Conference participations

Although various data support a role for the ambient Glu played in epilepsy, patterns associated with the cellular uptake of Glu have not been addressed in detail. We compared the effects of L-trans-pyrrolidine-2,4-dicarboxylate (tPDC), DL-threo-b-benzyloxyaspartate (DL-TBOA) and dihydrokainate (DHK) representing different transporter specificity and membrane-permeability profiles on recurrent seizure-like events (SLEs) evoked in hippocampal slices by low-[Mg2+] condition. In the presence of these inhibitors, the onset time and the amplitude of the first occurring SLE decreased. Uptake inhibitors frequently aborted SLEs sequential to the first occurring one. SLE duration either increased or decreased depending on the type and the concentration of the particular inhibitor applied. Simultaneous field potential and whole-cell current measurements indicated depolarization-induced inactivation of CA3 pyramidal neurons recorded after the development of the first SLE in the presence of the higher concentrations of the inhibitors (DHK 300 M; DL-TBOA 50 M; tPDC 500 M). Simultaneous measurements of intrinsic optical signal (IOS) also supported prolonged neuronal activation during SLEs. Neurochemical correlates were monitored during Glu uptake inhibition or SLEs through the analysis of perfused ACSF samples in combination with the use of HPLC-coupled tandem mass spectrometry (MS/MS) detection. In conclusion, our findings suggest a role for Glu transport in the genesis and maintenance of recurrent epileptiform discharges.

We have previously described a membrane target in subfractions of the nucleus accumbens (NAcc) that binds metabolites succinic acid (SUC) and gamma-hydroxybutyric acid (GHB), as well as the gap-junction blocker carbenoxolone (CBX). In the present study, we investigated the acute effect of SUC, GHB and CBX on propagating Ca2+ waves induced by extracellular ATP (100 μM) in NAcc slices loaded by the fluorescent intracellular Ca2+ indicator Fluo-4 AM. Confocal laser scanning microscopy has been used to detect changes in numbers and the fluorescence intensity of cells participating in the Ca2+ wave. We suggest, that ATP-induced Ca2+ wave propagates in astrocytes located in the NAcc, which can be significantly decreased by gap-junction blockers CBX (0.1 and 1 mM) and flufenamic acid (FFA 1 mM). Conversely, the non-selective P2 purinergic antagonist suramin (0.1 and 1 mM) had no significant blocking effect on the number and the fluorescence intensity of cells of ATP-induced Ca2+ waves in the NAcc, suggesting that such propagation depends mainly on the functional activity of gap-junctions. The effects of SUC and GHB were found as concentration-dependent. In low concentration (0.05 mM) SUC and GHB enhanced both the number and the fluorescence intensity of cells, while their applications in higher concentrations (0.2 and 2 mM) elicited an inhibitory effect of SUC. In addition, whole cell patch clamp recordings of tetrasodium Fluo-4 loaded NAcc medium spiny neurons showed no changes in electrical activity and intracellular Ca2+ level during the propagation of ATP-induced astrocytic Ca2+ wave. Our findings indicate that calcium waves propagate a mechanism involving of gap-junction mediated interactions between astrocytes in the Nacc. The facilitatory role in astrocytic Ca2+ wave propagation suggests that SUC may represent a strong candidate for metabolic signalling, and therefore, a benefit for further studies

Ambient level of gamma-aminobutyric acid (GABA), the major inhibitory neurotransmitter of the brain is mediated by neuronal and glial GABA transporters (GATs), members of the sodium and chloride ion-dependent solute carrier family. The neuronal GABA transporter subtype has already been proven to be a target for the antiepileptic drug Tiagabine, whereas druggability of glial GATs is far from understood. Recent advance in structure elucidation of a bacterial orthologue leucine transporter in complex with different solutes substantiates homology modeling of human GATs (hGATs) providing mechanistic clues for structure-based prediction of the potential of medicinal chemistry campaigns. A characteristic structural feature of the occluded conformation of hGATs is that similar extra- and intracellular gates are formed by middle-broken TM1 and TM6 helices. Binding crevice developed by unwound segments of broken helices facilitates symport causing GABA to fit TM1-bound Na+(1) closely inside. In accordance, a favored accommodation of substrate inhibitors with high docking score predicts efficient inhibition of transport through the neuronal hGAT-1 if the TM1 binding prerequisite were combined. Docking, molecular dynamics and uptake data indicate, that amino acids participating in substrate binding of hGAT-1 and glial (hGAT-2, hGAT-/3) subtypes are different. By contrast, substrate binding crevices of hGAT-2/3 cannot be distinguished, avoiding sensible prediction of efficient substrate inhibitors. Glial subtypes might be specifically distinguished by interfering Zn2+ binding in the second extracellular loop of hGAT-3. Formation of the unique ring-like Na+-GABA complex in the occluded binding crevices anticipates family member symporters exploring chemiosmotic energy via reversible chemical coupling of Na+ ion.

The balance between excitation and inhibition is of crucial importance for the proper functioning of the nervous system. Many neurological disorders, like epilepsy or ischemia are characterized by a disruption of this balance, usually by the abnormal intensification of excitatory signals. Here we demonstrate that the intense excitation leads to the emergence of a previously unrecognized endogenous defense mechanism by which the extracellular excitatory glutamate (Glu) is exchanged for the inhibitory γ-amino-butyric acid (GABA). This process is maintained by the concerted action of the glial Glu and GABA transporters, using the glial Glu/Na⁺ uptake as the driving force for the release of GABA. Taking advantage of the exclusive appearance of the Glu-GABA exchange mechanism during and following intense excitation periods, it may serve as an ideal target for the development of new pathomechanism-specific therapeutics.

The balance between excitation and inhibition is of crucial importance for the proper functioning of the nervous system. Many neurological disorders, like epilepsy or ischemia are characterized by a disruption of this balance, usually by the abnormal intensification of excitatory signals. Here we demonstrate that the intense excitation leads to the emergence of a previously unrecognized endogenous defense mechanism by which the extracellular excitatory glutamate (Glu) is exchanged for the inhibitory gamma-amino-butyric acid (GABA). This process is maintained by the concerted action of the glial Glu and GABA transporters, using the glial Glu/Na⁺ uptake as the driving force for the release of GABA. Taking advantage of the exclusive appearance of the Glu-GABA exchange mechanism during and following intense excitation periods, it may serve as an ideal target for the development of new pathomechanism-specific therapeutics.

