INTRODUCTION

Voltage-gated potassium currents

Voltage-gated potassium currents were among the first membrane currents to be recognized and investigated for their important roles in neuronal signaling, particularly in the repolarization of the action potential (Hodgkin and Huxley, 1952). The first K⁺-current characterized by Hodgkin and Huxley was termed delayed rectifier to describe kinetic that is characterized by delayed onset of activation (compared to the sodium current), followed by little or no inactivation, and outward rectification, i.e. a deviation from the linear current-voltage relationship of Ohm’s law.

Almost 20 years later a second type of voltage gated K⁺-current has been detected: the transient A-current (I_A) presents with rapid, transient activation in the subthreshold range of membrane potentials (-60 mV to -45 mV), fast inactivation, and fast recovery from inactivation (Connor and Stevens, 1971; Neher, 1971). These features suggest strong contribution to action potential or spike repolarization, and contribute to the regulation of the frequency of the repetitive firing [Baxter, 91; Connor, 71; Hille, 92; Liss, 01; Rudy, 88]. I_A contributes also to the resting membrane potential (RMP) by the so-called window current (an overlap in the voltage range of current activation and inactivation), and therefore I_A-channels exert a strong influence on steady state inactivation of Na⁺-channels. Thereby the A-current plays an important role in the signal processing in the dendrites, including the integration of the synaptic inputs, the filtering of fast synaptic potentials, the temporal regulation of action potential back-propagation from the soma into the dendrite, that is important for the induction of long-term potentiation by its voltage effect on Ca²⁺-permeation through NMDA receptors [Hoffman, 97; Johnston, 00; Schoppa, 99; Watanabe, 02]. Influencing action potential duration, I_A-channels strongly affect presynaptic Ca²⁺-influx and transmitter release [Pongs, 99], affecting synaptic and network excitability [Muller, 90; Muller, 91]. The importance of the somatodendritic A-current in regulating firing frequency was recently highlighted by the demonstration that the pacemaker activity of individual dopaminergic neurons in the substantia nigra, and hence the level of dopamine release, is directly correlated with the density of the A-current [Liss, 01]. For the investigation of function of those and other K⁺-currents (leak, Ca²⁺-dependent, etc.), availability of specific blockers (TEA, 4-AP, Cs⁺, dendrotoxin, charibdotoxin etc.) was crucial.

Molecular basis of voltage-gated potassium currents

Fast development of molecular biology, cloning and expression of K⁺-channels in the past 20 years made possible a better understanding of K⁺-channel physiology and pathophysiology. These studies have demonstrated that potassium channels constitute a very diverse subfamily of ion channels and are composed of variable combinations of subunits encoded in large multigene families. The Kv family of genes encodes subunits of tetrameric voltage-gated potassium channels and is divided into several subfamilies, based on sequence similarities and evolutionary relationship [Hille, 92; Coetzee, 99; Cooper, 98; Jan, 90; Pongs, 99; Pongs, 92; Rudy, 88]. This diversity contributes to the ability of specific neurons to respond uniquely to different inputs. Most of the cloned voltage-dependent channels appear to be members of four gene families. Simultaneous research on drosophila and rat resulted in two nomenclatures: voltage-gated potassium channel families can be identified by their Drosophila gene homologue (Shaker, Shab, Shaw and Shal) or by an alternative nomenclature (Kv1, Kv2, Kv3 and Kv4) [Wei, 90]. This classification has physiological consequences, because different subunits of the same family could combine to form functional tetrameric channel [Covarrubias, 91; Salkoff, 92].

Among Kv-α-subunit genes that encode the transient outward Kv channels (I_A) are Kv1.4, Kv3.3, Kv3.4, Kv4.1, Kv4.2 and Kv4.3.

