-2-associated death protein (BAD), enabling them to bind to the protein, which sequesters and inhibits them Akt also phosphorylates eNOS at Ser-1179, enhancing the production of NO as another mechanism to inhibit apoptosis. Endothelial cell survival pathways are known to involve Akt, and it remains to be seen whether NO plays a role in this process. Hsp27 associates with Akt and protects its kinase activity from heat stress and serum deprivation in PC12 embryonal carcinoma cells. In neutrophils, Hsp27 association is necessary for Akt activation. Act induced phosphorylation of Hsp27 results in its dissociation from Akt and enhanced neutrophil apoptosis. Hsp90 is a necessary chaperone for the Akt kinase by interacting both the Akt-activator 3-phosphoinositide-dependent protein kinase-1 (PDK1) and Akt itself. 

In connection with this, Hsp90 serves as a molecular scaffold to promote the Akt-induced phosphorylation and activation of eNOSHeat shock proteins in tumor cells Hsp, such as Hsp70, Hsp27, and Hsp90, which can inhibit apoptosis by direct physical interaction with apoptotic molecules, are also overexpressed in several tumor Moreover, due to the Hsp-induced simultaneous stabilization of various proteins, Hsp inhibitors (as opposed to, e.g., protein kinase inhibitors) target not only a specific molecule, but a number of molecules, which makes them potentially more effective in the induction of tumor cell apoptosis. The most important Hsp90 inhibitors are geldanamycin , its less toxic analogue, 17-allylamino-17-demethoxy- geldanamycin (17-AAG; , radicicol, and its more stable oxime derivatives ,, which have a higher affinity for Hsp90 than geldanamycin . Recently, new geldanamycin analogues and a third class of inhibitors, the purine scaffold inhibitors, were developed, and there are ongoing efforts to synthesize even more Hsp90-interacting drug candidates.

A recent report showed an important element of tumor specificity of Hsp90 inhibitors . When Hsp90 becomes activated, it forms a large complex with various co-chaperones in tumor cells; on the contrary, it is found in a latent, uncomplexed state in normal cells. The geldanamycin-derivative, 17-AAG, binds to the tumorspecific, complexed form of Hsp90, with a 100-fold higher affinity than the latent form in nontransformed cells This difference also raises the possibility that active Hsp90 behaves as a tumor selective catalyst in converting geldanamycin derivatives to their active conformation Hsp90 inhibition leads to the dissociation of various Hsp90 client proteins from the chaperone complex and to their consecutive degradation by the proteasome. Inhibition of Hsp90 induces apoptosis of various tumor cells. Hsp90 inhibition also leads to a defect in a number of proliferative signals including the Akt-dependent survival pathway. Moreover, inhibition of Hsp90 was shown to be successful in reducing chemoresistant tumors with poor prognosis  The C-terminal nucleotide binding pocket has a unique nucleotide  

Figure:10  Major pathways of mitochondrial apoptosis. Many proapoptotic and signal transduction pathways converge on mitochondria to induce mitochondrial membrane permeabilization. The PTP induces the mitochondrial translocation and multimerization of the proapoptotic protein, Bax. Bax, in turn, helps the permeabilization of the inner mitochondrial membrane resulting in the leakage of cytochrome c and other mitochondrial inter-membrane proteins that give a ‘‘go’’ signal for the execution phase of apoptosis. The second mitochondria-derived activator of caspases (Smac/DIABLO) is a mitochondrial protein that inhibits the IAP (such as XIAP, c-IAP1, and c-IAP2) after its release to the cytosol. IAP are known to block the processing of the effector caspases, caspase-3 and -9. Apart from Smac/DIABLO, there is another mitochondrial protein, HtrA2/Omi, that also inhibits IAP. The release of cytochrome c from mitochondria drives the assembly of the high molecular weight caspase-activating complex called apoptosome. The apoptosome contains oligomerized Apaf-1, which in presence of dATP and caspase-9, recruits and helps the autoactivating cleavage of caspase-3, an executioner of apoptosis. On the figure, several elements of ROS-induced apoptosis are also included, such as the formation of the highly toxic nitrogen reactive species (NRS). Finally, in the executioner phase of apoptosis, endonucleases, such as CAD, are activated.

binding specificity Major pathways of mitochondrial apoptosis. Many proapoptotic and signal transduction pathways converge on mitochondria to induce mitochondrial membrane permeabilization.

The PTP induces the mitochondrial translocation and multimerization of the proapoptotic protein, Bax. Bax, in turn, helps the permeabilization of the inner mitochondrial membrane resulting in the leakage of cytochrome c and other mitochondrial inter-membrane proteins that give a ‘‘go’’ signal for the execution phase of apoptosis. The second mitochondria-derived activator of caspases (Smac/DIABLO) is a mitochondrial protein that inhibits the IAP (such as XIAP, c-IAP1, and c-IAP2) after its release to the cytosol. IAP are known to block the processing of the effector caspases, caspase-3 and -9. Apart from Smac/DIABLO, there is another mitochondrial protein, HtrA2/Omi, that also inhibits IAP. The release of cytochrome c from mitochondria drives the assembly of the high molecular weight caspase-activating complex called apoptosome. The apoptosome contains oligomerized Apaf-1, which in presence of dATP and caspase-9, recruits and helps the autoactivating cleavage of caspase-3, an executioner of apoptosis (72).

