RO4938581, a novel cognitive enhancer acting at GABAA α5 subunit-containing receptors

ORIGINAL INVESTIGATION

RO4938581, a novel cognitive enhancer acting at GABAA α5subunit-containing receptors

Theresa M. Ballard & Frédéric Knoflach &

Eric Prinssen & Edilio Borroni & Jeffrey A. Vivian &

Jennifer Basile & Rodolfo Gasser & Jean-Luc Moreau &

Joseph G. Wettstein & Bernd Buettelmann &

Henner Knust & Andrew W. Thomas & Gerhard Trube &

Maria-Clemencia Hernandez

Received: 30 May 2008 /Accepted: 26 September 2008 / Published online: 21 October 2008# Springer-Verlag 2008

AbstractRationale GABAA α5 subunit-containing receptors areprimarily expressed in the hippocampus and their role inlearning and memory has been demonstrated recently byboth genetic and pharmacological approaches.Objectives The objective of the study is to evaluate thecognitive effects of a novel GABAA α5 receptor inverseagonist, RO4938581 in rats and monkeys.Materials and methods The in vitro profile was determinedusing radioligand binding and electrophysiological assaysfor the GABAA α1, α2, α3, and α5 receptors. Long-termpotentiation (LTP) was performed in mouse hippocampalslices. Cognitive effects were assessed in rats in the delayedmatch to position (DMTP) task and the Morris water maze.In monkeys, the object retrieval task was used. Pro-convulsant and anxiogenic potentials were evaluated in

mice and rats. In vivo receptor occupancy was determinedusing [3H]-RO0154513.Results RO4938581 is a potent inverse agonist at theGABAA α5 receptor, with both binding and functionalselectivity, enhancing hippocampal LTP. RO4938581 re-versed scopolamine-induced working memory impairmentin the DMTP task (0.3–1 mg/kg p.o.) and diazepam-induced spatial learning impairment (1–10 mg/kg p.o.).RO4938581 improved executive function in monkeys (3-10 mg/kg p.o.). Importantly, RO4938581 showed noanxiogenic and pro-convulsive potential. RO4938581dose-dependently bound to GABAA α5 receptors andapproximately 30% receptor occupancy was sufficient toproduce enhanced cognition in the rat.Conclusions The data further support the potential ofGABAA α5 receptors as a target for cognition-enhancingdrugs. The dual binding and functional selectivity offers anideal profile for cognition-enhancing effects without theunwanted side effects associated with activity at otherGABAA receptor subtypes.

Keywords GABAA receptors . Benzodiazepine .

Inverse agonist . Long-term potentiation . Cognition .

Delayed match to position .Morris water maze .

Object retrieval .Monkey . Rat

Introduction

The GABAA receptor chloride channel is the predominantinhibitory neurotransmitter receptor in the mammaliancentral nervous system and has been widely used as atarget for neuromodulatory drugs. Many compounds in

Psychopharmacology (2009) 202:207–223DOI 10.1007/s00213-008-1357-7

T. M. Ballard : F. Knoflach : E. Prinssen : E. Borroni :J. A. Vivian : J. Basile : R. Gasser : J.-L. Moreau :J. G. Wettstein : B. Buettelmann :H. Knust :A. W. Thomas :G. Trube :M.-C. Hernandez (*)CNS Research, F. Hoffmann-La Roche Ltd.,Grenzacherstrasse 124,CH-4070 Basel, Switzerlande-mail: [email protected]

J. A. Vivian : J. BasileCNS Research, Roche Bioscience,Palo Alto, USA

Present address:J. A. VivianDepartment of Integrative Systems Neuroscience,Merck Research Laboratories,PA 19486 West Point, USA

clinical use as anxiolytics, sedatives, hypnotics, or anti-epileptics increase GABAA receptor activation via theallosteric benzodiazepine (BZD) binding site. Such com-pounds have been termed “BZD receptor agonists,”whereas BZD-binding site ligands producing the oppositeeffect, i.e., decreasing receptor activation, are called“inverse agonists.” BZD “antagonists” are ligands whichbind to the receptor without modulating its function(Haefely 1989). BZD receptor inverse agonists have so faronly been tested in animal behavior experiments and in avery few exploratory human studies. The results showedbeneficial activity against impaired cognition; however,further clinical development of these compounds wasprevented by anxiogenic or proconvulsive side effects(Dorow et al. 1983; Duka et al. 1996; Jensen et al. 1987;Little et al. 1984; McNamara and Skelton 1993; Petersen etal. 1983; Venault et al. 1986).

There is a large unmet medical need for treatment ofcognitive deficits in aging western societies with millions ofpatients suffering from Alzheimer’s disease and other typesof dementias. Currently available therapies are based eitheron cholinesterase inhibition (e.g., donepezil) or on NMDAreceptor antagonism (memantine). However, the effective-ness of current therapies is small, and cholinesteraseinhibitors suffer from mechanism-related side effects (Birksand Harvey 2006; McShane et al. 2006). There is a clearneed for a novel therapy with improved efficacy and bettertolerability. Therefore, inhibition of GABAA receptorfunction remains an attractive alternative, provided com-pounds can be made selective for the receptor subtypemainly involved in memory formation (Maubach 2003).

GABAA receptors are pentamers mostly consisting oftwo α, two β, and a γ subunit. Several gene products areavailable for each of the subunits giving rise to a largenumber of receptor variants. The importance of different αsubunit subtypes has been elucidated by the generation oftransgenic mice lacking the normal diazepam sensitivity ofthe α1, α2, α3, or α5 subunit (α4 and α6 are diazepaminsensitive). The results suggest that α1 is responsible forthe sedative effects and α2 and perhaps α3 for theanxiolytic effects of agonistic BZD receptor ligands (Löwet al. 2000; McKernan et al. 2000; Möhler 2006; Rudolphet al. 1999). The consequences of a modified pharmacologyof the α5 subunit are less evident, but reduced expressionof the subunit could be associated with facilitated cognitionin hippocampal-dependent tasks (Collinson et al. 2006;Crestani et al. 2002). This is in line with the preferentiallocalization of α5 subunits in the hippocampus (Fritschyand Möhler 1995; Pirker et al. 2000). It was thereforehypothesized that a BZD site ligand with inverse agonismselective for α5 containing GABAA receptors shouldenhance cognitive function without anxiogenic and pro-convulsant side effects.

Selectivity of a BZD site ligand can be achieved bydifferent affinities to GABAA receptor subtypes (“bindingselectivity”). Alternatively, in the case of similar subtypeaffinities, different degrees of receptor modulation (“func-tional selectivity”) can be attempted, i.e., inverse agonismat α5, but (ant)agonism at other subtypes. A compoundmay also have a combination of both binding andfunctional selectivity although, so far, this is rare. Anumber of compounds with binding and/or functionalselectivity for α5 containing GABAA receptors haverecently been synthesized and tested on animal behavior(Liu et al. 1996; Quirk et al. 1996; Savic et al. 2008;Sternfeld et al. 2004). Most results confirm the hypothesisthat compounds with such a pharmacological profile canimprove cognitive function without CNS-mediated adverseeffects (Atack et al. 2006b; Collinson et al. 2006; Dawsonet al. 2006). However, an optimal combination of potency,efficacy, binding and functional selectivity, oral bioavail-ability and brain penetration has not yet been achieved.