Principal subunits

K v 4.2

Kv4.2 (KCND2, Shal family) protein is strongly expressed across the somatodendriticaxis of hippocampal CA1 pyramidal neurons [Sheng, 92; Maletic-Savatic, 95], and evidence suggests thatthe predominant A-type transient current throughout principalneurons of the CNS arises from channels formed by Kv4.2 [Serodio, 94]. By Northern blot analysis, [Zhu, 99; Isbrandt, 00] expression of a 6.8-kb transcript was detected only in brain, particularly in the amygdala, caudate nucleus, cerebellum, hippocampus, substantia nigra, and thalamus. Heterologous expression determined that KCND2 mediates the rapidly inactivating, A-type outward potassium current which is not under the control of the N-terminus as it is in Kv1 (Shaker) channels [Zhu, 99]. The somatodendritic A-type (Kv4.2) channels rapidly inactivate close to the action potential firing threshold and recover from inactivation in the millisecond time range [Pak, 91; Serodio, 94]. In 2001, Bähring and co-workers suggested that Kv4.2 channels in response to membrane depolarization accumulate in the closed-inactivated state, from which they directly recover, bypassing the open state [Bahring, 01]. These features distinguish somatodendritic from other A-type channels.

Pharmacologically, this channel is blocked in a dose-dependent manner by 4-amino pyridine (4-AP) and polyunsaturated fatty acids (PUFA), such as arachidonic acid (AA). The effectiveness of PUFAs, however, is dependent on the gene subfamily that encodes the channel. For example, AA is most efficient in blocking Shal (Kv4) channels, whereas for Shaker (Kv1) channels there are controversial findings [Villarroel, 96; Danthi, 03].

Kv1.4

Channels from Kv1 (KNCA, Shaker) subfamily, for instance Kv1.4, also contributes to somatic transient currents in hippocampal basket interneurons [Zhang, 95] and other neurons [Eder, 96; Pardo, 92]. Moreover, there is evidence for heteromultimerization of Kv α-subunits from the same subfamily. It has been shown that antibodies, specific to Kv1.4 co-immunoprecipitate Kv1.2 and Kv1.1 proteins in non-denaturated brain membrane extracts [Sheng, 93]. The co-expression of Kv1.4 and Kv1.2 in axons and terminals of many cells suggests the native A-type K⁺ current may result from Kv1.4/Kv1.2 heteromultimers within these compartments [Sheng, 93; Wang, 99]. Such heteromultimere channels will exhibit hybrid pharmacology, for example to TEA. Therefore, there are not enough high-affinity pharmacological tools available for Kv1.4 channels.

Kv1 transient currents inactivate with time constants that change with the voltage and recover from inactivation very slowly [Po, 93; Stuhmer, 89]. The inactivation mechanism of Kv1.4 is well understood and could be described as a ‘ball-and-chain’ interaction [Hoshi, 90; Hoshi, 91]. In case of Kv1.4 channel, inactivation domain of the N-terminal of the Kv-α subunit [Zagotta, 90] or of an accessory Kvβ subunit [Rettig, 94] binds to a ball receptor near the inner entrance of the pore [Isacoff, 91] thereby preventing ion flux through the open channel. The fast N-type inactivation of the Shaker channels is followed by a slower C-type inactivation, from which the channels recover slowly [Hoshi, 90].

Regulatory subunits of Kv4.2 and Kv1.4

A number of proteins have been discovered recently that have been shown to interact with and modify function of Kv-α proteins, including β-subunits (Kvβs), K⁺-channel Interacting Proteins (KChIPs), dipeptidyl aminopeptidase-like proteins (DPPX or DPP10), frequenin, postsynaptic density protein 95 (PSD95), filamin, etc. The physiological role of many of these proteins in native channels remains to be clarified.

β-s ubunits

Kvβ-subunits are proteins that are essential components of the Kv1 subfamily of voltage-gated potassium channels and are non-enzymatic homologues of aldo-keto reductases [Campomanes, 02; Gulbis, 99; McCormack, 94]. Transient K⁺ channels, formed by Kv1.4 proteins (and by other Kv1 α-subunits), complex with β-subunits that confer fast inactivation to otherwise non-inactivating Kv1 channels [Heinemann, 96; Rettig, 94]. Furthermore, in this case, interaction of Kv1 with beta subunits slows down the recovery from inactivation [Heinemann, 95; Heinemann, 96; Serodio, 96].

Another role assigned to the β-subunits is to act as chaperones during channel biosynthesis [Shi, 96] and thus to increase expression levels, an effect first described for the interaction of Kvβ2 with Kv1.4 [McCormack, 95].