Previously, it has been supposed that PPAR activation could also be effective in the regulation of neuronal death in ischaemic, neurodegenerative or inflammatory cerebral diseases. Firstly, PPARs have been described in brain and in spinal cord Beyond expression in cerebral or spinal blood vessels, (74) PPARs are also expressed in neurons and in astrocytes, whereas oligodendrocytes exclusively show PPARβ/δ expression The extent of this expression depends on the isoform of PPAR involved. PPARβ/δ has been found in numerous brain areas, while PPARα and PPARγ have been localized to more restricted brains areas Secondly, whatever the aetiology, neuronal death is induced by inflammatory and oxidative processes with a link between the two phenomena Inflammation and oxidative stress induce both necrotic and apoptotic neuronal death (73). The transcription factor NF-κB plays a key role in regulation of inflammation and oxidative stress leading to neuronal death, explaining why PPARs have been considered as possible targets for neuroprotectionPPARs are also able to inhibit the entry of inflammatory cells into the CNS (central nervous system) from the periphery by inhibition of chemokines, adhesion molecules and metalloproteinases 

PPAR and cerebral ischaemia 

Because fibrates, used as lipid-lowering agents, contribute to secondary prevention of stroke, it has been supposed that these PPARα activators could also preventively protect the brain against noxious biological reactions induced by cerebral ischaemia, such as oxidative stress and inflammation. More recently, (73).it has been demonstrated that PPARα agonists could also induce an acute neuroprotection when administered just before cerebral ischaemia or during the reperfusion period Administration of the PPARγ agonists troglitazone or pioglitazone 24 or 72 h before and at the time of cerebral infarction dramatically reduced infarction volume and improved neurological function following transient middle cerebral artery occlusion in rats. This effect is exerted in a dose-dependent manner. This neuroprotection has been reproduced by an intracerebroventricular administration of pioglitazone, proving that it is the activation of intracerebral PPARγ that confers neuroprotection and neurological improvement following ischaemic injury Moreover, a non-thiazolidinedione PPARγ agonist (L-796449) also had a neuroprotective effect in experimental stroke (74).

Cerebral mechanisms 

The neuroprotection observed after treatment with PPAR agonists is related to several mechanisms including both oxidative stress modulation and anti-inflammatory effect. PPARα agonist-induced neuroprotective effect is associated with a decrease in cerebral oxidative stress depending on the increase in activity of numerous antioxidant enzymes, in particular Cu/Zn superoxide dismutase and glutathione peroxidase This modulation of antioxidant enzymes is responsible for a decrease in ischaemia-induced reactive oxygen species production and lipid peroxidation (74).This effect on oxidative stress could be related to a direct effect on antioxidant enzymes expression, because PPREs (PPAR-response elements) have been found in the gene of Cu/Zn superoxide dismutase The neuroprotective effects of PPAR agonists are also related to inhibition of ischaemia-induced inflammatory markers (interleukin-1β, COX-2 and inducible nitric oxide synthase) The different PPARisoforms do not modulate the inflammatory pathways involved in neuroprotection in a similar manner. For instance, ischaemia-induced COX-2 overexpression is prevented by PPARγ agonists but not by PPARα agonists There is a link between PPARinduced modulation of oxidative stress and inflammation, since prevention of COX-2 induction results from oxidative stress inhibition The cellular target of these anti-inflammatory effects is probably microglial cells, since PPARγ agonists, such pioglitazone, are able to decrease microglial activation when administered intracerebrally The key target of this anti-inflammatory effect is NF-κB, which plays a crucial role in neuronal death PPARγ and PPARα activation is responsible for inhibition of the NF-Κb p65 monomer as well as induction of IκBα (inhibitory κB) (73). The role of suppression of activation of p38 mitogenactivated protein kinase has also been demonstrated recently Beyond this direct effect on ischaemia-induced deleterious pathways explaining neuroprotection, the challenge will be to demonstrate that a part of the neurological improvement induced by PPAR activators could be the result of neurorepair, since PPARγ s are also involved in the regulation of neural stem cell proliferation and differentiation 

Because PPARs are mainly expressed in cerebral vascular wall, in particular in endothelium, it has been supposed that vascular mechanisms could be involved in neuroprotection (73). Thus preventive neuroprotection by PPARα is associated with an improvement in middle cerebral artery sensitivity to endothelium-dependent relaxation unrelated to an increase in endothelial nitric oxide synthase expression More recently, it has been demonstrated that preventive or acute PPARα agonist-induced neuroprotection paralleled the prevention of ischaemia-induced endothelial dysfunction This vascular effect could be related to: (i) the prevention of ischaemia-induced vascular expression of adhesion molecules; (ii) the antioxidant effect of PPAR activation; and (iii) the inhibition of ischaemia-induced metalloproteinase expression. In addition, PPAR could also be involved in endothelial regeneration as has been demonstrated in other arterial areas  (74).

Cell Surface Signalling 

Neuronal excitability is the result of an imbalance of ions across a cell’s membrane. During ischemia this imbalance is disrupted and the altered Ca2+ homeostasis mediates an excitotoxic cascade culminating over hours to days. The cascade of events leading to this death can be divided into 3 stages (21, 22). The initial stage is the ischemic period itself (5-30 min for global and 60-90 min for focal) during which ionic gradients collapse. Much is known about the nature of the metabolic failure, however, the only neuroprotective strategy available at this time is to restore perfusion with tPA. The second stage is the reperfusion period in which there is a recovery of the cell’s energy state, ion homeostasis, and basic physiological functions. The second stage lasts for hours to days and is the time for neuroprotective strategies designed to prevent the death of postischemic neurons. Finally, there is the third stage in which there is another energy failure accompanied, eventually, by cell death. Because it is too late to affect protection at this stage, trophic responses in which regenerative capacities can be enhanced are the best therapeutic approach (21 - 44).