In this paper, we will describe the properties ofRO4938581, an imidazo-triazolo-benzodiazepine combin-ing both binding and functional selectivity at the α5containing GABAA receptor and will show that thecompound improves cognition in several paradigms oflearning and memory in rats and monkeys, with noanxiogenic or proconvulsant effects.

Material and methods

Plasmids and recombinant cell expression

cDNAs encoding the rat GABAA α1, α2, α3. and α5receptor subunits were subcloned into the pIRESpurovector, and the cDNAs encoding the rat GABAA β2 andβ3 and γ2 (short) receptor subunits were subcloned into thepIRESneo and pIREShygro vector, respectively (Clontech,Mountain View, CA, USA). Stable cell lines expressing ratGABAA α1β2γ2, α2β3γ2, α3β3γ2, and α5β3γ2 recep-tor subtypes were derived by transfection of pIRESplasmids containing the desired subunit cDNAs (α, β, γ,1:1:2 ratio) into HEK 293 cells, using the lipofectamine2000 kit according to the manufacturer’s instructions(Invitrogen, Carlsbad, CA, USA). The cells were grownin minimal essential medium (Invitrogen) supplementedwith 10% fetal calf serum (Invitrogen), 20 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES;Invitrogen), and 100 U/ml penicillin/100 μg/ml streptomy-cin (Invitrogen). The culture medium for the stable celllines contained, in addition, 0.3 μg/ml puromycin (Clon-tech, Mountain View, CA, USA), 300 μg/ml hygromycin B(Roche Diagnostics, Mannheim, Germany), and 200 μg/mlG418 (Invitrogen). For electrophysiological experiments,

208 Psychopharmacology (2009) 202:207–223

the cells were plated onto glass coverslips coated with poly-D-lysin (Sigma-Aldrich, St. Louis, MO, USA).

Membrane preparation and [3H] flumazenil-binding assays

Cells expressing the different GABAA receptor subtypeswere harvested and washed three times with cold phosphate-buffered saline and frozen at −80°C. The pellet wassuspended in cold Krebs–Tris buffer pH 7.5 containing (inmM) Tris 15, NaCl 120, KCl 100, CaCl2 25, and MgCl2 25and homogenized with a Polytron homogenizer (IKALabortechnik AG, Staufen, Germany) for 20 s at10,000 rpm. After centrifugation at 50,000×g for 60 min at4°C, the pellet was resuspended in cold Krebs–Tris bufferpH 7.5, and the protein content was measured using theBradford method (Bio-Rad Laboratories, Reinach, Switzer-land) with bovine serum albumin as the standard. Themembrane homogenate was frozen at −80°C before use. Theinhibition of 1 nM [3H]flumazenil ([3H]RO0151788) bindingby RO4938581 was measured in rat GABAA α1β2γ2,α2β3γ2, α3β3γ2, and α5β3γ2 receptor subtypes. From theIC50 the Ki was calculated using the Cheng-Prusoff equation(Ki=IC50/(1+[radioligand]/Kd); Cheng and Prusoff 1973)with respective Kd values for [3H]flumazenil binding of0.9, 3.9, 1.9, and 0.6 nM at the α1-, α2-, α3-, or α5-containing subtype, respectively. Nonspecific binding wasdetermined in the presence of 10 μM diazepam. Thepercentage inhibition of [3H]flumazenil binding, the IC50

and the Ki values were calculated using ActivityBase (IDBS;Guildford, Surrey, UK).

Whole cell patch-clamp of HEK293 cells

Experiments were performed with HEK293 cells express-ing the α1β2γ2, α2β3γ2, α3β3γ2, and α5β3γ2 GABAA

receptor subtypes. Cells plated on glass coverslips weretransferred to a Perspex chamber on the stage of NikonDiaphot inverted microscope and continuously perfusedwith a solution containing (in mM) NaCl 150, KCl 4, CaCl21.2, MgCl2 1, HEPES 10, pH 7.2 adjusted to 7.4 withNaOH and osmolarity adjusted to 340 mOsm with sucrose.Patch pipettes were pulled with an approximate tip diameterof 2 μm and a resistance of 2 MΩ with borosilicate glass(Clark Electromedical Instruments, Reading, UK) and filledwith a solution containing (in mM) CsCl 140, HEPES 10,ethylene glycol tetraacetic acid 11, MgCl2 1, CaCl2 1, andMg-ATP 4, pH adjusted to 7.2 with CsOH and osmolarityadjusted to 340 mOsm with sucrose. Cells were recorded inthe patch-clamp whole cell mode using an Multiclamp700B amplifier (Molecular Devices, Sunnyvale, CA, USA).GABA in the presence or absence of RO4938581 wasapplied to the cell for 2 s in 1 min intervals using amultibarrelled microapplicator pipette controlled by a

stepping motor (Bio-Logic SAS, Claix, France). For eachexperiment, at least three GABA control applications weregenerated and only cells showing stable GABA responseswere selected for the compound testing. Prior to applicationof a GABA–compound mixture, the same concentration ofthe compound alone was applied by bath perfusion for1 min. During the experiment, all bath solutions contained0.1% DMSO, which by itself was without detectable effecton the GABA responses. For data analysis, the pClamp dataacquisition program set (Molecular Devices) and Originsoftware package (OriginLab Corporation, Northampton,MA, USA) were used. For each concentration ofRO4938581, maximum current amplitudes were measuredand plotted as a function of the concentration. Currentamplitudes were fitted with the nonlinear least-squaresfitting routine of Origin using the equation:

I xð Þ¼Imax= 1þ x=IC50ð Þn½ �, where I is the measuredcurrent, Imax the maximum current amplitude, x theconcentration of RO4938581, IC50 is the half-maximumeffective concentration, and n the Hill slope.

Subjects

Mice Male and female DBA2/J mice (Charles River,France) aged from 19–21 days at time of testing, wereused in the audiogenic seizures test. Three- to four-month-old C57Bl/6J mice (RCC, Itingen, Switzerland) were usedin the LTP experiment.

Rats Wistar rats (RCC), 36- to 42-day-old, were used in thepopulation spike recording experiments. Male Sprague–Dawley rats (Charles River, France) with approximateweight of 200 g were used in the following experiments:spontaneous locomotor activity, neurological assessment,elevated plus maze, stimulus rats in the social approachavoidance (SAA) test (with approximate weight of 450 g).Male Fischer rats (F-344, Charles River, Germany) withapproximate weight of 200 g were used in the SAA test.Male Lister hooded rats (Harlan, Netherlands) with approx-imate weight of 450 g (at time of drug testing) were used inthe following tests: chain pull responding, delayed match toposition (DMTP) task and, with approximate weight of300 g, in the Morris water maze and in the in vivo bindingexperiments. All rodents were group-housed in holdingrooms at controlled temperature (20–22°C), humidity (55–65%), and 12-h light/dark cycle (lights on at 06:00 hours).All rodents were allowed ad libitum access to food and waterexcept those used in the operant conditioning tests, wherefood was limited to that earned in the test session and 12–15 g/rat at the end of the day. The experimental proceduresused in the present investigation received prior approvalfrom the City of Basel Cantonal Animal Protection Com-mittee based on adherence to federal and local regulations.