We have previously reported a membrane protein that binds the intermediary metabolite of gamma-hydroxybutyric acid
(GHB), succinic acid (SUC) and the gap-junction blocker carbenoxolone in nucleus accumbens (NA) synaptosomal
subfractions. In the present study, we explored propagating Ca2+ waves induced by extracellular ATP (100 μM) and
challenged them by possible inhibitors, intermediary metabolites and their combinations. Confocal laser imaging has
been used to detect changes in the number and the fluorescence intensity of cells loaded with the fluorescent indicator
of cytosolic Ca2+ ion Fluo-4 acetoxymethyl ester. We report for the NA, that ATP-induced Ca2+ wave propagates in
astrocytes that can be decreased by gap junction blockers carbenoxolone (0.1 and 1 mM) and flufenamic acid (1 mM).
Conversely the ATP-gated cation channel (P2x) antagonist suramin (0.1 mM) had no effect on the cell number and
fluorescence intensity of cells of ATP-induced Ca2+ waves in the NA, suggesting that propagation depends on
functional activity of gap-junctions. In addition, concentration-dependent effects of SUC and GHB were described. In
low concentration (0.05 mM) SUC and GHB enhanced both the number and the fluorescence intensity of cells, while in
higher concentration (0.2 and 2 mM) an inhibitory effect of SUC was observed. Our findings indicate that in the NA,
calcium waves propagate via the mechanism involving gap-junctional connections between glutamate receptor
containing astrocytes. The facilitatory role in astrocytic Ca2+ wave propagation suggests SUC as a strong candidate for
metabolic signalling and will therefore benefit from further study.

Here we report on dynamics of synaptic input to CA3 pyramidal cells and its contribution to spontaneous emergence of recurrent seizure-like events (SLEs) in juvenile (P10-13) rat hippocampal slices bathed in low-[Mg2+] artificial cerebrospinal fluid. In field potential recordings a short epoch of high frequency oscillation (HFO; 400-800 Hz) occupied the first ~10 ms of SLEs. Voltage-clamped pyramidal cells showed GABAergic input synchronized at a 10 ms time scale arriving at the instant of SLE onset in association with HFO, while the glutamatergic input lagged by ~10 ms. In association with HFO and GABAergic input CA3 pyramidal cell firing probability peaked in case recorded with intracellular [Cl-] unperturbed (cell-attached recordings). This arrangement of HFO, synchrony and temporal sequence of synaptic inputs together with precision of neuronal firing emerged gradually with preictal discharges. Clonic discharges at SLE end showed similar dynamics with comparable GABAergic input but reduced glutamatergic excitation. Glutamatergic input shows stereotyped waveform, suggesting threshold behaviour in evoking a synchronized discharge (either SLE, preictal or clonic). Application of GABAA receptor antagonist picrotoxin (1-100 mM) blocked GABAergic currents and reduced the inter-SLE interval, but did not block recurrent SLEs. Dynamic changes of GABAergic input set the threshold for excitation and regulate precise timing of CA3 pyramidal cell firing, and are thus key players in extreme susceptibility to seizures of the juvenile hippocampus.

The major inhibitory neurotransmitter gamma-aminobutyric acid (GABA) is removed from the extracellular space by integral membrane proteins, belonging to the solute carrier 6 (SLC6) family of sodium-coupled transporters, also known as neurotransmitter sodium symporters. Certain pathological conditions, including epilepsy can be treated by inhibition of the neuronal GABA transporter functions.
Based on structural information at atomic resolution available for a bacterial homologue (LeuTAa), models of the occluded conformational state of the human neuronal (hGAT-1) and glial (hGAT-2 and hGAT-3) GABA transporters were built. Models were validated by docking a selected set of transportable substrates and inhibitors (ligands), and by subsequent molecular dynamics (MD) calculations. All docked ligands were found to interact with one of the two sodium ions Na+(1) proved to be essential for substrate-binding. The conformation of GABA was changed during MD calculations, resulting in a ring structure formed by intramolecular H-bonding between amino and carboxyl groups, the latter being in a strong electrostatic interaction with Na+(1), suggesting that hGAT-1, hGAT-2 and hGAT-3 may transport GABA in complex with Na+(1). Unraveling the zinc binding sites possibly present in the extracellular loops and negatively charged regions of the membrane-embedded helices of hGAT-2 and hGAT-3 is intimately linked to the goal of elucidating their distinguishable zinc ion-sensitivity and to the development of new antiepileptic agents in the future.

Recent data indicate the involvement of astrocytes in the development of cocaine-dependence as they show mGluR5-dependent increases of [Ca2+]i (PNAS 104, 2007, 1995-2000) and an enhanced expression of the intermediate filament glial fibrillary acidic protein (Eur J Neurosci 17, 2003, 1273-1278). Here we studied the action of gamma-hydroxybutyrate (GHB), a recently emerged recreational drug, on Ca2+ signalling in astrocytes in rat (10-24 days old) slices of the nucleus accumbens (NAc) and ventral tegmental area (VTA) incubated with the Ca2+ indicator dye Fluo-4AM and TTX (1mM). Responding cells were injected with Alexa-green, and identified as astrocytes by their morphology and S100b immunoreactivity. The number of astrocytes showing increases in [Ca2+]i was measured under different conditions, including the addition of antagonists for GABAB (CGP54626, 10mM), glutamate (CNQX, 10mM; APV, 20mM; MPEP, 30mM) and putative GHB (NCS382, 300mM) receptors. In the NAc, GHB, but not baclofen, induced transient [Ca2+]i elevations (GHB, 2 mM: 3.7±0.6 astrocytes; baclofen, 20mM: 0.4±0.3 astrocytes). The number of astrocytes responding to GHB application was unaffected by CGP54626 (2.5±0.8) and NCS382 (3.7±0.4) but was reduced by CNQX (1.7±0.6), MPEP (0.7±0.3) or APV (2.1±0.7). The GHB-induced [Ca2+]i elevations were abolished in a medium containing 0 mM Ca2+ (0.5±0.2) or after application of the SERCA inhibitor cyclopiazonic acid (10 mM; 0.4±0.4), but persisted in slices incubated with bafilomycin (4 mM; 2.9±0.4). In the VTA, GHB (2 mM) also induced transient [Ca2+]i elevations (in 4.6±0.7 astrocytes), which were unaffected by NCS382 (3.45±0.87), but partially antagonized by CGP54626 (2.9±0.5).
These data show a clear effect of GHB on astrocytes of the NAc and VTA, that might have a role in the rewarding properties of this drug. In the NAc, this action of GHB appears not to involve GABAB or putative GHB receptors.
(Supported by the Wellcome Trust, grants 68690 &71436, and NIDA, grant 5R21DA14830)