K ⁺ -channel Interacting Proteins (KChIPs)

Potassium channel interacting proteins (KChIPs) were found to co-immunoprecipitate and co-immunolocalize with Kv4 α-subunits in brain and heart [An, 00]. All KChIP subunits possess a high sequence homology region to neuronal calcium sensor-1 (NCS-1; frequenin) and calsenilin/DREAM, which belong to the superfamily of EF-hand-containing proteins. Therefore, a role for cytoplasmic calcium in the regulation of Kv4 channel function has been suggested [An, 00; Weiss, 01].

It has been demonstrated that frequenin itself is responsible for a Ca²⁺-dependent enhancement of Kv4 expression and modulation of kinetic behaviour [Nakamura, 01]. In agreement, KChIPs interact with NH₂ terminus of Kv4 proteins and enhance surface expression, thereby increasing I_Acurrent densities when expressed in heterologous systems and in native tissues [Liss, 01; Hatano, 02; An, 00]. For example, KChIP4a delays channel opening, but when the channel opens, it favours its open state by disrupting fast inactivation and slowing channel closing [Holmqvist, 02]. In contrast, KChIP1 speeds up inactivation from closed state and accelerates channel closing [Beck, 02]. Coexpression of KChIP1 in oocytes and mammalian cells results in increased current densities, slowed onset of inactivation, and accelerated recovery from inactivation (Nakamura et al., 2001; Hatano et al, 2002). Moreover, it has been shown that a defect in the KChIP2 gene results in complete loss of transient outward current (I_to) in cardiac myocytes [Kuo, 01].

Postsynaptic density protein (PSD-95)

PSD-95, a scaffolding synapse-associated protein that possesses PDZ domains, associates with Kv4.2 and Kv1.4. It has been shown that PSD-95 influences Kv1.4 channel clustering [Hsueh, 98] and Kv4.2 channel surface expression without affecting total channel levels [Wong, 02].

Dipeptidyl peptidase-like protein (DPPX)

DPPX (also known as DPP6 or BSPL) is a glycoprotein that is structurally related to the dipeptidyl aminopeptidase and cell adhesion protein CD26, but with unknown function. DPPX is co-expressed with the principal subunits of I_A-Kv4 in a somatodendritic pattern in central nervous system (CNS) neurons. DPPX associates with the pore-forming subunit Kv4 and drastically increases their intracellular trafficking and membrane targeting [Nadal, 03].

Moreover, function reconstitution experiments in Xenopus oocytes and CHO cells demonstrated that DPPX increases the rate of inactivation of Kv4.2 currents, considerably decreases time to peak, shifts the voltage-dependence of both activation and steady-state inactivation to the left and also increases the rate of recovery from inactivation [Nadal, 03].

Molecular basis of I_K(V)

Several candidate genes exist for being the molecular determinants of the delayed rectifier potassium current in pyramidal neurons. Among them are Kv2.1, Kv2.2 and Kv1.1 [Blaine, 01]. This non-inactivating potassium current, known also as sustained potassium current, plays a major role in membrane repolarization during an action potential. Kv2 family is widespread throughout the CNS and in hippocampus, is localised on somata and dendrites of pyramidal cells and interneurones. Pharmacologically the delayed rectifier currents are sensitive to TEA and to high concentrations of 4-AP.

Structure of a Kv channel

After the cloning and expression of Kv-channel proteins it became possible to study the structure of Kv channels after cristallization and x-ray crystallography of the channel proteins. Kv potassium channels are tetramers and their monomers contain six transmembrane α-helical segments (S1-S6). An intramembrane loop between S5 and S6 (P-loop) possess specific sequence of amino acids that are responsible for potassium ions selectivity. Four subunits of voltage-gated potassium channel form the pore surrounded by integral membrane domains that can "sense" membrane voltage and open the pore in response to its change.

The "voltage sensor" (S4) features positively charged arginine residues, that enable it to move at the protein-lipid interface, within the membrane electric field, thereby allowing membrane voltage to drive the pore between closed and opened states (MacKinnon, 2003). Both, N- and C-termini are in the cytosol and play an essential role in inactivation of the channels, as well as in regulatory subunits binding.