Figure: 7 Schematic illustration of cell surface mediators of excitotoxicity cascade.

Potassium Channels 

K1 channels are a major contributor to a cell’s resting potential and their activation helps to maintain a hyperpolarized resting membrane potential. It should not be surprising, then, that K1 efflux occurs much sooner after the onset of ischemia than either the Na1 or Ca2+ influx, which does not occur until ATP levels have fallen by more than 50% (23). There are a number of different types of K1 channels. The metabolic nature of an ischemic insult suggests that ATPsensitive K1 channels, which are activated by a decrease in ATP, would be one of the first channels to respond during ischemia. In hypoxic rats, the early K1 efflux in the dorsal hippocampus could be blocked by pretreatment with 4-aminopyridine (4-AP) a blocker of voltage-activated K1 channels found predominantly in the dendritic portion of hippocampal pyramidal neurons (22). At later times after ischemia, other K1 channels are activated including the Ca2+-activated K1 channels (23), activated by an increase in intracellular Ca2+ and the ATP-sensitive K1 channels (24-27). Astrocytes attempt to buffer this increase in [K1]e. They switch to anaerobic glycolysis and swell 5 to 10 times their normal size. Eventually, astrocytes are no longer able to cope with the increase in [K1]e and they lyse (22). These events all occur during the first stage of ischemia. There is little known regarding the behavior or expression of K1 channels in the second stage but given their role in in vitro neuronal apoptosis (28), a closer examination of these channels is warranted. The development of therapeutic agents targeting K1 channels in the brain will not be easy. Specificity will be a problem because K1 channels are ubiquitously expressed throughout the body (30).

Sodium Channels 

Na1 channels also play an important role in neuronal excitability and they are as widely expressed as K1 channels. While Na1 channels have received little attention by those studying ischemia in gray matter, they have been extensively studied in ischemic white matter (31-33). Because of the differences in the mechanisms underlying excitation in white matter, it would be surprising if the mechanisms of ischemic death were the same in white matter and gray matter. In cerebral ischemia there is a pronounced Na1 influx at the end of the first stage of ischemia (Fig 7). This is coincident with the Ca2+ influx and the anoxic depolarization associated with energy failure (34,35) Although problems of specificity exist for therapeutic agents targeting Na1 channels, there is evidence using local anesthetics that suggests inhibition of Na1 channels is neuroprotective. Some anticonvulsants that inhibit Na1 channels are neuroprotective in vivo (37). Lamotrigine and its derivatives BW1003C87 and BW619C89 are protective in models of focal and global ischemia. Similarly, the Na1 channel blockers riluzole and its derivative RP66055 are also protective in both focal and global models of cerebral ischemia. The role of Na1 channels in cerebral ischemic death should not be over emphasized because the agents mentioned previously are not selective (40). Riluzole antagonizes N-methyl- D-aspartate (NMDA) receptors and lamotrigine and its derivatives inhibit Ca2+ channels. Furthermore, many of the Na1 channel antagonists exhibit severe cardiovascular effects that eliminate them as therapeutic agents for the treatment of stroke (38).

Ca2+ channels have received a lot of attention in studies of cerebral ischemia because Ca2+ influx and the disruption of Ca2+ homeostasis play an important role in ischemic cell .In addition to the Ca2+ influx through voltage-gated Ca2+ channels, a much larger portion of the influx occurs through ligand-gated ion channels (42). Of the voltage-gated Ca2+ channels there are primarily 5 types, L, T, N, R, and P/Q, which are  SMALL, MORLEY, AND BUCHAN defined by their subunit molecular biology and pharmacology. L-type channels are thought to activate gene responses because they are located mainly on the cell bodies of neurons and to a lesser degree on dendrites. Much of the insight on the role of the various Ca2+ channel subtypes in ischemic neuronal injury comes from studies using pharmacological agents (43). Dihydropyridines have been used with limited success in animal models of cerebral ischemia. Nimodipine, chosen for its high blood brain barrier permeability, is neuroprotective in some models of focal and global ischemia but isradipine66 and AT-22767 failed to reduce lesion volume in focal models of ischemia (40). 

Several clinical trials using Ca2+ channel antagonists have shown no neuroprotective efficacy. Because of the importance of glutamate in ischemia, the Ca2+ channels involved in glutamate release provide an attractive therapeutic target. The conotoxin, SNX-111, which specifically blocks N-type channels, is neuroprotective in both focal and global models of ischemia. Antagonists of Q-type (SNX-230), and P-type (daurisoline), Ca2+ channels failed to provide neuroprotection against ischemia although the efficacy of SNX-111 was impressive in animal models, this efficacy was not translated into success in clinical trials for the treatment of stroke (42). As with K1 and Na1 channel antagonists, agents targeting voltage-gated Ca2+ channels will be faced with the problem of specificity because of the ubiquitous expression of these channels (41).

There are several factors responsible for the cytotoxicity of COX-2 in the setting of cerebral ischemia, which result in neuronal injury. The increased production of free radicals and PGE2 are among the most recognized mechanisms of toxicity linked to increased COX-2 activity. However, there are also two other processes, modulated by COX-2, which could potentially lead to neuronal death: 1) promotion of cell cycle activity by increasing cyclin D1 expression, and 2) metabolism of endocannabinoids  Cerebral ischemia results in a substantial increase in the availability of arachidonic acid, the substrate for the COX enzymatic pathway. A wide body of experimental evidence supports the theory that COX catalytic activity in linked to the production of free radicals (46). 