Gamma-hydroxybutyrate (GHB) has recently emerged as a major recreational drug, with associated public health problems. GHB has diverse neuropharmacological and neurobiological properties, including activation of both gamma-aminobutyric acid type B (GABAB) receptors and putative GHB receptors. An action on the mesocorticolimbic dopamine pathways has been suggested to be responsible for the addictive properties of GHB and for its withdrawal symptoms.
Results from preclinical studies have been equivocal with regard to GHB abuse potential. GHB is not self-administered by monkeys experienced in the self-administration of phencyclidine or methohexital, but is self-administered by naïve mice (0.05-0.1 mg/kg/infusion), even though only the first day of self-administration was evaluated in the latter case. In rats, oral GHB is self-administered and induces conditioned place preference.
Here we report that in naïve Wistar rats GHB does not maintain intravenous self-administration when available under continuous reinforcement (FR1, 20s time out) in a wide range of doses (from 0.01 to 1 mg/0.1 ml/infusion). When GHB rewarding properties are evaluated in an unbiased conditioned place preference (CPP) paradigm, it clearly elicits a significant CPP. When a standard conditioning procedure is used (i.e. alternating days of drug and vehicle conditioning), 10 pairings with 350 mg/kg GHB but not 175 mg/kg induce CPP. However, 350 mg/kg GHB can also elicit a clear CPP with 5 or 8 pairings, when the interval between drug and vehicle conditioning is shorten to 6 h (i.e. one drug and one vehicle conditioning per day). Preference for the oral GHB side develops regardless of whether testing is carried out in the drugged or un-drugged state, thus excluding potential state-dependent effects as an explanation of the GHB-induced CPP. These data, and results with place conditioning methodology during GHB administration in the ventral tegmental area and the nucleus accumbens, support the existence of rewarding properties of GHB in rats, which might involve interactions with different receptors present both on neuronal and astrocytic membranes (see Emri et al., this meeting).
(Supported by the Wellcome Trust, grants 68690 &71436, and NIDA, grant 5R21DA14830)

Growing amount of experimental data suggests an intricate interplay between inhibitory and excitatory neurotransmission. Extracellular levels of Glu and GABA depend not only on the balance between synaptic release and uptake but also on extrasynaptic release and the metabolic state of the tissue. We examined the effect of Glu (0.1 mM) on the radiolabelled GABA release from rat brain hippocampal slices. Inhibition of the dominant GABA transporter subtype (GAT1) by the non-transportable inhibitor NNC-711 (0.1 mM) disclosed a Glu-induced GABA release (169±12% of the control). The effect could not be blocked by inhibiting Glu receptors with CNQX (0.01 mM), AP5 (0.05 mM) and MCPG (0.5 mM). No change of Glu-induced GABA release (169±24%) was observed in nominally Ca²⁺ free, high-[Mg²⁺] (20 mM) media, suggesting that the observed GABA release is only partially attributable to some receptor-mediated effect of Glu. In contrast, application of the non-transportable Glu transporter inhibitor D,L-TBOA (0.1 mM and 1 mM) decreased the effect of Glu on the GABA release to 135±13% and 122±5%, respectively. Therefore we conclude that the mechanism underlying Glu-induced GABA release is probably the reversal of GABA transporters upon activation of Glu transporters.

Increasing evidence that tonic inhibition contributes to information processing in the nervous system emphasizes the importance of establishing the sources of ambient transmitter causing tonic inhibition. With the application of ramifying biological models at different levels of complexity, we disclose several lines of evidence that substrate activation of Glu transporters increases ambient GABA level both in vitro and in vivo. This Ca²⁺-independent elevation of [GABA]_o acts through glial GABA transporter reversal and can be eliminated by preventing Glu uptake. Underlying these phenomena, the mechanistic model provided establishes the first direct link between excitatory and inhibitory neurotransmission. The coordinated activation of Glu and GABA transporters represents a negative feedback mechanism to modulate tonic inhibition and combat intense excitation in diseases like epilepsy or ischemia. Acknowledgments: 1/A/005/2004 NKFP MediChem2, Transporter Explorer AKF-050068, GVOP-3.2.1.-2004-04-0210/3.0, Lundbeck Foundation and Danish MRC (22-03-0250), FP6 BNII No LSHM–CT 2004-503039.

Previously, a central intermediate in the TCA cycle succinate was postulated to mimic some of the actions of gamma-hydroxybutyric acid (GHB) (Molnár et al., 2006). Here we report the presence of specific, pH_o-dependent succinate binding sites in both human nucleus accumbens (NA) and rat forebrain synaptic membranes.
Specific [³H]Succinate binding sites in human NA and in rat forebrain synaptic membrane fraction were characterized by the rank order of affinity of NCS-382>GHB~succinate>carbenoxolone>>citrate~lactate. There was no interaction between specific GABA-B receptor ligands and these sites. Effective concentrations for inhibition of [³H]Succinate uptake were different (IC_50,SUCC=6.7±0.2μM, IC_50,GHB>1mM, IC_50,CIT=1.0±0.3mM). [³H]Succinate release from rat NA punches was enhanced by 2mM succinate or 0.1 mM Glu, but not by 2mM GHB. Simultaneous [¹⁴C]GABA and [³H]Succinate release measurements showed, that 50mM KCl evoked substantial decrease of [³H]Succinate release, whereas [¹⁴C]GABA release was enhanced. We conclude the existence of a pH_o-dependent synaptic membrane site for the intermediary metabolite succinate. The pharmacological and functional properties of this recognition site may possibly suggest the existence of a hemichannel-like target for succinate.

Acknowledgements: MediChem2 (1/A/005/2004 NKFP) and Transporter Explorer (GVOP 3.1.1-2004-05-00).

Membrane transporters are responsible for the clearance of L-Glutamate (Glu) from the synaptic cleft. Homology model based on crystal structures of a related archeal transporter (Yernool et al., Nature, 431:811, 2004; Boudker et al., Nature, 445:387, 2007) and key residues that take part in structural rearrangements of the human Glu transporter EAAT1 (Leighton et al., J. Biol. Chem., 281:29788, 2006) has been derived.
Molecular docking and dynamics calculations were applied to investigate binding interactions of the membrane transporter of Glu, and by comparison, the extracellular binding domain of type 1 metabotropic Glu receptor (mGlu1). Ligands, including the native transport substrate Glu, the specific Glu transport inhibitor L-trans-2,4-PDC and the non-substrate gamma-aminobutyric acid (GABA) were docked to the binding site of mGlu1 or Glu transporter modell. In line with transport data, ligands in mGlu1 were ranked as Glu > L-trans-2,4-PDC > GABA. Molecular dynamics studies revealed that Glu adopts bent and extended conformations in mGlu1 and Glu transporter, respectively. Initial molecular motions of Glu transporter helices possibly associated with function were disclosed (Simon et al., Lett. Drug Design Disc., 3:293, 2006). Acknowledgements: Center of Excellence on Biomolecular Chemistry QLK2-CT-2002-90436, Transporter Explorer AKF-050068, 1/A/005/2004 NKFP MediChem2.