Psychopharmacology (2009) 202:207–223 209

Monkeys Adult (5–12 years; 6–9 kg) male cynomolgusmacaques (Macaca fascicularis) were housed in same-sexpairs in a colony room maintained at 21±2°C, 40±10%humidity, and on a normal 12-h light/dark cycle (lights onat 07:00 hours). Food (Purina High Protein no. 5045) andwater were available ad libitum. All experimental proce-dures were approved by the Institutional Animal Care andUse Committee of Roche Palo Alto and were in accordancewith National Institutes of Health guidelines.

Population spike recordings in rat hippocampal slices

Rats were anesthetized in a 2.5% isoflurane/96.5% oxygenmixture and sacrificed. The hippocampi were dissected and400-μm slices were transversely cut with a tissue chopper(Sorvall, Newton, USA). The slices were then maintainedin a submerged chamber and perfused at room temperaturein a solution containing (in mM): NaCl 124, KCl 5, MgSO4

2, CaCl2 2, KH2PO4 1.25, NaHCO3 25, D-glucose 11oxygenated with 95% O2/5% CO2 (pH 7.4, 307 mOsm).Population spikes (PS) were recorded from the CA1 regionof the hippocampus similarly as described by Kemp et al.(1987). A glass micropipette (1–3 MΩ) containing 2 MNaCl was positioned in the stratum pyramidale andinsulated bipolar platin/iridium electrodes were positionedin the Schaffer collaterals for orthodromic stimulation(0.1 ms, 60–90 μA). The stimulus intensity that induced aPS near the maximum amplitude was used as the teststimulus. PS were amplified with an Cyberamp 380amplifier (Molecular Devices), filtered at 2.4 KHz anddigitized at 20 KHz with a Digidata 1322 acquisition board(Molecular Devices) for subsequent storage on a personalcomputer (Compaq Deskpro EN).

To obtain concentration–response relationships, PSwere evoked at 30-s intervals and increasing concen-trations of isoguvacine were applied to the slice until amaximum inhibition of the PS was reached. For eachconcentration of isoguvacine, the PS amplitude obtainedfrom four averaged responses was measured. Isoguva-cine was then removed from the salt solution. After afull recovery of the PS, RO4938581 was preapplied tothe slice for 15 min, and then another isoguvacineconcentration response curve was generated in thepresence of the compound. PS amplitudes were fittedwith the nonlinear least-squares fitting routine of Originusing the equation:

PS xð Þ¼PSmax= 1þ x=IC50ð Þn½ �, where PS is the measuredPS amplitude, PSmax the maximum PS amplitude, x theconcentration of isoguvacine, IC50 the half-maximumeffective concentration, and n the Hill slope. The data areexpressed as mean±SEM. IC50s were analyzed using a two-tailed, paired t test.

Long-term potentiation

Mice were anesthetized in a 2.5% isoflurane/96.5%oxygen mixture and decapitated. The hippocampi weredissected and 400-μm slices were cut with a Sorvalltissue chopper. Slices were maintained in an interfacechamber and perfused at 35°C with a simple saltsolution containing (in mM): NaCl 124, KCl 2.5,MgSO4 2, CaCl2 2.5, KH2PO4 1.25, NaHCO3 26,D-glucose 10, sucrose 4, gassed with 95% O2, and 5%CO2 (pH 7.4, 307 mOsm). Field excitatory postsynapticpotentials (fEPSPs) were recorded from the CA1 stratumradiatum with a glass micropipette (1–3 MΩ) containing2 M NaCl and evoked by stimulation of the Schaffercollaterals with insulated bipolar platin/iridium electrodes>500 μm away from the recording electrode. The stimulusstrength was adjusted to evoke fEPSPs equal to 30% ofthe relative maximum amplitude without superimposedpopulation spike. After stable baseline recordings, long-term potentiation (LTP) was induced by a brief primingstimulus (ten stimuli at 100 Hz) followed by a burststimulus (four times ten stimuli at 100 Hz at an intervalof 20 s). The duration of the stimulation pulses wasdoubled during the tetanus. This stimulus paradigm hasbeen found to be sensitive to allosteric modulators ofGABA receptors (Seabrook et al. 1997). For the analysis,fEPSP slopes were expressed as a percentage of thebaseline values recorded 10 min before the primingstimulation. Results from several slices were expressedas mean±SEM.

Compounds and administration

RO4938581, RO0194603, [3H]-RO0154513, [3H]-RO0151788, L-655,708, and diazepam were synthesizedat F.Hoffmann-La Roche for in vivo administration,RO4938581 was prepared in 0.3% Tween 80 v/v 0.9%saline (rat), or in 7.5% gelatine v/v 0.2% saline (mouse), orin HEC (hydroxy-ethyl-cellulose 250 HHX) (monkey).RO4938581 was administered per os (p.o.) with a fixedpretreatment time of 1 h (rat), 2 h (mouse), and 6.7 h(monkey). The pretreatment times were selected from theappropriate pharmacokinetics data as the timecorresponding to maximal drug plasma concentration.Scopolamine hydrobromide (Sigma) was prepared in 0.9%saline and administered s.c. 30 min prior to testing.Diazepam was prepared in 0.3% Tween 80 v/v 0.9% salineand was administered i.p. 30 min prior to testing.RO0194603 was prepared in 0.3% Tween 80 v/v salineand administered either p.o. with a pretreatment time of30 min (rat) or i.p. with a pre-treatment time of 15 min(mouse). All doses of compounds are expressed as that ofthe base.


Determination of receptor occupancy

Lister hooded rats were pretreated with vehicle orRO4938581 at dose 1, 3, and 10 mg/kg, p.o. (three ratsper dose group), and 50 min later, received intravenously0.1 mCi/kg of [3H]-RO0154513 (specific activity 53 Ci/mmol). To determine the amount of nonspecific-bindinggiven by the radioligand, an additional group of rats wastreated with the selective GABAA α5 ligand L-655,708(Atack et al. 2006b) at dose 10 mg/kg, i.p. 30 min beforeadministration of the radioligand. Preliminary experimentsshowed that L-655,708 at dose 10 mg/kg producedmaximal occupancy of GABAA α5 receptors in the rathippocampus. All rats were sacrificed 10 min afteradministration of the radioligand; brains were rapidlyremoved and frozen in dry ice. Sagittal sections (10 μmthick) were cut in a cryostat, mounted on Histobond glassslides (Marienfeld Laboratories Glassware, Germany),dried at room temperature and exposed, together withtritium microscales, to tritium-sensitive imaging plates(BAS-TR2025) for 5 days. The imaging plates werescanned in a Fujifilm BAS-5000 high-resolution phosphorimager, and the amount of [3H]-RO0154513 bound to thehippocampus was quantified with an MCID M2 imageanalysis system (Imaging Research, St. Catherines,Ontario, Canada). To obtain the specific binding (SB),radioactivity in the hippocampus of L-655,708-treatedanimals (i.e., nonspecific binding) was subtracted from theradioactivity in the hippocampus of vehicle- andRO4938581-treated animals. The extent to which specificbinding is reduced relative to vehicle-treated animalsreflects the degree to which RO4938581 occupies thebenzodiazepine site of the GABAA α5 receptor popula-tion. Receptor occupancy was calculated according to thefollowing equation:

% receptor occupancy ¼ 1� SB RO4938581=SB vehicleÞ � 100ð

where “SB RO4938581” is the specific binding measuredin the hippocampus of rats pretreated with RO4938581,and “SB vehicle” is the specific binding measured in thehippocampus of rats pretreated with the vehicle.