Oxidative stress is considered to be one of the major determinants of ischemic neuronal death Detailed biochemical investigations have demonstrated that free radicals are indeed produced by the peroxidase step of the COX reaction in which PGG2 is converted to PGH2. Although it has become customary to consider reactive oxygen species (ROS), and specifically superoxide anion (O2•-), to be the primary radical produced by COX activity during inflammation, there is no direct evidence for this notion. On the contrary, the two major types of radicals so far known to be involved in COX activity are tyrosyl radicals on proteins and carbon-centered radicals on the substrate arachidonic acid (50). Due to the characteristically short half-life of free radicals, and the technical difficulties associated with their direct measurement in biological systems, there is still debate on the chemical nature of the free radical(s) involved in COX-2-mediated oxidative stress during inflammation and cerebral ischemia (49).

However, there is an overwhelming line of evidence indicating that enhanced COX activity following cerebral ischemia and excitotoxicity is associated with oxidative damage. Pharmacological inhibition of COX-2 with either nimesulide or rofecoxib resulted in a significant reduction in measures of oxidative injury in the hippocampus following global cerebral ischemia in gerbils. These COX-2 inhibitors prevented ischemia-induced glutathione depletion and the increase in lipid peroxidation, as assessed by the levels of lipid hydroperoxides, malondialdehyde (MDA) and 4-hydroxy-alkenals. In a rat model of global forebrain ischemia, nimesulide treatment reduced lipid peroxidation and prevented the depletion of reduced glutathione following reperfusion. Similarly, indomethacin treatment significantly reduced 8-hydroxy-deoxyguanosine (8-OH-dG), a highly sensitive marker of DNA oxidation, in the ischemic hippocampus.

In in vivo models of excitotoxicity, COX-2 has been also demonstrated to be a significant source of free radicals. Treatment with nimesulide significantly reduced oxidative injury in the rat hippocampus after systemic kainate injection. A microdialysis study in the hippocampus of freely moving rats showed that the COX inhibitors flurbiprofen and indomethacin, or the COX-2 selective inhibitor NS-398 effectively reduced 8-epi- PGF2α (15-F2t-IsoP), a reliable marker of free radical-mediated lipid peroxidation, following infusion of NMDA. Interestingly, using the same rat model of excitotoxicity, inhibition of COX-1 with SC-560 significantly attenuated the increase in hippocampal 8-epi-PGF2α levels induced by NMDA. In support of this study, it has been reported that the COX-1 inhibitor valeryl salicylate reduced measures of oxidative stress in the gerbil hippocampus following temporary global ischemia This suggests that COX-1 may also contribute to oxidative injury following excitotoxicity and brain ischemia (48).

iNOS and NOX 

Most brain pathologies are accompanied by inflammation, during which the production of NO (nitric oxide; mainly fromiNOS) and/ or superoxide (O+) plus H2O2 (mainlyfrom NOX) are increased. NO and superoxide, and derivative RNS (reactive nitrogen species) and ROS (reactive oxygen species) are, at low concentrations signaling molecules, for example regulating cell proliferation, but at high concentrations are key cytotoxic molecules of innate immune defence against pathogens (51-54). They are also implicated in the pathology of most brain diseases, .iNOS is not normally expressed in the brain, but

Figure 8  How inflammation is induced by pathogens or damage

Inflammatory mediators such as LPS (lipopolysaccharide) and cytokines cause its expression in microglia and astrocytes, and possibly in neuron (55, 56). Once expressed iNOS produces high levels of NO continuously. Phagocytic cells such as neutrophils, macrophages and microglia have a specific NOX known as PHOX (phagocytic NADPH oxidase), consisting of subunits gp91 (glycoprotein), p22, p47, p67, p40 and Rac. In the healthy, non-inflamed brain, PHOX is expressed at high levels inmicroglia (and possibly at low levels in astrocytes and neurons). However, PHOX is not active Hypoxia, iNOS expression or PHOX activation may be relatively benign, or even cytoprotective, when present alone, but hypoxia combined with NO results in cell death via respiratory inhibition, and PHOX activation combined with iNOS expression results in death via peroxynitrite production unless acutely stimulated by for example TNFα (tumour necrosis factor α), IL-1β (interleukin 1β), chemokines, arachidonate, ATP or phagocytosis, when it produces high levels of superoxide extracellularly, which dismutate to H2O2 (57).

NO-induced neuronal death 

Scientists have identified various modes and mechanism by which inflammatory-activated glia kill neurons in culture The first mode is mediated by high levels of iNOS expression in gliaI It has been shown that resulting high levels of NO induce neuronal death by causing inhibition of mitochondrial cytochrome oxidase in neurons (58-60). NO inhibition of neuronal respiration caused neuronal depolarization and glutamate release, followed by excitotoxicity via the NMDA (N-methyl-D-aspartate) receptor. This excitotoxicity may be potentiated by a second mechanism, as NO from iNOS results in glutamate release from astrocytes via calcium release from intracellular stores stimulating exocytosis of vesicular glutamate. Thus inflammatory-activated astrocytes maintained a higher extracellular glutamate level, which was probably insufficient to induce excitotoxicity alone, but may well be sufficient if, in addition, neuronal respiration is inhibited so that NMDA receptor is activated by both depolarization and glutamate. However, the above mechanism requires relatively high levels of NO or iNOS expression, and iNOS can be expressed in vitro or in vivo apparently with little or no neuronal death. Indeed NO from iNOS may be protective by blocking brain cell death. On the other hand, low levels of iNOS expression may synergize with other conditions to induce cell death (61-63). For example, hypoxia strongly synergizes with NO or iNOS expression to induce neuronal death via respiratory inhibition. This is because NO inhibits cytochrome oxidase in competition with oxygen, so that NO greatly increases the apparent Km of neuronal respiration for oxygen. This sensitization to hypoxia is potentially important in stroke, where both inflammation and hypoxia may coexist.