In response to light, the concentration of cGMP drops in photoreceptor cells due to phosphodiesterase 6 (PDE6) activation. Within the phosphodiesterase family, the most closely related to PDE6 enzyme is PDE5, which is a target for inhibitory pharmaceuticals – like Sildenafil – aimed to treat erectile disfunctions. Due to the structural similarity between PDE6 and PDE5, however, these inhibitors also inhibit PDE6 in photoreceptor cells.
In accordance with our earlier results obtained in the isolated rat retina (Barabás P. et al., 2003), we found that PDE6 inhibitor Zaprinast (1-10 mikroM) and Sildenafil (1-10 mikroM) enhanced the amplitude of light-evoked electrical responses to 300 % and 200 %, respectively. The temporal pattern of their effects was similar. To explain the paradoxically increased light responses observable in the presence of PDE6 inhibitors, we examined binding interactions between PDE6 and Sildenafil, and between PDE6 and Zaprinast in docking experiments, utilizing the known structures of PDE5 crystallized with inhibitors.
We built a homology model on the frame of PDE5 and docked Zaprinast or Sildenafil into this model. The two inhibitors bound similarly to the catalytic centre of PDE6. Moreover, PDE-inhibitors not only occupied the place of cGMP, the native substrate within the binding pocket, but also interacted with hydrophobic amino acids near the entrance of the binding pocket, where normally the inhibitory gamma subunit of PDE6 would bind. On the basis of these results we hypothesize that catalytic site inhibitors hold back gamma subunit binding to the binding pocket. This increases the probability of enzyme activation, leading to increased light responses at the level of photoreceptor cells (Simon Á. et al., 2006).

γ-amino butyric acid (GABA) and glutamate (Glu) are the major inhibitory and excitatory neurotransmitters of the central nervous system, respectively. Their function maintains the balance and provides proper function of the brain. Although cooperation between inhibitory and excitatory signalling is a widely known phenomenon, direct coupling between these processes remained undisclosed.
By the application of ramifying biological models at different levels of complexity in combination with different analytical (radiotracer studies, NMR, HPLC, in vivo microdyalizis) approaches we disclosed several lines of evidence that extracellular Glu evokes increase in ambient GABA concentration. This process, observed in different brain regions, is mediated by glial Glu and GABA transporters. Based on the experimental results, we provide a model that directly couples inhibitory and excitatory neurotransmission. According to this model, coordinated activation of glutamate and GABA transporters determine the ratio of the ambient concentrations of major excitatory and inhibitory neurotransmitters, establishing a negative feedback mechanism.
We envision that the discovery of Glu uptake-coupled GABA release will support the development of new pathomechanism-specific treatments for diseases characterized by intense excitation, such as epilepsy or ischemia.

A synaptic binding site for gamma-hydroxybutyric acid (GHB) – a naturally occurring metabolite of succinic acid - interacting succinate has been disclosed in rat and human (nucleus accumbens, NA) sub-cellular fractions using [3H]GHB (Molnár T. et al. 2006). To address the presumed recognition site for succinate, the pharmacological profile of [3H]Succinate binding to synaptic membranes prepared from rat forebrain and human NA samples has been investigated (Molnár T. et al. 2007a, 2007b).
Specific [3H]Succinate binding sites in the human NA synaptic membrane fraction showed a strong pH-dependence and were characterized by binding of succinate (IC50,SUCC = 2.9 ± 0.6 microM), GHB (IC50,GHB = 2.2 ± 1.0 microM) and gap-junction blocker carbenoxolone (IC50,CBX = 24 ± 7 microM). A similar [3H]Succinate binding profile was found in rat forebrain synaptic membrane fractions.
To our knowledge, this is the first report on the existence of membrane recognition sites for intermediary metabolites succinate/GHB. The pharmacological profile and the pH-dependence of binding conclusively suggest that intermediary metabolites may target some gap-junction constituent connexin protein.

Molecular details of target-ligand interactions are essential for understanding the mechanisms of drug action and design. Recently determined high-resolution crystal structures of membrane-embedded neuronal target molecules provide a unique opportunity to gain insight into molecular mechanisms underlying Gluergic and GABAergic drug action.
Based on the high-resolution three-dimensional structures of the extracellular binding domains of ionotropic and metabotropic Glu receptors, molecular interactions in the binding crevices of agonists and antagonists were disclosed and stability centres of the receptors were determined. Target structure-based modelling has been substantiated in measurements of ligand binding and function (Nyikos L. et al. 2002, Kovács I. et al. 2004, Lasztóczi B. et al. 2006). Docking and subsequent molecular dynamics calculations revealed possible motions of helices harbouring the binding crevice of a bacterial Glu-transporter homologue (Simon Á. et al. 2006). Similarly, homology modelling of interactions between ligands and human GABA transporter subtype provide an opportunity to predict substrate function (Palló A. et al., in preparation).
Screening on structure-based homology models of Gluergic and GABAergic targets, validated by experimental binding and functional data may help to understand potential drug-target interactions, providing an opportunity to design novel, more efficient drugs to treat brain diseases such as epilepsy and ischemic insult.

A központi idegrendszeren ható vegyületek működési mechanizmusának megismeréséhez elengedhetetlenül szükséges a célfehérje-ligandum kölcsönhatás molekuláris szintű ismerete, melyekhez szükséges a membránba ágyazott idegi célfehérjék nagyfelbontású szerkezete. Az ionotróp és metabotróp glutaminsav (Glu) receptor extracelluláris kötődoménjének ismeretében meghatároztuk a zárt és nyitott kötőhelybe illeszkedő ligandumok kölcsönhatásait és a kötőhely stabilitási centrumait. Glu-transzporter homológ fehérjén végzett molekuladinamikai számításokkal rámutattunk a kötőhely környékén elhelyezkedő hélixek elmozdulására, míg a gamma-aminovajsav (GABA) transzporter homológján végzett dokkolási vizsgálatok a neurotranszmitter felvétel gátlásának egy lehetséges módjára mutattak rá. A retinában található foszfodiészteráz 6 enzim modelljén a gátló gamma alegység szerepét valószínűsítettük zaprinast és sildenafil hatására fellépő fényválasz növekedésben, míg a G-fehérje kapcsolt szomatosztatin receptor modelljén a TT-232 gyógyszerjelölt molekula néhány lehetséges illeszkedését határoztuk meg. Molekulamodellezési és dokkolási eljárásokkal kapott eredményeink, az osztályon folyó kísérletekkel együtt a gyógyszerjelölt molekulák és célfehérjéik működésének megértését segíthetik elő.

Gamma-aminobutyric acid (GABA) and glutamate (Glu) are the two major inhibitory and excitatory neurotransmitters in the central nervous system. They are released in inhibitory or excitatory chemical synapses between contacting neurons, while transmission is completed (and modulated) by the removal of these substances from the extracellular space into glial and neuronal cells by specific transporters. We examined the effect of Glu on the release of GABA into the extracellular space of the hippocampal brain slice of the rat. We found that inhibition of the primary GABA transporter reveals a GABA efflux which can be triggered by Glu. This GABA release is only partially attributable to a direct, receptor-mediated effect of Glu on synaptic release, as the effect cannot be blocked either by inhibiting glutamate receptors or inhibiting synaptic release itself. On the other hand, Glu-evoked GABA release is markedly decreased by TBOA, a specific inhibitor of Glu transport, suggesting that activation of Glu transporters results in GABA release through reversal of GABA transporters. This interplay of Glu and GABA transport may represent a direct regulatory feedback mechanism between inhibitory and excitatory neurotransmission.