Spontaneous locomotor activity

Sprague–Dawley rats were placed into a novel test chamberwhich consisted of a Plexiglas® box (41 cm×41 cm×30.5 cm)with sawdust bedding on the floor. Movement of the animalresulted in interruption of an array of photobeams fromhorizontally located infrared sources placed around the testchamber (Omnitech Electronics, Dartmouth, Canada). Theeffect of RO4938581 (0.1, 1, 10 mg/kg; n=8/group) on totalhorizontal activity was recorded over a 1-h period.

Neurological assessment

Sprague–Dawley rats were observed over a 5- to 10-minperiod and were assessed for body posture and gait, rightingreflex, grasping reflex, grip strength, rotarod performance(at 8 and 16 rpm) and observed for any other symptoms(see Higgins et al. 2001 for further details). The effect ofRO4938581 (3, 10, 30 mg/kg, n=6/group) was comparedto vehicle.

Operant conditioning tests

The operant conditioning boxes were obtained from MedAssociates (St. Albans, VT, USA), and the protocols wererun by the Kestrel Control system from ConclusiveMarketing (Harlow, UK) operating on an IBM compatiblePC. For both tests, a Latin-squares design was used toassess drug effects in rats that had reached an asymptoticlevel of performance. Rats were tested twice weekly with atleast a 2-day interval between test sessions. Rats weretrained between test days to maintain baseline performance.

Chain pull for food reward under VI20 schedule

Food-deprived Lister hooded rats were previously trained topull a ceiling chain for food reward (45 mg Formula PNoyes pellet). Schedule requirements were graduallyincreased to a final value of VI20 (variable interval of20 s), and animals were ready for drug testing once stableresponding had been attained. The effect of RO4938581 (1,3, 10, 30 mg/kg, n=12) was assessed on the total number ofchain pulls and number of food rewards obtained during the30-min session.

Delayed match to position task

The training procedure and experimental conditions wereidentical to that described by Higgins et al. (2004). In thistask, food-deprived Lister Hooded rats were previouslytrained to press a lever during the sample stage and thenthey have to remember this response during the choicestage. During the choice stage, two levers are presented,and subjects have to press the same lever as the sample (i.e., matching) to receive a food reward. There is a variabledelay interval (1–24 s) between the sample and choicestage, so that at short delays (1–4 s), the choice accuracy ishigh, and as the delay interval increases (>8 s), the animalsbecome progressively worse because of the increasedmemory load. The parameters measured included percentcorrect [total no. correct/(total no. correct + incorrect)×100]both for the total session and for each delay interval.Omissions and latency to press the sample lever, choicelever, and collect the food reward (magazine latency) were


also measured to give an assessment of drug effects onoverall performance. The effect of RO4938581 wasassessed alone (0.1, 1, and 10 mg/kg; n=12) and versusscopolamine (0.03 mg/kg)-induced impairment (0.1, 0.3,and 1.0 mg/kg p.o.; n=11).

Morris water maze

The water maze is a large pool (2 m diameter and 0.7 m high)filled with water made opaque using a white artificial opacifier(E308, Bronx Chemicals, West Yorkshire, UK) and sur-rounded by extramaze cues. The water temperature (21°C)was constant throughout the experiment. The maze wasarbitrarily divided into four quadrants: NE, NW, SE, SW;and a colorless perspex circular platform (d=10 cm) waspositioned at the center of one of these quadrants, 1–2 cmbelow the water surface. A computer tracking system (HVSImage Ltd., UK), was used to analyze each rat’s swim pathon-line. Each rat started at a sequential position on each trial,and the maximum length of each trial was 60 s. Platformpositions were assigned in a semirandom order, ensuring thateach position was used an equivalent number of times withineach treatment group. If a rat found the platform during thetrial, it was left on the platform for 10 s. If the rat did not findthe platform by the end of the trial, it was guided toward it,allowed to climb onto the platform, and left there for 10 s.There was an intertrial interval (ITI) of 10 min betweentrials, during which the animals were returned to their homecage. Rats were initially pretrained to find a hidden platformposition (two trials per day for 5 days). Subjects weresemirandomly divided into treatment groups, based on theirperformance on the last training day.

Test day Subjects had to learn to locate a hidden platformplaced in a new position in the maze. The test sessionconsisted of six acquisition trials and a probe trial (platformremoved) to assess the extent of spatial learning. The effectof RO4938581 (1, 3, 10 mg/kg) versus a diazepam (6 mg/kg)-induced impairment was assessed (n=9–10/group). Themean latency, path length, and swim speed were assessedover the six trials. In the probe trial, the percent time spentsearching for the previous platform in each quadrant (left,platform, right, and opposite), and the number of platformcrossings were measured. Due to the duration of theprotocol, testing was carried out over 2 days (half of eachdose group was tested on each day).

Object retrieval task in adult monkey

This task requires a monkey to reach into a clear acrylic box(dimensions: 5×5×5 cm) with one open side, to retrieve afood treat (cubes of apple or pear, 1–2 cm3). The box was

positioned in front of the monkey and outside of the homecage, with the open side facing left, right, or toward themonkey. Food treats were placed on the outer edge, inneredge, or deep within the box. A test session consisted of 17trials with nine “easy” food retrievals (i.e., placement of thefood reward on the inner or outer edges of the box or whenthe opening is toward the monkey) and eight “difficult” foodretrievals (i.e., placement of the food reward deep within thebox and the open side facing left or right). The order ofpresentation never varied, there were no contingencies forincorrect reaches (i.e., monkeys typically acquired the treatafter the incorrect reach) or dropped treats, and trials wereterminated if there were no reaches within 3 min. For moredetails, see Rutten et al. (2008). Administration ofRO4938581 (0.1, 0.3, 1, 3, and 10 mg/kg, n=12) andbehavioral measurement were completed blind. Compounddoses were administered in a pseudorandom order.

Anxiety tests: social approach avoidance and elevated plusmaze

A tracking system (Ethovision, Noldus, The Netherlands)was used for both test procedures.

SAA The apparatus consists of a black plastic test box dividedinto two separate compartments, social and nonsocial,connected by a sliding door. The social compartmentcontained a subchamber delimited by a perforated transparentwall in which the stimulus (Sprague–Dawley) rat wasconfined. The test started by introducing a test rat (Fischer)into the nonsocial compartment for a 3-min habituationperiod. After 3 min, the sliding door was opened, allowingthe rat to move freely between the two compartments for10 min (see Nicolas and Prinssen 2006, for further details).The effect of RO4938581 (0.3, 1, 3 mg/kg, n=10/group) ontime spent in the hidden zone (within the nonsocial area) wasassessed. The positive control, RO0194603, was adminis-tered at 3 mg/kg in a separate experiment (n=7-8/group).