PHOX 

A second mode that identified was the dual-key (iNOS and PHOX) mechanism of inflammatory neurodegeneration that activation of iNOS or PHOX alone caused little or no neuronal death, but when both were activated together, this caused massive neuronal death mediated by peroxynitrite. inflammatory neurodegeneration induced by TNFα, IL-1β, prion peptide, LPS, IFNγ (interferon γ ), arachidonate, ATP and/or PMA was mediated by this mechanism under particular conditions (64). Simultaneous activation of PHOX and iNOS in microglia resulted in the disappearance of NO, appearance of peroxynitrite , and massive death of cocultured neurons that was prevented by inhibitors of iNOS or PHOX or by scavengers of superoxide or peroxynitrite. Importantly however, activation of PHOX alone caused no neuronal death, but did activate the microglia to proliferate  and release TNFα and IL-1β in response to fibrillar β-amyloid . PHOX has been shown by many other laboratories to be a key regulator of inflammatory activation of microglia , and thus potentially a target for anti-inflammatory strategies (65).

A growing number of recent investigations have established a critical role for leukocytes in propagating tissue damage after ischemia and reperfusion in stroke (70). Experimental data obtained from animal models of middle cerebral artery occlusion implicate inflammatory cell adhesion molecules, chemokines, and cytokines in the pathogenesis of this ischemic damage.

Cell adhesion molecules 

Three structural classes of cell adhesion molecules influence leukocyte migration, homing, and cell-cell interactions during the inflammatory response: (1) selectins, (2) integrins, and (3) proteins of the immunoglobulin superfamily.

The selectins are glycoproteins that mediate low-affinity endothelial-leukocyte interactions, thereby promoting the margination and rolling of leukocytes via interactions of carbohydrate residues. This family of cell adhesion molecules includes P-, E-, and L-selectin. P-selectin is found on platelets as well as endothelial cells, and its counterreceptor on leukocytes contains the oligosaccharide sialyl Lewis X . Pre-existing cytoplasmic stores of P-selectin in endothelial cell Weibel-Palade bodies permit its rapid mobilization to the cell surface within minutes of endothelial cell activation by thrombin, (70). complement, and histamine. 

Presumably, this early relocalization of P-selectin aids in the preliminary adhesion of leukocytes; however, there is evidence to suggest it plays a continuing role because its expression is increased in post-ischemic cerebral vasculature These results suggest a role for P-selectin in ischemic cerebral injury via the promotion of leukocyte adhesion and the development of the postischemic no-reflow state. The role of E-selectin in the pathogenesis of ischemic stroke is less well established. Unlike P-selectin, there is no preformed pool of E-selectin However, it is known that E-selectin is found in endothelial cells and leukocytes and is important in the development of the inflammatory response in vivo supports the hypothesis that leukocyte adhesion molecules contribute to neutrophil accumulation and the ensuing reperfusion injury. However, there are conflicting data from studies of human serum, in which soluble isoforms of adhesion molecules can be quantified, presumably after they are shed from activated cell surfaces. 

One study of serum samples of 22 patients with acute stroke revealed an initial increase in soluble E-selectin levels that persisted for 24 hours. In contrast, another study of 51 patients presenting with ischemic symptoms within 24 hours found no differences in circulating E-selectin levels (71). These conflicting results make the relevance of E-selectin in human cerebral ischemia unclear, although its role has been substantiated heavily in other primate experiments. L-selectin is also found in endothelial cells and leukocytes. In addition to promoting neutrophil rolling, L-selectin mediates neutrophil attachment to endothelial cells via interactions that are independent of integrin CD18. However, further data as to its mechanism have yet to be forthcoming. 

Leukocytes and the acute inflammatory response 

The leukocytes that originate from the myeloid stem cell include the monocytes and neutrophils, Leukocytes circulate in the bloodstream and enter the tissues only when they are recruited to sites of infection or inflammation. Under quiescent conditions, neutrophils do not interact closely with the cerebral endothelium. However, subsequent to the interruption of CBF that follows arterial occlusion, the acute inflammatory response is initiated by the adherence of neutrophils to the ischemic endothelium. If reperfusion is established and circulating blood returns through the vessels, it carries the additional neutrophils underlying reperfusion injury to the sites of tissue ischemia. After adhesion to the microvasculature, the neutrophils must cross the blood-brain barrier. 

Once neutrophils penetrate into ischemic brain, tissue damage is incited through their release of oxygen-free radicals and proteolytic enzymes. Recently, much interest has been shown in cerebroprotective strategies targeted specifically at neutrophils (70). Experimental models of stroke with neutrophil depletion, inhibition of neutrophil adhesion, and inhibition of neutrophil function have all been J. Huang et al. / Surgical Neurology 66 (2006) 232–245 233 shown to have reductions in infarct sizes and improved outcomes. Specifically, protein kinase C has been shown to play a significant role in neutrophil adhesion, degranulation, and superoxide generation. Knockout mice that were deletionally mutant for protein kinase C demonstrated diminished infarct volumes when subjected to transient cerebral ischemiaIn addition, neutrophils have also been implicated in ischemic injury as a source of MMP-9, a protease that degrades the basal lamina and mediates breakdown of the blood-brain barrier after cerebral parenchymal injury . Although the exact mechanism remains to be determined, it has been shown that MMP-9 is released into ischemic brain concurrently with neutrophil accumulation within ischemic microvessels. Whether these neutrophils are the true source of the surge in MMP-9 expression is unclear; however, it is known that neutrophils express MMP-9 under normal conditions (71).