Growing amount of experimental data suggests an intricate interplay between inhibitory and excitatory neurotransmission. Glutamate (Glu) and gamma-amino butyrate (GABA) are two major neurotransmitters of the central nervous system, whose extracellular levels depend not only on the balance between synaptic release and uptake but also on extrasynaptic release and the metabolic state of the tissue. We examined the effect of Glu (0.1 mM) on the radiolabelled GABA release from rat brain hippocampal slices. Inhibition of the dominant GABA transporter subtype by the non-transportable GAT1 inhibitor NNC-711 (100 µM) disclosed a Glu-induced GABA release (169 ± 12 % of the control), which was Ca2+-independent. The effect could not be blocked by inhibiting Glu receptors with 10 µM CNQX, 50 µM AP5 and 500 µM MCPG. As the effect in nominally Ca2+-free, high Mg2+ (20 mM) solution was the same (169 ± 24 %), the observed GABA release is only partially attributable to a direct, receptor-mediated effect of Glu on the synaptic release of GABA. In contrast, 0.1 mM and 1 mM D,L-TBOA decreased the effect of Glu on the GABA release to 135 ± 13 % and 122 ± 5 %, respectively. Because Glu-induced GABA release is markedly decreased by TBOA, a specific inhibitor of Glu transport, a plausible mechanism is the reversal of GABA transporters upon activation of Glu transporters. This transporter-mediated interplay represents a direct link between inhibitory and excitatory neurotransmission. Financial supports of 3.1.1-2004-05-00 Transporter Explorer and 1/A/005/2004 NKFP Medichem2 are acknowledged.

The balance between inhibition and excitation is dominated by GABA and Glu, the major inhibitory and excitatory neurotransmitters in the brain. Interplaying inhibitory and excitatory neurotransmissions has been demonstrated in the past decade, however, direct coupling between these functionally antagonistic neurotransmitters has remained undisclosed.
We discovered that activation of Glu transporters induces inhibitory feedback by increasing extracellular GABA level. A Ca²⁺-independent GABA efflux into the extracellular space has been identified, which can be triggered by Glu transporter substrates both in vitro and in vivo. This GABA release was eliminated by the blockade of Glu transporters with the non-transportable inhibitors DHK and TBOA, but only partially affected by the inhibition of Glu decarboxylase with semicarbazide. We propose a neuronal and a glial mechanism responsible for Glu-induced GABA release, both of them are operating through the reversal of GABA transporters. This transporter-mediated interplay represents a direct link between inhibitory and excitatory signalling, which may function as a negative feedback mechanism to avoid hyperexcitability. Under physiological conditions, Glu-induced GABA release may contribute to tonic inhibition, whereas it can provide a new therapeutic strategy to combat intense excitation in diseases such as epilepsy and ischemia.

Gamma-aminobutyric acid (GABA) transporter subtype GAT-1 performs termination of synaptic transmission and replenishes neuronal GABA pools. Certain pathological conditions, including epilepsy can be treated by inhibition of GAT-1 functions.
Based on the experimental structure of a bacterial homologue, in silico model of the human G

GABA and glutamate are the major inhibitory and excitatory neurotransmitters in the brain. Their uptake into axon terminals and glia is mediated by specific transporters (GAT1-GAT4 and EAAT1-EAAT5 for GABA and Glu, respectively). However, recent findings demonstrate the existence of a previously unrecognised mechanism that transports GABA in a Glu-sensitive manner (Héja et al., J. Med. Chem. 2004, 47, 5620-5629). To explore this mechanism, the non-transportable GABA uptake inhibitor, NNC-711 was used to cover the dominant GABA transporter GAT1 in rat cerebrocortical homogenate and the effect of Glu transport inhibitors on [³H]GABA uptake was examined. In this pharmacologically isolated system, Glu transporter substrates, like L-Glu, L-Asp, D-Asp, cysteic acid and t-PDC did inhibit GABA transport. Non-transportable Glu uptake inhibitors DL-TBOA and DHK had no significant effect on this transport. L-Glu and t-PDC were less potent in inhibiting GABA uptake in the presence of 100 µM TBOA or 100 µM DHK suggesting that they acted intracellularly. The pharmacological profile of the Glu-sensitive GABA transport process was similar to that of GAT3 (GAT4). Involvement of GAT3 and/or GAT4 in the Glu-sensitive GABA transport was also supported by its regional distribution. Inhibition of GABA uptake in the presence of 100 µM NNC-711 by L-Glu and t-PDC was more pronounced in GAT3 (GAT4)-rich human brain regions including corpus callosum, choroid plexus and spinal cord as well as the cortical white matter than in the cortical gray matter. In vivo administration of the GAT3 (GAT4) inhibitor β-alanine increased both GABA and Glu level in the contralateral cortex. In vitro and in vivo lines of evidence described so far suggest the presence of a previously unrecognized Glu-sensitive GABA transport process in the CNS.

Phosphodiesterase 6 (PDE6) is a key enzyme of the phototransduction cascade decreasing intracellular cGMP concentration upon light activation. A close homologue of PDE6, PDE5 is a target of competitive catalytic site inhibitors including sildenafil, the active principle of Viagra^TM. Since sildenafil binds to PDE6 with high affinity, concerns have been arisen on its possible visual side-effects, supported by in vitro [1] and in vivo [2] data.
Recently disclosed crystal structures of PDE5 in complex with ligands, including GMP and sildenafil offered an opportunity to build a homology model of PDE6, and disclose its catalytic binding domain. PDE6 has been modelled in the ligand-bound form using PDE5 templates in complex with GMP and sildenafil (Protein Data Bank code 1TBF and 1T9S), and subsequently binding interactions between the PDE6 catalytic site and GMP, sildenafil and another PDE5/6 inhibitor zaprinast have been explored.
It was found, that sildenafil is able to fit into the catalytic pocket of PDE6 similarly to that of PDE5. This finding is in line with data characterising sildenafil with low PDE5 versus PDE6 selectivity [3]. By measuring the effect of sildenafil and zaprinast (1 µM, 10 µM) on the light-evoked field potential of the dark-adapted isolated rat retina, we found, that - paradoxically - both inhibitors enhanced the amplitude of light-responses. Binding interactions of GMP and these cGMP-analogue inhibitors with the binding pocket residues showed that the inhibitors, but not GMP interacted with Phe778 and Met759 (sildenafil) or Met759 (zaprinast). These are the key residues known to be involved in the interaction between the catalytic binding domain and the inhibitory (γ) subunit of PDE6. Differential binding modes of GMP and the inhibitors established by docking calculations underlying competition between the catalytic site inhibitors and the inhibitory γ subunit provide an explanation for the paradoxical activation of PDE6 by catalytic site inhibitors observed [4].
Acknowledgements: This work was supported by 1/A/005/2004 NKFP MediChem2, QLK2-CT-2002-90436 Center of Excellence and OTKA F043569.