EPM Sprague–Dawley rats were placed on the maze for a5-min period (see Martin et al. 2002 for further details).Light intensity was set at lowest possible (i.e., 60 lx in themiddle of the maze) in order to increase sensitivity to detectanxiogenic-like activity. The effect of RO4938581 (0.3, 1, and3 mg/kg; n=10/group) on time spent on the open arms wasassessed. The positive control, RO0194603, was adminis-tered at 3 mg/kg in a separate experiment (n=12/group).

Audiogenic seizures

The apparatus for acoustic stimulation consisted of a soundisolated wooden chamber (inside dimensions, 45×45×


45 cm), with a Plexiglas three-layered insulated front doorpermitting full-view observation. Uniform sound distribu-tion in the chamber was ensured by four high-powertweeters HF50 from Celestion Co (Ipswich, UK) built intothe ceiling of the chamber. Two pairs of mice were placedin separate rectangular Plexiglas transparent boxes (21×44×21 cm) with sawdust bedding where they couldsimultaneously be observed. Testing involved exposure toa 14-kHz sinusoidal tone at 70 dB measured 2 cm above thefloor for 1 min. Sound stimulation began immediately aftertransfer of the mice to the test chamber. After vehicletreatment, such acoustic stimulation typically induced tonicconvulsions in 25% of mice. The proportion of mice thatexhibited tonic convulsions (n=8/group) following treat-ment with RO4938581 (3, 10, 30, 100 mg/kg) or thepositive control RO0194603 (0.03 mg/kg) were recorded.

Data analysis

Horizontal activity across time bins was analyzed by arepeated measures analysis of variance (ANOVA), followedby a one-factor ANOVA and post hoc Newman–Keuls testat each time bin. Total horizontal activity in 1 h and gripstrength were analyzed by a one-factor ANOVA. Nonpara-metric data (horizontal wire test, rotarod) were analyzed byKruskal–Wallis ANOVA. Chain pull and DMTP data wereanalyzed by a one-factor repeated-measures ANOVA,except for choice accuracy across delays which wasanalyzed using a two-factor repeated measures ANOVA.In all significant cases, this was followed by a post hocNewman–Keuls test. All water maze data were analyzedwith a one-factor ANOVA, except for time spent in eachquadrant for each treatment group which was analyzed witha repeated measures ANOVA. All significant cases werefollowed by a post hoc Newman Keuls test. Monkey objectretrieval data were analyzed with a one-factor repeated-measures ANOVA, followed in significant cases by a posthoc Dunnett analysis. SAA and elevated plus maze datawere analyzed with a one-factor ANOVA, except for thepositive control which was analyzed separately with anunpaired t test. Number of mice per group exhibiting tonicconvulsions compared to the vehicle group were analyzedby a Chi square test. In all studies, the accepted level ofsignificance was p<0.05.

Results

Binding affinity

The structure of RO4938581 (3-bromo-10-difluoromethyl-9H-imidazo[1,5-a][1,2,4]triazolo[1,5-d][1,4]benzodiaze-pine) is shown in Fig. 1. Inhibition of [3H]flumazenil

binding showed that the compound binds with higheraffinity (4.6 nM) to the BZD binding site of the ratGABAA α5β3γ2 receptor subtype than to receptorscontaining α1, α2, or α3 subunits in conjunction with β3and γ2 subunits (80–185 nM, see Table 1). Comparableaffinities were also observed for transiently expressedhuman GABAA receptors (data not shown).

Effects of RO4938581 on GABA-induced responsesin HEK293 cells

To determine the maximal allosteric modulation of theGABA responses in HEK293 cells stably expressing theα1β2γ2, α2β3γ2, α3β3γ2, and α5β3γ2 GABAA recep-tor subtypes, a representative inverse agonist, the β-carboline β-CCM was tested. In all cells, β-CCM(1 μM) decreased the current induced by a GABAconcentration evoking an ∼20% maximum response. Themaximum inhibition for GABA-induced current inhibitionby β-CCM were −54±3%, −39±3%, −40±4%, and −54±4% for the α1β2γ2, α2β3γ2, α3β3γ2, and α5β3γ2receptor-expressing cells, respectively (n=8–14, data notshown). RO4938581 concentration dependently inhibitedthe GABA-induced current in cells expressing α5β3γ2receptors, (Fig. 2). The maximum inhibition was −46±5%,n=6; the pIC50 value and Hill slope calculated from theconcentration–response curve were 8.13±0.05 and 1.45±0.4, respectively. In contrast to β-CCM, RO4938581 wasnot or only slightly effective at the other receptor subtypes(α1β2γ2, α2β3γ2, α3β3γ2; Fig. 2).

N

N

N

NN

Br

F

FFig. 1 Structure of RO4938581(3-bromo-10-difluoromethyl-9H-imidazo[1,5-a][1,2,4]triazolo[1,5-d][1,4]benzodiazepine)

Table 1 Affinity of RO4938581 for rat recombinant GABAA receptorsubtypes

GABAA receptor subtype Ki (nM)

α1β2γ2 174±28α2β3γ2 185±5α3β3γ2 80±16α5β3γ2 4.6±1.4

Inhibition of [3 H]flumazenil binding was used to measure affinity ofRO4938581 at the different GABAA receptor subtypes. Data shown asmean±SD from two separate determinations.


Modulatory effect of RO4938581 on population spikes

The hippocampal brain slice preparation can be used toassess quantitatively the potencies of compounds acting atthe GABAA receptor. In particular, BZD receptor ligandshave been shown to modulate the inhibition of synapticallyevoked CA1 population spikes (PS) by isoguvacine, aGABA-mimetic compound (Kemp et al. 1987). Wetherefore used this protocol to examine the effect ofRO4938581 on native GABAA receptors in rat hippocam-pal slices. As described by Kemp et al. (1987), isoguvacineinduced a concentration-dependent decrease in PS ampli-tude reaching complete inhibition at 128 μM (Fig. 3).When RO4938581 (300 nM) was applied in the absence ofisoguvacine, no change of the PS amplitude was observed.However, the concentration response curve for PS inhibi-tion by isoguvacine was shifted to the right by RO4938581(300 nM, Fig. 3), i.e., the compound acted as a negativemodulator of isoguvacine. The pIC50 values in the absenceand presence of 300 nM RO4938581 were 4.57±0.07(27 nM) and 4.35±0.05 (45 nM), respectively, and signifi-cantly different (four slices, p<0.05, two-tailed t test).