Stroke is a disease that affects the blood vessels that supply blood to the brain. Cerebral ischemia occurs when the amount of oxygen and other nutrients supplied by blood flow is insufficient to meet the metabolic demands of brain tissue. In ischemic stroke, the blood supply to the brain is disrupted by cerebrovascular disease. If ischemia is of long duration or severe, death is rapid and necrotic. However, if ischemia is transient or incomplete, the genes that execute programmed cell death may be activated. Which process occurs depends on the duration and severity of the ischemic insult.

After coronary heart disease (CHD) and cancer of all types, stroke is the third commonest cause of death worldwide. Asians have a lower rate of CHD and a higher prevalence of stroke. Among the Asians, the number who died from stroke was more than three times that for CHD. Due to severity of this disease it is necessary to find out key targets that can treat the disease. There are some targets that may treat brain stroke.

Hsp can be used as novel molecular targets for pharmacological and therapeutic interventions both to prevent and to cause apoptosis. Hsp play a major role in apoptotic signaling events. However, their proapoptotic role is balanced and usually overcome by the wellknown Hsp-induced cytoprotection. This finely tuned balance is not only a key point in regulating cell death or survival but also serves as a switch between the two forms of cell death, apoptosis and necrosis.

PPAR agonists exert neuroprotective effects in models of cerebral ischaemia, neurodegenerative diseases and multiple sclerosis, with some clinical data confirming these experimental results. These results have been essentially obtained with PPARã and PPARá activators, while the PPARâ/ä pathway remains largely unexplored despite interest in the target. Developments of new and more potent PPAR activators as well as combined action of the different isoforms of PPAR are also future prospects in terms of neuroprotection and also in terms of neurorepair.

The evidence for the role of P-selectin in leukocyte-endothelial interactions in human stroke is clearer than the data for E-selectin. A greater understanding of the role of P-selectin, in particular, may inform the search for future pharmacological targets in the treatment and prevention of leukocyte rolling and adhesion, an early step toward brain infarction.

Additional understanding of the interactions between leukocytes and ischemic endothelial cells might elucidate more target-specific therapies in the future.

1) American Heart Association/American Stroke Association. Heart Disease and Stroke Statistics, 2008 Update At-a-Glance. http://www.americanheart.org/downloadable/heart/1200082005246HS_Stats%202008.final.pdf. Accessed March 27, 2008.

2) Kelly BM, Pangilinan PH Jr, Rodriguez GM. The stroke rehabilitation paradigm. Phys Med Rehabil Clin N Am. 2007;18(4):631-650.

3) Alberts MJ, Atkinson R. Risk reduction strategies in ischaemic stroke: the role of antiplatelet therapy. Clin Drug Invest. 2004;24(5):245-254.

4) Johnston SC, Gress DR, Browner WS, Sidney S. Shortterm prognosis after emergency department diagnosis of TIA. JAMA. 2000;284(22):2901-2906.

5) Johnston SC.Transient ischemic attack. N Engl J Med. 2002;347(21):1687-1692.

6) Sacco RL, Adams R, Albers G, et al. Guidelines for prevention of stroke in patients with ischemic stroke or transient ischemic attack: the American Academy of Neurology affirms the value of this guideline. Stroke. 2006;37(2):577-617.

7) Albers GW, Caplan LR, Easton JD, et al. Transient ischemic attack—proposal for a new definition. N Engl J Med. 2002;347(21):1713-1716.

8) Coutts SB, Simon JE, Eliasziw M, et al. Triaging transient ischemic attack and minor stroke patients using acute magnetic resonance imaging. Ann Neurol. 2005;57(6):848-854.

9) Johnston SC, Rothwell PM, Nguyen-Huynh MN, et al. Validation and refinement of scores to predict very early stroke risk after transient ischaemic attack. Lancet. 2007;369(9558):283-292.

10) Van der Worp HB, van Gijn J. Acute ischemic stroke. N Engl J Med. 2007;357(6):572-579.

11) Kennedy J, Hill MD, Ryckborst KJ, Eliasziw M, Demchuk AM, Buchan AM; FASTER Investigators. Fast assessment of stroke and transient ischaemic attack to prevent early recurrence (FASTER): a randomised controlled pilot trial. Lancet Neurol. 2007;6(11):961-969.

12) Jamieson DG, Parekh A, Ezekowitz MD. Review of antiplatelet therapy in secondary prevention of cerebrovascular events: a need for direct comparisons between antiplatelet agents. J Cardiovasc Pharmacol Ther. 2005;10(3):153-161.

13) Kaufmann AM, Firlik AD, Fukui MB, Wechsler LR, Jungries CA,Yonas H. Ischemic core and penumbra in human stroke. Stroke.1999;30(1):93-99.

14) Liu Y, Mituska S, Hashizume K, Hosaka T, Nukui H. The sequential change of local cerebral blood flow and local cerebral glucose metabolism after focal cerebral ischaemia and reperfusion in rat and the effect of MK-801 on local cerebral glucose metabolism. Acta Neurochir (Wien). 1997;139(8):770-779.