References
[1] P. Barabás, Zs. Riedl, J. Kardos, Neurochemistry International, 2003, 43, 591-195.
[2] J.K. Luu, A.V. Chappelow, T.J. McCulley, M.F. Marmor, American Journal of Ophthalmology, 2001, 132, 388-394.
[3] A. Laties, E. Zrenner, Progress in Retinal and Eye Research, 2002, 21, 485-506.
[4] Á. Simon, P. Barabás, J. Kardos, Neurochemistry International, 2006, in press.

Phosphodiesterase 6 (PDE6) is a key enzyme of the phototransduction cascade, decreasing intracellular cGMP concentration upon activation. PDE5, a close homologue of PDE6, is a target of new cGMP analogue inhibitors like sildenafil, applied in treating erectile dysfunction. Possible visual side-effects via high-affinity sildenafil binding to PDE6, however was a cause for concerns. By measuring long-term effects of sildenafil and another PDE5/6 inhibitor zaprinast (1µM, 10µM) on the amplitude of electric light response (ELR) in the dark-adapted isolated rat retina, it was found that both inhibitors enhanced and not inhibited ELR amplitudes. Moreover, these paradoxical enhancements were smaller for the more efficacious inhibitor sildenafil. To explain these phenomena, a PDE5 crystal structure-based (PDB code 1TBF) homology model of the PDE6 catalytic domain has been built and binding interactions between GMP, sildenafil, zaprinast and the catalytic binding pocket residues were compared. Docking GMP, sildenafil and zaprinast showed that inhibitors, but not GMP, interacted with Phe778 and Met759 (sildenafil) or Met759 (zaprinast), the key residues involved in the interaction between the catalytic binding domain and the PDE6-specific inhibitory subunit γ. These findings indicate that the enhanced ELR may be explained as a result of competition between the catalytic site inhibitors and the γ subunit.
Acknowledgements: This work was supported by the NKFP 1/A/005/04 (MediChem2), OTKA F043569, GVOP-3.1.1.-2004-05-0068/3.0 Transporter Explorer and Center of Excellence QLK2-CT-2002-90436.

GABA and glutamate are the major inhibitory and excitatory neurotransmitters in the brain. Their uptake into axon terminals and glia is mediated by specific transporters (GAT1-GAT4 and EAAT1-EAAT5 for GABA and Glu, respectively). However recent findings demonstrate the existence of a previously unrecognised mechanism that transports GABA in a Glu-sensitive manner (Héja et al., J. Med. Chem. 2004, 47, 5620-5629). To explore this mechanism, the non-transportable GABA uptake inhibitor, NNC-711 was used to cover the dominant GABA transporter GAT1 in rat brain homogenate and the effect of Glu transport inhibitors on GABA uptake was examined. In this pharmacologically isolated system, Glu transporter substrates, like L-Glu and t-PDC, as well as Gln did inhibit GABA transport with IC₅₀ values of 11 μM, 3.5 μM and 47 μM, respectively. Dopamine, serotonine as well as the non-transportable Glu uptake inhibitors DL-TBOA and dihydrokainate had no significant effect on this transport (IC₅₀>1000 μM). The pharmacological profile suggests that this Glu-sensitive GABA transport process is mediated by a known GABA transporter subtype, GAT3.

Acknowledgement: CBCH QLK2-CT-2002-90436, 1/A/005/2004 NKFP MediChem2 and GVOP-3.1.1.-2004-05-0068/3.0 AKF-050068 Transzporter Explorer.

Early formation of reactive oxygen species (ROS) in electron-microscopically controlled mitochondrial sub-fractions isolated from the rat cerebral cortex was studied by utilising sub-second kinetic measurements in combination with fluorescence detection of dihydroethidium oxidation. Sub-second formation of ROS in mitochondria exposed to a condition mimicking the excitotoxic situation ([Na⁺]=16.5 mM, [Ca²⁺]=2.5 μM) in the presence of 10 mM ADP, and by the enzymatic hypoxanthine oxidisation was completely inhibited by para-benzoquinone (200 mM). Substantial increase in ROS formation was observed when extramitochondrial Ca²⁺ ion concentration ([Ca²⁺]EXT) was 2.5 µM or higher, with a half-saturation time of about 30 ms. Mitochondrial Ca²⁺ ion ([Ca²⁺]MIT) enhancements, monitored with the use of the fluorescent calcium ion indicator rhod-ff, occurred with an order of magnitude lower half-saturation time, indicating that the increase of [Ca²⁺]MIT preceded ROS formation. Similar dependence of [Ca²⁺]MIT enhancement on [Ca²⁺]EXT associated with an earlier onset suggested that the enhanced [Ca²⁺]MIT did induce ROS formation in mitochondria.

Investigating GABA and Glu transport properties of novel secoergoline derivatives containing both GABA and glutamate bioisosteric motifs, identical effectiveness of compounds on the two distinct processes has been experienced. The question arose whether this unusual lack of selectivity were related to a previously unrecognised mechanism that is able to transport both GABA and glutamate (Héja et al., J. Med. Chem. 2004, 47, 5620-5629). To explore the possibility, the non-transportable GABA uptake inhibitor, 1-(2-benzhydrylideneaminooxyethyl)-1,2,5,6-tetrahydropyridine-3-carboxylic acid (NNC-711) was used to cover the known GABA transporters in rat brain homogenate and the effect of Glu transport inhibitors on GABA uptake was examined. In this pharmacologically isolated system, glutamate transporter substrates, like L-Glu and pyrrolidine-2,4-dicarboxylic acid (t-PDC), as well as glutamine did inhibit GABA transport with IC50 values of 11.0±0.2 μM, 3.5±0.1 μM and 46.8±2.1 μM, respectively. Dopamine and the non-transportable Glu uptake inhibitor DL-TBOA had no significant effect on this transport (IC50>1000 μM). These findings may indicate the existence of a novel GABA/Glu transport mechanism in the CNS, although it is not clear at present whether it is mediated by a known or a new type transporter.
Acknowledgement: CBCH QLK2-CT-2002-90436, MediChem2 1/A/005/2004 NKFP and Transzporter Explorer AKF-050068

Phosphodiesterases (PDEs) are hydrolytic enzymes that decrease the levels of cyclic nucleotides (cGMP and cAMP) in the cell. A retina-specific effector enzyme in the phototransduction cascade, PDE6, hydrolyses cGMP. High-resolution experimental structure of PDE6, however, has not been disclosed as yet.
PDE5, a cGMP hydrolyzing isoform, and a close homologue of PDE6, is abundantly expressed in the corpus cavernosum of the male reproductive organ. Inhibitors of PDE5 (Sildenafil, Vardenafil and Tadalafil) are effectively applied in treating erectile dysfunction. Since they are able to cross the blood-retina barrier, concerns have been arisen on possible visual side-effects of these drugs, supported by in vitro (Barabás et al., Neurochem. Int. 43:591) and in vivo (Luu, Am. J. Ophthalmol. 132:388) data.
The first X-ray structures of PDE5 in complex with inhibitors, including sildenafil have been released in 2004. These high-resolution crystal structures provided the opportunity to build a homology model of PDE6, and to investigate binding interactions between the PDE6 catalytic site and the inhibitors. PDE6 has been modelled in the ligand-bound form using PDE5 as a template.
It is found, that sildenafil is able to fit into the catalytic pocket of PDE6 similarly to that of PDE5. This finding is in line with data, that characterise sildenafil with low PDE5 versus PDE6 selectivity (Laties and Zrenner, Prog. Ret. Eye Res. 21:485).