Effect of RO4938581 on hippocampal LTP

The plasticity of Schaffer collaterals—CA1 synaptic trans-mission was investigated in the absence and presence ofRO4938581 (Fig. 4). To induce long-term potentiation, wehave selected a weak conditioning stimulus that was

previously shown to activate GABAA receptors and to besensitive to BZD receptor ligands (Seabrook et al. 1997). Incontrol experiments, the mean fEPSP slopes 1–5 min afterthe priming stimulus and 1–5 min after the burst stimuluswere enhanced compared to the baseline period before thepriming stimulation, but returned to baseline 1 h afterstimulation. In the presence of 300 nM RO4938581, themean fEPSP slopes were also enhanced 1–5 min after thepriming and burst stimuli compared to the baseline period,but, in contrast to control experiments, did not return tobaseline 1 h after tetanic stimulation. Thus, RO4938581was able to induce early LTP from a subthreshold tetanicstimulation paradigm.

Receptor occupancy in the rat hippocampus

The degree of receptor occupancy produced by RO4938581was measured by evaluating its ability to block in vivo thebinding of the GABAA α5 receptor radioligand RO0154513(Hadingham et al. 1993; Lingford-Hughes et al. 2002;Maeda et al. 2003). Following administration of [3H]-RO0154513, the highest specific binding was observed inthe hippocampus (data not shown) in agreement with theknown distribution of the GABAA α5 receptor (Pirker et al.2000). Pretreatment of rats with RO4938581 dose-dependentlydecreased the binding of [3H]-RO0154513 in the hippo-campus. The degree of GABAA α5 receptor occupancyproduced by RO4938581 at doses of 0.1, 1, and 10 mg/kgp.o. was 30%, 74%, and 90%, respectively (Fig. 5).

Effect of RO4938581 on spontaneous locomotor activity,neurological assessment, and chain pull (VI20) test

RO4938581 had a significant effect on horizontal activity[F(33,208)=3.4, p<0.001], reaching significance at 10 mg/kg p.o. from 0–5 min (p<0.001), 5–10 min (p<0.05) and10–15 min (p<0.01; Fig. 6a). Both 0.1 and 1 mg/kgsignificantly (p<0.01) reduced horizontal activity onlyduring the 10- to 15-min time bin. RO4938581 did nothave a significant effect [F(3,28)=1.3, NS] on totalhorizontal activity during 1 h (Fig. 6b).

The purpose of the neurological assessment test is todetermine compound effects on neurological signs, motorperformance, and muscle strength. There was no effect ofRO4938581 on body posture, gait, reflexes, and otherneurological signs. RO4938581 had no effect on gripstrength [F(3,20)=0.48, NS], horizontal wire test (H=2.4,NS) and rotarod performance at 8 rpm (H=3.0, NS) and16 rpm (H=3.0, NS; Table 2).

Food-deprived Lister-hooded rats were previouslytrained to pull a ceiling chain for food reward on a VI20schedule. This variable time period leads to a high level ofresponding based on the unpredictability of food presenta-

-9 -8 -7 -6 -5

-50

-40

-30

-20

-10

0%

ch

ang

e o

f G

AB

A r

esp

on

se

log( [RO4938581] )

Fig. 2 Inhibition of the GABA response by RO4938581 in HEK293cells expressing the α1β2γ2 (empty square), α2β3γ2 (emptytriangle), α3β3γ2 (filled triangle), and α5β3γ2 (filled circle) subunitcombinations. Current amplitudes obtained from GABA applicationsin the presence of different concentrations of RO4938581 are plottedas a function of the drug concentration. Values are normalized to thecontrol responses obtained with GABA alone. Symbols represent themean±SEM of four to six cells. The sigmoidal curve for α5β3γ2 wasgenerated with the mean pIC50 and Hill slope


tion. Hence, the high motoric demand of the test makes itsensitive to disruption by drug effects on motivation and/ormotor function. RO4938581 had no effect on the number ofchain pulls [F(4,44)=2.4, p=0.06] and the number of foodrewards obtained [F(4,44)=1.7, NS] during the 30-minsession (Table 2).

RO4938581 attenuates scopolamine-induced workingmemory impairment in the DMTP task

DMTP is a test of spatial working memory, which isdependent on hippocampal function (Aggleton et al. 1992;

a. b.

-5 -40

20

40

60

80

100

% r

edu

ctio

n o

f P

S

log( [isoguvacine] )

2 mV 10 ms

Fig. 3 Effect of RO4938581 on the inhibition of population spike(PS) by isoguvacine recorded from hippocampal CA1 neurons. aSuperimposed PS obtained in the same hippocampal slice and evokedin the absence (straight line) and presence of isoguvacine (64 μM,dotted line) and in the presence of isoguvacine 64 μM with

RO4938581 (300 nM; dashed line). b Concentration–response curvesfor isoguvacine in inhibiting PS generated in the absence (opencircles) and presence (filled circles) of RO4938581 (300 nM). Thepoints represent the mean±SEM of four slices. The sigmoidal curvesare fits through these points

a b c d

0 20 40 60 80 100 120

100

150

200

EP

SP

slo

pe

(% b

asel

ine)

time (min)

a b

c

d

priming burst

�

Fig. 4 Time courses of field excitatory postsynaptic potentials(fEPSPs) generated in the absence (empty circles) and presence (filledcircles) of 300 nM RO4938581. fEPSPs were recorded from the apicaldendritic layer of the CA1 region after stimulation of the Schaffercollateral commissural pathway at 30-s interval. After a stable baselinerecording, a priming stimulus (ten events at 100 Hz) and a burststimulus (four times ten events at 100 Hz every 20 s) were applied tothe slices at time points indicated by the bars. This stimulationparadigm induced LTP only in slices incubated in RO4938581 20 minbefore the priming stimuli and until the end of the experiment. Thetraces on top are averages of four responses obtained at the timesindicated by the letters. Bars: 20 ms, 0.8 mV

Veh. 0.1 1 100

20

40

60

80

% O

ccu

pan

cy

100

RO4938581 (mg/kg, po)

Fig. 5 Occupancy of rat hippocampal α5-containing GABAA

receptors by RO4938581. Receptor occupancy was measured 60 minafter oral administration of RO4938581 at dose 0.1, 1, and 10 mg/kg.Receptors were labeled by [3H]-RO0154513 administered 10 minbefore sacrificing the animals. Bars represent mean±SD of occupancymeasured in three Lister Hooded rats


Hampson et al. 1999; Higgins et al. 2002). In the firstexperiment, the effect of RO4938581 administered alone onperformance of the DMTP task was assessed (Fig. 7 a,b).RO4938581 did not have an effect on percent correctresponses across delays [F(15,165)=1.0, NS; Fig. 1a] ortotal percent correct responses during the session [F(3,33)=0.6, NS; Fig. 1b]. There was a slight but significant increasein latency to make a correct response [F(3,33)=3.6, p<0.05] and magazine latency [F(3,33)=5.3, p<0.01] follow-ing 10 mg/kg (Table 3).

In the second experiment, the effect of RO4938581versus a scopolamine-induced impairment was assessed(Fig. 7c,d). Scopolamine significantly reduced percentcorrect responses in a delay-dependent manner (p<0.01 at16 and 24 s), indicating an impairment of working memoryin the DMTP task. RO4938581 significantly (p<0.01)reversed the scopolamine impairment at 0.3 and 1 mg/kgp.o. at both delay intervals to control levels (Fig. 1c). At0.1 mg/kg p.o., there was a partial, but significant (p<0.05)reversal at the 24-s delay only. Scopolamine significantly(p<0.001) reduced total percent correct responses (Fig. 1d),which was significantly reversed by RO4938581 at 0.1 (p<0.01), 0.3 (p<0.01), and 1 mg/kg (p<0.01).