15) Li SC, Schoenberg BS, Wang C, et al. Cerebrovascular disease in the People’s Republic of China: epidemiologic and clinical features. Neurology 1985; 35: 1708-13.

16) Tapas Kumar BANERJEE MD FRCP (London), Neurology Asia 2006; 11 : 1 – 4

17) Shyamal Kumar DAS DM (Neurology), Neurology Asia 2006; 11 : 1 – 4

18) Steven h. Graham robert w. Hickey,molecular pathophysiology of stroke,1317-1326

19) Baron JC. Mapping the ischaemic penumbra with PET: implications for acute stroke treatment. Cerebrovasc Dis 1999; 9: 193-252.

20) Schlaug G, Benfield A, Baird A et al. The ischaemic penumbra. Operationally defined by diffusion and perfusion MRI. Neurology 1999; 53: 1528-37.

21) Martinou J-C, Dubois-Dauphin M, Staple JK, et al: Overexpression of BCL-2 in transgenic mice protects neurons from naturally occurring cell death and experimental ischemia. Neuron 13:1017-1030, 1994

22) Chen J, Graham SH, Nakayama M, et al: Apoptosis repressor genes Bcl-2 and Bcl-x-long are expressed in the rat brain following global ischemia. J Cereb Blood Flow Metab 17:2-10, 1997

23) Siesjo B, Zhao Q, Pahlmark K, et al: Glutamate, calcium, and free radicals as mediators of ischaemic brain damage. Ann Thorac Surg 59:1316-1320, 1995

24) Siesjo BK: Pathophysiology and treatment of focal cerebral ischemia: Part I: Pathophysiology. J Neurosurg 77:169-184, 1992

25) Zetterstrom TSC, Vaughan-Jones RD, Grahame- Smith DG:Ashort period of hypoxia produces a rapid and transient rise in [K1]e in rat hippocampus in vivo which is inhibited by certain K1 channel blocking agents. Neuroscience 67:815-821, 1995

26) Duchen MR: Effects of metabolic inhibition on the membrane properties of isolated mouse primary sensory neurons. J Physiol 424:387-409, 1990

27) Ben-Ari Y: Effects of glebenclamide, a selective blocker of ATP-K1 channel, on anoxic response of hippocampal neurons. Pflugers Arch 414:111-114, 1989

28) Ben-Ari Y, Krnjevic K, Crepel V: Activators of ATPsensitive K1 channels reduce anoxic depolarization in CA3 hippocampal neurons. Neuroscience 37:55- 60, 1990

29) Xie Y, Zacharias E, Hoff P, et al: Ion channel involvement in anoxic depolarization induced by cardiac arrest in rat brain. J Cereb Blood Flow Metab 15:587- 594, 1995

30) Yu SP, Yeh CH, Sensi SL, et al: Mediation of neuronal apoptosis by enhancement of outward potassium current. Science 278:114-117, 1997

31) Stys PK, Waxman SG, Ransom BR: NA1-Ca2+ exchanger mediates Ca2+ influx during anoxia in mammalian central nervous system white matter. Ann Neurol 30:375-380, 1991

32) Stys PK, Waxman SG, Ransom BR: Ionic mechanism of anoxic injury in mammalian CNS white matter: Role of Na1 channels and NA1-Ca2+ exchanger. J Neurosci 12:430-439, 1992

33) Xie Y, Dengler K, Zacharias E, et al: Effects of the sodium channel blocker tetrodotoxin (TTX) on cellular ion homeostasis in rat brain subjected to complete ischaemia. Brain Res 652:216-224, 1994

34) Boening JA, Kass IS, Cottrell JE, et al: The effect of blocking sodium influx on anoxic damage in the rat hippocampal slice. Neuroscience 33:263-268, 1989

35) Stys PK, Ransom BR, Waxman SG: Tertiery and quaternary local anesthetics protect CNS white matter from anoxic injury at concentrations that do not block excitability. J Neurophysiol 67:236-240, 1992

36) Smith SE, Meldrum BS: Cerebroprotective effect of lamotrigine after focal ischaemia in rats. Stroke 26:117-122, 1995

37) Graham SH, Chen J, Lan J, et al: Neuroprotective effects of a use-dependent blocker of voltage dependent sodium channel, BW619C89, in rat middle cerebral artery occlusion. J Pharmacol Exp Ther 269:854-859, 1994

38) Leach MJ, Swan JH, Eisenthal D, et al: BW619C89, a glutamate release inhibitor, protects against focal cerebral ischaemic damage. Stroke 24:1063-1067, 1993

39) Wiard RP, Dickerson MC, Beek O, et al: Neuroprotective properties of the novel antiepileptic lamotrigine in a geerbil model of global cerebral ischemia. Stroke 26:466-472, 1995

40) Gilland E, Malgorzata P-S, Andine P, et al: Hypoxicischaemic injury in the neonatal rat brain: effects of pre- and post-treatment with the glutamate release inhibitor BW1003C87. Dev Brain Res 83:79-84, 1994

41) Lekieffre D, Meldrum BS: The pyrimidine-derivative, BW1003C87, protects CA1 and striatal neurons following transient severe forebrain ischaemia in rats. A microdialysis and histological study. Neuroscience 56:93-99, 1993

42) Pratt J, Rataud J, Bardot F, et al: Neuroprotective actions of riluzole in rodent models of global and focal ischaemia. Neurosci Lett 140:225-230, 1992

43) Wahl F, Allix M, Plotkine M, et al: Effect of riluzole on focal cerebral ischaemia in rats. Eur J Pharmacol 230:209-214, 1993

44) Struzmann JM, Mignani S, Debarnot F, et al: Pharmacological profile and neuroprotective activities

45) Muralikrishna Adibhatla R, Hatcher JF, Department Of Neurological Surgery University of Wisconsin, Madison,WI-53792-3232, USA

46) Tsai A, Kulmacz RJ. Tyrosyl radicals in prostaglandin H synthase-1 and -2. Prostaglandins Other Lipid Mediat 2000; 62(3):231-254.