Fig. 1 Membrane topology of a voltage gated potassium channel. Scheme of a Kv monomer. The six tansmembrane segments are shown in yellow (S1-6). A T1 recognition domain is indicated in green. Charges (+ and -) are indicated for some of the residues in the transmembrane domains of the voltage sensor (S1-4). The pore region includes S5-S6 intermembrane loop (Deutsch, 2003). Inset: Macrocomplex of Kv channel assemble with auxiliary subunits (dark blue). T1 domains associate with beta subunits. For simplicity the C termini are not shown. (Gulbis et al., 2000).

Arachidonic Acid

“The arachidonic acid cascade is arguably the most elaborated signaling system neurobiologists have to deal with”-D. Piomelli [Piomelli, 93]. It not only generates multiple messenger molecules (~ 20, an estimate limited to the brain), but these molecules may act both within and from the outside of neurons, aiming at intracellular as well as extracellular targets.

Arachidonic acid structure

Arachidonic acid (AA) is a polyunsaturated fatty acid (PUFA) consisting of a 20-atom-long carbon chain with 4 double bonds and a carboxylic group at one of the ends (20:4). The carbon-carbon double bonds in AA and other polyunsaturated fatty acids can lead to free-radical formation. Reactions with oxygen form unstable lipid peroxide compounds containing the same unstable oxygen-oxygen bond found in hydrogen peroxide and thus contributes to cellular oxidative stress. ETYA is a structural analogue of AA with chemically more stable triple bonds instead of double bonds (Fig. 2B).

Fig. 2 Structure of arachidonic acid and ETYA

Arachidonic acid release

In neurons, a receptor-dependent event, leading to Ca²⁺-influx or release of Ca²⁺from intracellular stores, is required for the brain lipases to release arachidonate from the phospholipids of the cell membrane. Several neuromodulators stimulate the deacylation of membrane phospholipids, causing release of free arachidonate. These include excitatory amino acids (e.g. glutamate), biogenic amines (e.g. serotonin and histamine), and peptides (e.g. bradykinin). Even though the final effect of these various substances on arachidonate turnover is similar, they may use different mechanisms to achieve it.

At least three distinct phospholipases are thought to generate free arachidonic acid, either directly (PLA₂) or indirectly (PLC and PLD). A number of studies have shown that all phospholipases are activated by neurotransmitters. Previous reports indicate that the key enzyme responsible for agonist-induced AA release is cytosolic PLA₂ (cPLA₂) [Lin, 92; Roshak, 94]. cPLA₂ is a cytosolic 85-kDa Ca²⁺ ^-dependent phospholipase and is activated by both an increase in intracellular free Ca²⁺ ^-concentration ([Ca²⁺ ^-]i) and Ser-505 phosphorylation by mitogen-activated protein kinase (MAPK) or protein kinase C [Leslie, 97]. Hormones and growth factors also regulate PLA₂ activity [Loo, 97], [Trevisi, 02; Leslie, 04]. In addition, H₂O₂ dose-dependently enhances cell membrane associated PLA₂ activity and stimulates AA-release [Chakraborti, 93; Yasuda, 99]. It was also suggested that H₂O₂ stimulates PLA₂ through PKC [Chakraborti, 93].

Fig. 3 Simplified scheme of arachidonic acid release events. IP3, inositol triphosphate, PLC, phospholipase C, PLA2 phospholipase A2, MAG, monoacyl glycerol DAG, diacylglycerol G, GTP-coupled protein, PIP2, phosphoinositol biphosphate

Most researchers believe that a G protein ensures the coupling of receptors with PLA₂. Jelsema and Axelrod [Jelsema, 87], have first to demonstrate inhibition of PLA₂ by G proteins.

Arachidonic acid metabolism

The major pathways of arachidonic acid metabolism have been discovered in most animal tissues, including brain. This pathways are controlled by lipoxygenase (LOX), cyclooxygenase (COX), and cytochrome (Cyt) P450 [Nishiyama, 92; Nishiyama, 93]. AA is converted into a large number of biologically active metabolites such as leukotrienes, lipoxins, prostaglandins, and thromboxanes, termed with the common name eicosanoids . These metabolites seem to play significant roles in several important processes, including vascular contraction and cell growth [Gong, 95; Anderson, 97].