Acknowledgements: This work was supported by the NKFP 1/A/005/04 (MediChem2), OTKA F043569 and Center of Excellence QLK2-CT-2002-90436

Phosphodiesterases (PDEs) are hydrolytic enzymes that decrease the levels of cyclic nucleotides (cGMP and cAMP) in the cell. A retina-specific effector enzyme in the phototransduction cascade, PDE6, hydrolyses cGMP. High-resolution experimental structure of PDE6, however, has not been disclosed as yet.
PDE5, a cGMP hydrolyzing isoform, and a close homologue of PDE6, is abundantly expressed in the corpus cavernosum of the male reproductive organ. Inhibitors of PDE5 (Sildenafil, Vardenafil and Tadalafil) are effectively applied in treating erectile dysfunction. Since they are able to cross the blood-retina barrier, concerns have been arisen on possible visual side-effects of these drugs, supported by in vitro (Barabás et al., Neurochem. Int. 43:591) and in vivo (Luu, Am. J. Ophthalmol. 132:388) data.
The first X-ray structures of PDE5 in complex with inhibitors, including sildenafil have been released in 2004. These high-resolution crystal structures provided the opportunity to build a homology model of PDE6, and to investigate binding interactions between the PDE6 catalytic site and the inhibitors. PDE6 has been modelled in the ligand-bound form using PDE5 as a template.
It is found, that sildenafil is able to fit into the catalytic pocket of PDE6 similarly to that of PDE5. This finding is in line with data, that characterise sildenafil with low PDE5 versus PDE6 selectivity (Laties and Zrenner, Prog. Ret. Eye Res. 21:485).

Acknowledgements: This work was supported by the NKFP 1/A/005/04 (MediChem2), OTKA F043569 and Center of Excellence QLK2-CT-2002-90436.

The destruction of somatostatin (SST)-containing cells is the first sign of most central nervous system (CNS) diseases, indicating an essential role for SST in CNS functioning. Increased CNS level of SST and formation of memory traces may also be related (Nyitrai et al., 2003: Eur. J. Pharmacol. 478, 111). To design specific SST analogues and/or peptidomimetics, however, it is mandatory to know the type of SST receptor (SSTR1-4) involved, and the conformation of the receptor-ligand complex. When unavailable, it can be disclosed by homology modelling of the receptor followed by ligand docking. A cyclopeptide analogue of SST, TT-232, a promising drug candidate for inhibition of tumour and inflammatory diseases, has been selected for docking (Simon et al., 2004: Biochem. Biophys. Res. Commun. 316, 1059).
SSTR family members contain seven transmembrane (7TM) helices and their X-ray structures are not available, therefore application of homology modelling is required. For SSTR1 homology modelling, transmembrane helices of bovine rhodopsin, the only 7TM receptor structure disclosed so far, are available. By exploring the sequence of rhodopsin helices as a template, the binding crevice of TT-232 in SSTR1 can be determined using computational modelling and a subsequent docking procedure. TT-232, showing a highly rigid ring conformation as revealed in NMR measurements, interacted with the following amino acid residues of SSTR1: Val133, Asp137 (helix 3), Arg197 (helix 4), Phe287, Gln291, Asn294 (helix 6), Ser305 and Tyr313 (helix 7). Docking results were validated by binding experiments performed in brain tissue membrane suspensions.
Information on the nature of binding interactions between TT-232 and SSTR1 may help to develop a strategy for the design of TT-232 peptidomimetics in the future.

This work was supported by the 1/047/2001 NKFP (MediChem) project and research grant OTKA F043569

Phosphodiesterases (PDEs) are hydrolytic enzymes that decrease the levels of cyclic nucleotides (cGMP and cAMP) in the cell. A retina-specific effector enzyme within the phototransduction cascade, PDE6, hydrolyses cGMP. High-resolution experimental structure of PDE6, however, has not been disclosed yet. A close structural homologue of PDE6, the cGMP hydrolysing PDE5, is a major drug target for medicinal chemistry, as inhibitors of PDE5 are effectively applied in treating erectile dysfunction. The first X-ray structures of PDE5 in complex with inhibitors, including sildenafil, have been released in 2004. These high-resolution crystal structures provided the opportunity to build up a homology model of PDE6, and to investigate binding interactions between the PDE6 catalytic site and the inhibitors, sildenafil and zaprinast. To elucidate binding characteristics, PDE6 has been modelled using the X-ray structure of PDE5 as a template. Similar to PDE5, sildenafil and zaprinast were seen to accommodate a binding site corresponding with that of PDE5. In measurements of function, both zaprinast [1] and sildenafil increased the light-evoked extracellular field potential (EFP) of the isolated rat retina, however the EFP response to sildenafil obeyed slower kinetics. Our findings may indicate either different binding characteristics and/or different mechanism of action for these inhibitors. Acknowledgements: This work was supported by the NKFP 1/A/005/04 (MediChem2), OTKA F043569 and Center of Excellence QLK2-CT-2002-90436.

Reference:
[1] P. Barabás, K. Antal, J. Kardos, Neurosci. Lett., 2004, 357, 195-198.

Cyclothiazide (6-chloro-3,4-dihydro-3-(2-norbornene-5-yl)-2H-1,2,4-benzothiadiazine-7-sulphonamide-1,1-dioxide, CTZ) impairs desensitisation of type AMPA glutamate receptor (GluR2). Allosteric binding of CTZ molecules in the dimeric interface region of the flip GluR2 extracellular domain crystals has been evidenced [1]. To disclose the binding characteristics of CTZ, effects of CTZ on the orthosteric agonist and antagonist binding parameters have been studied in measurements of binding of [3H](S)-F-Willardiine (agonist) and [3H]-NBQX (antagonist) to functional [2-3] AMPA receptors in freshly isolated rat brain membrane homogenates. CTZ displaced the high-affinity binding of [3H](S)-F-Willardiine by an apparently competitive mechanism (KI=60 μM) and lowered ([CTZ]=100 M) the affinity of both, the antagonist and the high-affinity agonist binding to one fifth [4]. The affinity-reduction of orthosteric agonist and antagonist binding reflects the active (non-desensitised) state of the receptor, that is attained at expense of thermodynamic work [4-6]. Molecular mechanics calculations showed that CTZ caused minor structural alterations within the orthosteric (recognition) and allosteric (modulatory) binding sites, nor did it caused significant alterations in the dimeric GluR2 subunit structures [4]. These findings together with the robust affinity-reduction of the recognition site indicate that the structural changes underlying desensitisation occur beyond the extracellular binding domain. The allosteric CTZ binding-induced increase of agonist efficacy may be interpreted in terms of a mechanism [7] involving AMPA receptor desensitisation sequential to activation. Acknowledgement: This work was supported by grants OTKA F43569, 1/047 NKFP MediChem, and Bolyai BO 00226/03.