RO4938581 attenuates diazepam-induced spatial learningimpairment in the Morris water maze

The Morris water maze is a test of spatial learning andmemory which is dependent on hippocampal function (Morris1984). Diazepam did not have an effect on swim speed[F(4,43)=1.5, NS], but significantly increased the latency(p<0.01) and path length (p<0.01) traveled to find thehidden platform position (data not shown). RO4938581

significantly reversed the diazepam-induced increase inlatency [F(4,43)=4.3, p<0.01] and path length [F(4,43)=6.0, p<0.001]. During the probe trial, vehicle-treated ratsspent significantly more time in the platform quadrantcompared to the left, right, and opposite quadrants [F(3,24)=9.6, p<0.001; Fig. 8a]. In contrast, diazepam-treated rats spent significantly less time in the platformquadrant (p<0.01; Fig. 8a) and had significantly fewerplatform crossings (p<0.01; Fig. 8b) compared to thevehicle group, indicating an impairment in spatial learn-ing. RO4938581 significantly attenuated the diazepam-induced impairment at 1 [F(3,27)=6.2, p<0.01], 3 [F(3,27)=3.9, p<0.05] and 10 mg/kg [F(3,27)=27.9, p<0.001]. Furthermore, the time spent in the platform

a. b.

0

500

1000

1500

2000

5 1 15 25 30 35 40 45 50 55 60

***

***

0

500

1000

1500

2000

5 10 20

Time (min)

Ho

rizo

nta

l act

ivit

y

***

***

0

1500

3000

4500

V 0 .1 1 10

Dose (mg/kg p.o.)

Fig. 6 The effect of RO4938581 on spontaneous locomotor activity overa 1-h period: a time course of horizontal activity in 5 min time bins:vehicle (empty circle); 0.1 (filled circle); 1.0 (filled triangle), 10 (empty

square) mg/kg; b total horizontal activity in 1 h. Data are presented asmean±SEM (n=8 Sprague–Dawley rats per dose group). Statistics: *p<0.05, **p<0.01, ***p<0.001 vs. vehicle

Table 2 Effect of RO4938581 on measures of neurological assess-ment and chain pull responding for food on a variable interval of 20 s(VI20)

Dose(mg/kg)

Grip strength(g)

Horizontalwire test (s)

Rotarod (s)16 rpm

Chain pull(VI20)

Counts Rewards

Veh 888±53 51 (41–60) 120 608±37 83±11 – – – 563±44 80±33 924±47 51 (34–60) 120 538±42 78±310 921±33 57 (42–60) 120 563±41 82±130 857±47 29 (27–53) 120 539±43 81±1

Chain pull data are presented as mean±SEM (n=12 Lister Hoodedrats). Neurological data are presented as: mean±SEM or as medianwith interquartile range in parentheses (n=6 Sprague–Dawley rats perdose group).


quadrant at 10 mg/kg (p<0.01; Fig. 8a) and the number ofplatform crossings at 1 (p<0.05), 3 (p<0.05), and 10 mg/kg (p<0.01; Fig. 8b) were significantly higher thandiazepam.

RO4938581 enhances executive function in the object-retrieval task in monkeys

The object-retrieval (detour-reaching) task is used to assessprefrontal executive function in monkeys (Rutten et al.2008). RO4938581 increased first reaches (% correct)

during difficult trials [F(5,55)=18.2, p<0.001], reachingsignificance at 3 and 10 mg/kg p.o. (Fig. 9). There were noeffects on easy trials (performance throughout the studywas 100% correct on this trial type).

RO4938581 does not induce an anxiogenic-like effectin the SAA and elevated plus maze

The SAA and EPM tests can be used to assess potentialanxiogenic-like effects of compounds (Nicolas and Prinssen2006). RO4938581 did not have an effect [F(3,36)=0.9,

a.Effect of RO4938581 administered alone on performance of the DMTP task

b. Effect of RO4938581 versus scopolamine-induced impairment of the DMTP task

V V 0.1 0.3 1

***

## ## ##

V V 0.1 0.3 1

V SCOP

***

## ## ##

40

50

60

70

80

90

100

0 8 16 24Delay (s)

% c

orr

ect

40

50

60

70

80

90

100

% c

orr

ect

70

80

90

100

V0 1

To

tal %

co

rrec

t

70

80

90

100

To

tal %

co

rrec

t

Dose (mg/kg)

0.1 10

***

**

##

##

###

##

***

**

##

##

###

##

0 8 16 24Delay (s)

Fig. 7 a The effect of RO4938581 administered alone, on short-termmemory function measured by choice accuracy across delay intervalin the delayed match to position (DMTP) task, at 0.1 (emptydiamond); 1.0 (empty triangle) and 10 (empty square) mg/kgcompared to vehicle (empty circle); n=12 Lister Hooded rats. b Theeffect of RO4938581 at 0.1 (filled diamond), 0.3 (filled triangle), and1.0 (filled square) mg/kg versus SCOP [scopolamine (filled circle)]-

induced working memory impairment [compared to vehicle (emptycircle)]; n=11 Lister Hooded rats. Data are presented as mean percentcorrect responses ±SEM. On the left hand side of the figure, the dataare presented by delay interval; on the right hand side, the data arecollapsed across delay for clarity. Statistics: **p<0.01, ***p<0.001:versus vehicle (V) treated group; #p<0.05, ##p<0.01: versusscopolamine-treated group


NS] on time spent in the hidden area of the SAA test(Fig. 10a). Moreover, RO4938581 did not have an effect[F(3,36)=0.3, NS] on time spent on the open arms of theelevated plus maze (Fig. 10b). The positive control,RO0194603, significantly increased the time spent in thehidden area in the SAA test (Fig. 10a) and significantlyreduced the time spent on the open arms of the elevatedplus maze (Fig. 10b), indicating an anxiogenic-likeprofile.

Lack of proconvulsant activity of RO4938581 followingan auditory stimulus

The proconvulsant potential of RO4938581 was investi-gated in naive 19- to 21-day-old DBA/2J mice (a strainthat exhibits age-dependent susceptibility to seizurefollowing an auditory stimulus). RO4938581 did nothave any proconvulsant activity up to 100 mg/kg inDBA2/J mice following an auditory stimulus (Table 4).

In contrast, the positive control, RO0194603, significantlyincreased the number of mice demonstrating tonic seizures(p<0.001).

Discussion

We have identified a new imidazo-triazolo-benzodiazepinecompound, RO4938581, that displays both binding selectiv-ity and inverse agonist functional selectivity for GABAA α5subunit-containing receptors. Since GABAA α5 receptors arepredominantly located in the hippocampus (Fritschy andMöhler 1995; Pirker et al. 2000), and there is evidence thatthese receptors play a role in hippocampal-dependentlearning and memory (Atack et al. 2006b; Collinson et al.2006; Crestani et al. 2002; Dawson et al. 2006), weevaluated the effect of RO4938581 in two rodent cognitiontests dependent on hippocampal function: the DMTP task(Aggleton et al. 1992; Hampson et al. 1999; Higgins et al.