In the brain, for example, 12-lipoxygenase products were shown to inhibit glutamate release from hippocampal mossy fiber nerve endings [Freeman, 91], whereas 5-lipoxygenase metabolites were found to increase the activity of muscarine-inactivated M-K⁺ channels in rat hippocampal CA1 neurons [Schweitzer, 90]. Some lipoxygenase metabolites of AA also activate the cardiac,muscarinic-activated K⁺ channel, stimulate BK_Ca channels in rat pituitary tumor cells[Duerson, 96], and modulate S-type K⁺ channels in Aplysia [Piomelli, 87]. K⁺ channel modulation by lipoxygenase products has also been reported in a number of non-neural cells, including heart myocytes [Kim, 90]. 12-Hydroxyeicosatetraenoic acid (12-HETE) is a neuromodulator that is synthesized during ischemia. Its neuronal effects includeattenuation of calcium influx and glutamate release as well asinhibition of AMPA receptor (AMPA-R) activation. 12-HETEreduces ischemic injury in the heart, but it canalso reduce neuronal excitotoxicity. 12-HETE could protect neurons from excitotoxicity by activatinga G_i/o-protein-coupled receptor, which limits calcium influx throughvoltage-gatedchannels [Hampson, 02].

Fig. 4 Simplified scheme of arachidonic acid metabolism. LOX, lipooxigenase; COX, cyclo-oxigenase; CytP450, cytochrome; HPETE, hydroperoxyeicosatrienols, HETE, hydroxyleicosatrienols; EETs, epoxyeicosatrienoics

The eicosanoids produce a wide range of biological effects on inflammatory responses, on the intensity and duration of pain and fever [Samad, 01; Vane, 76], and on reproductive function [Zahradnik, 92]. Eicosanoids also play important roles in acid secretion [Eberhart, 95], regulating blood pressure through vasodilation [Pomposiello, 01] or vasoconstriction [Yu, 03], and inhibiting [Petroni, 95] or activating [Davi, 99] platelet aggregation and thrombosis.

Modulation of voltage-gated potassium channels by arachidonic acid

AA and its metabolic products have been shown to modulatea large number of ligand- and voltage-gated ion channels in avariety of systems [Bevan, 87; Ordway, 91]. The potential role of arachidonic acid in mediating K⁺-channel modulation and presynaptic inhibition of neurotransmitter release-one example of a second messenger role for the eicosanoids-was first suggested by J. Schwartz , based on experiments with Aplysia. AA has been shownto activate a large conductance (160 pS) K⁺-channel directly in cardiac atrial muscle [Kim, 89]and a small conductance (23 pS) K⁺-channel in smooth muscle [Ordway, 91].In central neurons,AA either depresses or enhances K⁺-currents through lipoxygenase or cyclooxygenase metabolites [Keyser, 90; Schweitzer, 90; Zona, 93]. With respect to K⁺-channels in expression systems, AA has been shown to inhibitcurrents directly through channels formed by members of the Kv4(Shal) subfamily of voltage-dependent potassium channels, whereasother members of Kv1 (Shaker) subfamilies were relatively insensitiveto AA [Villarroel, 96].

High concentrations of arachidonic acid (#SYMBOL#1µM) directly modulate I_A in CA1 hippocampal neurons [Colbert, 99; Keros, 97]. Bittner and Müller showed that intracellular AA is highly potent in selectively inhibiting the A-current in cultured hippocampal neurons. Our group recently demonstrated, that extremely low concentration of AA (1 pM) attenuates I_A, but not the delayed rectifier, in CA1 pyramidal neurons from acute brain slices [Angelova, 06].

Oxidative Stress

Cells constantly generate reactive oxygen species (ROS) during aerobic metabolism. ROS are produced as by-products of oxidative metabolism, in which energy activation and electron reduction are involved. Bearing an unpaired electron, these molecules can react with a large number of neuronal targets, beginning chain reactions that generate different radical species as intermediate products.

ROS include free radicals such as the superoxide anion (O₂ ^-), hydroxyl radicals (·OH), and the non-radical hydrogen peroxide (H₂O₂). They are particularly transient species due to their high chemical reactivity and can react with DNA and proteins, i.e. enzymes, ion channels and lipids. Lipid peroxidation represents a chain reaction that results in massive oxidation of membrane lipids. ROS production is enhanced severely in several disease states such as hypertension, diabetes mellitus and atherosclerosis, but also during epileptic seizures (Kovacs et al, 2002).