References:
[1] Y. Sun, R. Olson, M. Horning, N. Amstrong, M. Mayer, E. Guoaux, Nature, 2002, 417, 245-253.
[2] É. Szárics, G. Nyitrai, I. Kovács, J. Kardos, Neurochem. Int., 1999, 36, 83-90.
[3] É. Szárics, L. Nyikos, P. Barabás, I. Kovács, N. Skuban, E. Temesváriné-Major, O. Egyed, P.I. Nagy, J. Kökösi, K. Takács-Novák, J. Kardos, Mol. Pharmacol., 2001, 59, 920-928.
[4] I. Kovács, Á. Simon, É. Szárics, P. Barabás, L. Nyikos, J. Kardos, Neurochem. Int., 2004, 44, 271-280.
[5] J. Kardos, D.J. Cash, J. Neurochem., 1990, 55, 1095-1099.
[6] J. Kardos, I. Kovács, E. Simon-Trompler, F. Hajós, Biochem. Pharmacol., 1991, 41, 1141-1144.
[7] J. Kardos, L. Nyikos, Trends Pharmacol. Sci., 2001, 22, 642-645.

Here we report on the inhibitory effects of a series of novel secoergoline derivatives on [³H]GABA and [³H]D-aspartate uptake in plasma membrane vesicle suspensions isolated from the rat cerebral cortex. Three of twelwe secoergoline derivatives (8, 9 and 11) containing bioisosteric sequences of GABA and Glu inhibited both GABA and Glu uptake whereby they appeared to be equipotent inhibitors with IC50 values between 270-1100 μM. In the presence of GABA and Glu transport-specific non-transportable inhibitors, inhibition of GABA and Glu transport by 8, 9 and 11 proceeded in two phases. The two phases of inhibition were characterised by IC₅₀ values between 4-180 nM and 360-1020 μM and different selectivity sequences. These findings may indicate the existence of some mechanism possibly mediated by a previously unrecognised GABA-Glu transporter.
Derivatives with the cis, but not the trans configuration (8 vs. 7 and 11 vs. 12) of bulky ester groups showed significant inhibitory effect. The cis-trans selectivity can be explained by docking these secoergolines in a three-dimensional model of the second and third transmembrane helices of GABA transporter type 1. Based on modelling studies, three residues (Asn-137, Ser-133 and Thr-89) have been identified (besides the previously recognized Tyr-140), which could possibly be involved in ligand binding.

Increased levels of somatostatin (SST) in the central nervous system (CNS) may help the formation of memory traces (1), and the destruction of SST-containing cells is the first sign of most central nervous system (CNS) diseases. Various types of SST receptors (SSTR1-4) are the target molecules for SST. A cyclopeptide analogue of SST, TT-232 is a promising drug candidate for both tumour inhibition and treatment of inflammatory diseases. To design CNS-specific drugs, however, it is mandatory to know the conformation of the receptor-ligand complex (2,3,4,5). When unavailable, it can be disclosed by homology modelling and using methods of bioinformatics.
Since X-ray structures of SSTR family members, belonging to the seven-transmembrane helices containing class (7TM) of receptors are not available, application of homology modelling is required. For SSTR1 homology modelling, transmembrane helices of bovine rhodopsin, the only 7TM receptor structure disclosed so far, are available. By exploring the sequence of rhodopsin helices as a template, the binding crevice of TT-232 in SSTR1 can be determined using computational modelling and subsequent docking procedure. TT-232, showing a highly rigid ring conformation as revealed in NMR measurements, interacted with the following amino acid residues of SSTR1: Val133, Asp137 (helix 3), Arg197 (helix 4), Phe287, Gln291, Asn294 (helix 6), Ser305 and Tyr313 (helix 7) (6). Docking results were validated by binding experiments performed in brain tissue membrane suspensions (6).
Information on the nature of binding interactions between TT-232 and SSTR1 may help to develop a strategy for the design of TT-232 peptidomimetics in the future.

* This work was supported by the 1/047/2001 NKFP (MediChem) project and research grant OTKA F043569,
** Protein Modelling Group of HAS at Eötvös Loránd University of Sciences, Budapest,
*** Department of Organic Chemistry, Eötvös Loránd University of Sciences, Budapest,
**** Peptide Biochemistry Research Group of HAS at Semmelweis University, Budapest.
REFERENCES

1. Nyitrai G., Kékesi A.K., Emri Zs., Szárics É., Juhász G., Kardos J.
GABAB receptor antagonist CGP-36742 enhances somatostatin release in the rat hippocampus in vivo and in vitro
Eur. J. Pharmacol. 478, 111-119 (2003)

2. Nyikos L., Simon Á., Barabás P., Kardos J.
Ligand-specific conformations of an ionotropic glutamate receptor
Protein Eng. 15, 717-720 (2002)

3. Kovács I., Simon Á., Szárics É., Barabás P., Héja L., Nyikos L., Kardos J.
Cyclothiazide binding to functionally active AMPA receptor reveals genuine allosteric interaction with agonist binding sites
Neurochem. Int. 44, 271-280 (2004)

4. Gogolák P., Simon Á., Horváth A., Réthi B., Simon I., Berkics K., Rajnavölgyi É., Tóth G.K.
Mapping of a protective helper T cell epitope of human influenza A virus hemagglutinin
Biochem. Biophys. Res. Commun. 270, 190-198 (2000)

5. Rajnavölgyi É., Nagy N., Thuresson B., Dosztányi Zs., Simon Á., Simon I., Karr R.W., Ernberg I., Klein E., Falk K.I.
A repetitive sequence of Epstein-Barr virus nuclear antigen 6 comprises overlapping T cell epitopes which induce HLA-DR restricted CD4+ T lymphocytes.
Int. Immunol. 12, 281-293 (2000)

6. Simon Á., Czajlik A., Perczel A., Kéri Gy., Nyikos L., Emri Zs., Kardos J.
Binding crevice for TT-232 in a homology model of type 1 somatostatin receptor
Biochem. Biophys. Res. Commun. 316, 1059-1064 (2004)