47) Candelario-Jalil E, Gonzalez-Falcon A, Garcia-Cabrera M, Alvarez D, Al-Dalain S, Martinez G et al. Assessment of the relative contribution of COX-1 and COX-2 isoforms to ischemia-induced oxidative damage and neurodegeneration following transient global cerebral ischemia. J Neurochem 2003; 86(3):545-555.

48) Miyamoto O, Tamae K, Kasai H, Hirakawa H, Hayashida Y, Konishi R et al. Suppression of hyperemia and DNA oxidation by indomethacin in cerebral ischemia. Eur J Pharmacol 2003; 459(2-3):179-186.

49) Candelario-Jalil E, Alvarez D, Merino N, Leon OS. Delayed treatment with nimesulide reduces measures of oxidative stress following global ischemic brain injury in gerbils. Neurosci Res 2003; 47(2):245-253.

50)  Al-Majed AA, Al-Yahya AA, Asiri Y, Al-Gonaiah MA, Mostafa AM. Nimesulide prevents oxidative stress damage following transient forebrain ischemia in the rat hippocampus. Res Commun Mol Pathol Pharmacol 2004; 115- 116:49-62.

51) Klegeris, A., McGeer, E.G. and McGeer, P.L (2007) Curr. Opin. Neurol. 20,

351–357

52)  Zipp, F. and Aktas, O (2006) Trends Neurosci. 29, 518–527

53) Lucas, S.M., Rothwell, N.J. and Gibson, R.M (2006) Br. J. Pharmacol. 147

(Suppl. 1), S232–S240

54) Brown, G.C. and Bal-Price, A (2003) Mol. Neurobiol. 27, 325–355

55) Block, M.L., Zecca, L. and Hong, J.S (2007) Nat. Rev. Neurosci. 8, 57–69

56) Wyss-Coray, T (2006) Nat. Med. 12, 1005–1015

57) Engelhardt, B. and Ransohoff, R.M (2005) Trends Immunol. 26,

485–495

58) Floyd, R.A (1999) Free Radical Biol. Med. 26, 1346–1355

59) Zekry, D., Epperson, T.K. and Krause, K.H (2003) IUBMB Life 55, 307–313

60) Murphy, S (2000) Glia 29, 1–13

61) Davies, K.J (1999) IUBMB Life 48, 41–47

62) Forman, H.J. and Torres, M (2002) Am. J. Respir. Crit. Care Med. 166,

S4–S8

63) Heneka, M.T. and Feinstein, D.L (2001) J. Neuroimmunol. 114, 8–18

64) Brown, G.C. and Cooper, C.E (1994) FEBS Lett. 356, 295–298

65) Bal-Price, A. and Brown, G.C (2001) J. Neurosci. 21, 6480–6491

66) Benveniste H, Drejer J, Schousboe A, et al: Elevation of the extracellular concentrations of glutamate and aspartate in rat hippocampus during transient cerebral ischemia monitored by intracerebral microdialysis. J Neurochem 43:1369-1374, 1984

67) Obrenovitch TP, Richards DA: Extracellular neurotransmitter changes in cerebral ischaemia. Cerebvasc Brain Metab Rev 7:1-54, 1995

68) Goll, D. E., Thompson, V. F., Li, H., Wei, W., & Cong, J (2003). The calpain system. Physiol Rev 83, 731– 801.

69) Bao, J. J., Le, X. F., Wang, R. Y., Yuan, J., Wang, L., Atkinson, E. N., LaPushin, R., Andreeff, M., Fang, B., Yu, Y., & Bast Jr., R. C (2002). Reexpression of the tumor suppressor gene ARHI induces apoptosis in ovarian and breast cancer cells through a caspase-independent calpaindependent pathway. Cancer Res 62, 7264– 7272.

70) Connolly Jr ES, Winfree CJ, Prestiagiacomo C, Kim S, Naka Y, Solomon, RA, Pinsky DJ, Exacerbation of cerebral injury in mice which express the P-selectin gene: identification of P-selectin blockade as a new target for the treatment of stroke. Circ Res 1996;81:304-10.

71) Diacovo TG, Roth SJ, Buccola JM, Bainton DF, Springer TA. Neutrophil rolling, arrest, and transmigration across activated, surface-adherent platelets via sequential action of P-selectin and the beta 2–integrin CD11b/CD18. Blood 1996;88:146-57.

72) Henshall, D. C., Araki, T., Schindler, C. K., Lan, J. Q., Tiekoter, K. L., Taki, W., & Simon, R. P (2002). Activation of Bcl-2-associated death protein and counter-response of Akt within cell populations during seizure-induced neuronal death. J Neurosci 22, 8458–8465.

73) Ouk, T., Petrault, O., Gautier, S., Gele´ , P., Laprais, M., Duriez, P., Bastide, M. and Bordet, R (2005) J. Cereb. Blood Flow Metab. 25, (Suppl. 1), S56

74) Collino, M., Aragno, M., Mastrocola, R., Benetti, E., Gallicchio, M., Dianzani, C., Danni, O., Thiemermann, C. and Fantozzi, R (2006) Free Radical Biol. Metab. 41, 579–589