The high metabolic rate of neurons implies a high baseline ROS production. Correspondingly, healthy brain cells possess high concentrations of both enzymatic and small molecule antioxidant defenses.

Antioxidants

To counteract the effects of oxidative stress, cells have developed two important defense mechanisms: a thiol reducing buffer (GSH and thioredoxin), and enzymatic systems (SOD, catalase, and glutathione peroxidase). The most important thiol is the cellular redox buffer GSH, present within cells at a millimolar concentration. Thiols also exist in different cysteine-containing compounds such as amino acids (cysteine, taurine, homocysteine), peptides (GSH, Coenzyme A), and proteins (thioredoxin, glutaredoxin, albumin, glutathione peroxidase, peroxiredoxin, redox factor-1, heat shock protein, etc).

Additionally, thiol groups located on various molecules act as redox sensitive switches, thereby providing a common trigger for a variety of radical-mediated signalling events.

Glutathione ( GSH )

The GSH system is probably the most important cellular defense mechanism that exists in the cell. The tripeptide GSH (Glu-Cys-Gly) not only acts as free radical scavenger but also functions in the regulation of the intracellular redox state. Recycling of the oxidized disulfide form (GSSG, Fig.5 A) maintains adequate levels of the reduced form of gluthathione (GSH, Fig.5 B). The system consists of GSH, glutathione peroxidase (GPx) and glutathione reductase. GPx catalyses the reduction of ROS and converts GSH to GSSG. GSSG is then reduced back to GSH by glutathione reductase that, in turn, is recycled by NAD⁺. GSH substrates include, but are not limited to hydrogen peroxide and hydroxyl radicals:

Fig. 5 Chemical structure of glutathione

The ability of the cell to regenerate GSH either by reduction of GSSG or by new synthesis of GSH is an important factor in the efficiency of that cell in managing oxidative stress. Under normal conditions, more than 95% of the GSH in a cell is reduced and so the intracellular environment is usually highly reducing [Carmody, 99]. In typical mammalian cells, the ratio of [GSH]/[GSSG] in the cytosol is 30:1, thereby maintaining very reduced redox potential (RP) of approximately 220 mV [Hwang, 92].

Ascorbic acid

Ascorbate (Vitamin C) can not be synthesised by mammals and therefore needs to be taken up through the diet. It is playing an essential role as an easily oxidized anti-oxidant and free radical scavenger. The oxidized form of ascorbic acid is known as dehydroascorbic acid. Reactive oxygen species oxidize ascorbate to dehydroascorbate. ROS are reduced to water while the oxidized forms of ascorbate are relatively stable and unreactive, and do not cause cellular damage. Ascorbic acid is also essential for the recovery of oxidized tocopherols.

Fig. 6 Chemical structure of ascorbic acid.

α-Tocopherol and Trolox C

The term Vitamin E represents eight structurally related compounds, called also tocopherols and tocotrienols, each with their own subfamilies of α-(Fig.7 B), β-, δ- and γ-substances. They are also equipped for an antioxidant function, despite their lack of a thiol (-SH) group. In its function as a chain-breaking antioxidant, tocopherols rapidly transfers its phenolic H⁺ atom to a lipid peroxyl radical, converting it into a lipid hydroperoxide and a relatively stable radical form of the vitamin [Machlin, 87]. The tocopheryl radical can be reduced to tocopherol by ascorbic acid or reduced glutathione or alternatively to be further oxidized to quinone [Bast, 02].

Fig. 7 Chemical structure of α-tocopherol and Trolox C

Trolox C (6-hydroxy-2,5,7,8-tetrametylchroman-2-carboxylic acid; Fig.7 B) is an α-tocopherol derivative that is lacking the phytyl tail, which enables its water-solubility. On the other hand, the aromatic structure remains unchanged implying preserved antioxidant properties.

Modulation of ion channels by ROS

Oxidative stress has met continuously growing interest for its roles in the physiology and pathophysiology of immune defense, ischemia, and neurodegenerative disease, such as Parkinson's disease, Alzheimer's dementia, and multiple sclerosis. Reactive oxygen species modify membrane lipids as well as a variety of proteins, including membrane channels, affecting cellular functions in both detrimental and protective ways [Kourie, 98].

Ruppersberg demonstrated in 1991 that inactivation of A-type currents through Kv1.4 channels is dependent on the cellular redox state in a cysteine-dependent manner [Ruppersberg, 91]. Oxidation decreases the conductance of Kv1.3, Kv1.4, Kv1.5, Kv3.4 [Duprat, 95] and Kv3.3 channels [Vega-Saenz de, 92], while it enhances K⁺-currents through human ether-a-gogo-related gene (HERG) K⁺-channels [Taglialatela, 97]. Modulation of K⁺-channel activity by cellular oxidative stress has emerged also as a significant determinant of vascular tone. Different kinds of ROS have been reported to modify various types of K⁺-channels in vascular tissues [Liu, 02; Sobey, 97; Pomposiello, 99]. Specific oxidative modulation of voltage activated K⁺-currents in cultured hippocampal neurons, as well as in slice CA1 pyramidal neurons by hydrogen peroxide and 1 pM AA has been demonstrated by the Müller lab [Bittner, 99; Muller, 02; Angelova, 06].

Hippocampal Formation

The entorhinal cortex (EC) is a part of the “limbic system” [MACLEAN, 52]. For a long time it was thought that the entorhinal cortex is primarily related to olfactory functions, however more recently it has been shown to also be involved in functions such as emotional and cognitive processes [Squire, 92]. The entorhinal cortex does receive major inputs from the olfactory bulb and cortex. Lately it has become clear from many studies, e.g. [O'Keefe, 78; Zola-Morgan, 90; Pouzet, 99], that the entorhinal cortex, together with the hippocampus, contributes importantly to learning and memory.

The entorhinal cortex functions as a gateway to the hippocampal formation, because its output via the perforant pathway is the major cortical source of input to the hippocampus. Furthermore, together with the subiculum it also provides the major output of the hippocampus [Witter, 89].

The cells that give input to the hippocampus are predominantly present in layers II and III, most neurons in layer II are modified pyramidal cells, or stellate cells, whereas layer III predominantly consists of small pyramidal neurons (Fig. 8). In the rat, the layer II cells of the EC have been shown to project primarily to the dentate gyrus [Steward, 76; Ruth, 82; Ruth, 88; Dolorfo, 98]. The entorhinal cortex layer III cells have been shown to project predominantly to CA1 and subiculum [Steward, 76].

Fig. 8 Schematic drawing of a section through rodent entorhinal cortex (modified after Lorente de Nó, 1934).

Hippocampal formation damage in neurological diseases

The entorhinal cortex is damaged in various brain diseases and in degenerating disorders, such as temporal lobe epilepsy [Du, 93], schizophrenia [Arnold, 95], frontotemporal dementia [Frisoni, 99] and Alzheimer’s disease [Braak, 92; Braak, 95; Braak, 96; Gomez-Isla, 96]. In schizophrenia, there are controversial findings regarding the damage in the entorhinal cortex [Akil, 97; Krimer, 97].

Changes in hippocampal formation in temporal lobe epilepsy

Selective neuronal loss and active gliosis are the basic morphologic changes observed in temporal lobe epilepsy [Mathern, 97]. The entorhinal cortex is probably one of the primary sites in which temporal lobe seizures propagate and reverberate [Spencer, 94]. In humans, early cell loss is most profound in the entorhinal cortex layer III [Du, 93]. In experimental studies of status epilepticus of rats, layer III of the medial portion of the entorhinal cortex is also suggested to be the most vulnerable [Du, 95]. At late stages, these changes are also prominent in CA1 region and in part of the subiculum.

Changes in hippocampal formation in Alzheimer’s disease

The entorhinal cortex is one of the cortical formations that are selectively vulnerable in Alzheimer’s disease. In AD, the damage in entorhinal cortex occurs very early in the disease course. Layer II shows neuronal loss and tangles formation in preclinical stages of AD and in patients with mild dementia (Braak and Braak, 1991). In later stages, amyloid accumulation and cell loss affect area CA1 and the subiculum.