Table 3 Effect of RO4938581 on measures of motivation and/or motor performance in the delayed match to position (DMTP) task

Dose (mg/kg) Number of trials Latency (s)

Correct Incorrect Missed Sample Correct Magazine

Vehicle 84±1 9±2 3±1 4.1±0.5 1.61±0.06 1.05±0.070.1 85±2 7±1 4±1 4.6±0.5 1.60±0.07 1.06±0.061 83±2 8±1 5±2 4.7±0.3 1.70±0.06 1.07±0.0610 82±2 9±1 5±2 5.0±0.6 1.75±0.09* 1.14±0.07**

Data are expressed as mean±SEM (n=12 Lister Hooded rats). Statistics: *p<0.05, **p<0.01 vs. vehicle group.

a. b.

0

1

2

3

4

5

6

V V 1 3 10V

##

##

##

V V 1 3 10

##

##

##

V DIAZEPAM

0

10

20

30

40

50

60

V V 1 3 10

++

+++ +++

+

## ##

0

10

20

30

40

50

60

% t

ime

in q

uad

ran

t

##

DIAZEPAM

No

. pla

tfo

rm c

ross

ing

s

Fig. 8 The effect of RO4938581 at 1, 3, and 10 mg/kg versusdiazepam-induced spatial learning impairment in the water mazeduring the probe trial: a percent time spent in left (dotted bar),platform (filled bar), right (empty bar) and opposite (diagonallystriped bar) quadrants; b number of platform crossings. Data are

presented as mean±SEM (n=10 Lister Hooded rats per dose group).Statistics (within treatment groups): +p<0.05, ++p<0.01, +++p<0.001 platform quadrant vs. left, right and opposite quadrants.Statistics (between treatment groups): #p<0.05, ##p<0.01 vs.diazepam-treated group


2002) and the Morris water maze (Morris 1984). We usedtwo different pharmacological approaches (cholinergic andGABAergic) to induce a cognitive impairment in each of thetasks, i.e., scopolamine and diazepam, respectively. In theDMTP task, RO4938581 reversed the scopolamine-inducedworking memory impairment, and in the Morris water maze,it reversed the diazepam-induced spatial learning impair-ment. In addition, RO4938581 also improved the perfor-mance of nonpharmacologically impaired adult monkeys inthe object retrieval task, which is a test requiring input fromthe prefrontal cortex (Diamond et al. 1989; Wilkinson et al.1997). RO4938581 did not show anxiogenic or proconvul-sant activity and did not induce adverse neurological effects,

or impair muscle strength, motor coordination, and respond-ing for food on a VI20 schedule.

Diazepam impairs spatial learning in naïve rats in theMorris water maze (Ballard et al. 2005), and it has beenproposed that BZD agonists may induce memory impair-ment by modulating hippocampal function (McNaughtonand Morris 1987). The current study used an abbreviatedversion of the standard acquisition protocol. Subjects werepretrained to the task, but had to learn a new spatialposition on the test day. Diazepam impaired the ability ofthe rats to find the spatial location of the hidden platform ata dose which did not affect swim speed. RO4938581 wasable to reverse the spatial learning impairment. This effectis not simply due to displacement of diazepam from theBZD binding site, since RO4938581 does not reverse otherbehavioral effects of diazepam, such as impairment ofrotarod performance (Ballard 2005). In contrast, com-pounds with no binding selectivity, fully reverse diaze-pam-induced rotarod impairment. Since RO4938581 hasbinding selectivity over the other subtypes, this indicatesthat reversal of diazepam impairment in the water maze ismediated via inverse agonism at the GABAA α5 receptor.

Nonselective GABAA inverse agonists attenuate scopol-amine-induced cognitive impairment in rats and humans(Duka et al. 1996; Kawasaki et al. 1996), In the presentstudy, using an α5 selective inverse agonist, we obtainedevidence for an interaction between GABAA α5 receptorsand the cholinergic system in a working memory test, theDMTP task. We speculate that reversal of the scopolamine-induced impairment could be due to an increase inhippocampal acetylcholine levels, since a nonselectiveGABAA partial inverse agonist, S-8510, has been shownto have this effect in vivo (Kawasaki et al. 1996). However,to verify whether GABAA α5 inverse agonists have an

25

50

75

100

V 0.3

Dose (mg/kg)

% c

orr

ect

**

25

50

75

100

0.1 3 10

**

1

Fig. 9 The effect of RO4938581 administered alone at 0.1, 0.3, 1, 3,and 10 mg/kg on percent correct (first reaches) during difficult trials ofthe object retrieval task in adult cynomolgus macaques. Data arepresented as mean±SEM (n=12). Statistics: **p<0.01 vs. vehicle (V)

a. b.**

0

150

300

450

V 0.3 1 3 V RO

**

0

150

300

450

V 0.3 1 3 V RO

**

0

150

300

450

V 0.3 1 3 V RODose (mg/kg) Dose (mg/kg)

Tim

e (s

) in

the

hid

den

are

a

0

20

40

60

V 0.3 1 3 V

**

0

20

40

60

V 0.3 1 3

**

0

20

40

60

V 0.3 1 3 V RO

Tim

e (s

) on

op

en a

rm

**

Fig. 10 The effect of RO4938581 administered at 0.3, 1, 3 mg/kg on:a time spent in the hidden zone in the social approach avoidance test(n=10 Fischer rats per dose group); and b time spent on the open armsof the elevated plus maze (n=10 Sprague–Dawley rats per dose

group). Data are presented as mean±SEM. Statistics: **p<0.01 vs.vehicle (V). RO is RO0194603 (3 mg/kg), which was tested as apositive control in a separate experiment (n=7–8 per dose group)


effect on acetylcholine release, such compounds should beassessed during microdialysis of the hippocampus in vivo.

In vivo binding experiments in the Lister hooded ratrevealed that doses of RO4938581 (0.1-1 mg/kg) whichreversed the cognitive impairment induced by scopolaminein the DMTP occupied approximately 30-70% of theGABAA α5 receptor subtype. These results are in agree-ment with previous studies where an α5 functionallyselective compound produced cognition enhancing effectsat a minimal effective dose that corresponded to a receptoroccupancy of approximately 25% (Dawson et al. 2006). Adose-range of 1-10 mg/kg was active in the Morris watermaze in Lister hooded rats, which corresponds to receptoroccupancies of 70-90%. This is higher than the value foundby Dawson et al. 2006, however, since all doses weresignificantly active, a lower dose-range should be tested infuture studies.

In rat hippocampal slices, RO4938581 modulatedGABAergic inhibition of CA1 pyramidal cells, as shownby the rightward shift of the isoguvacine concentration–response relation. The extent of the shift was less than thatcaused by the nonselective inverse agonists DMCM and β-CCM (Kemp et al. 1987; Knoflach, unpublished data). Asexpected for an inverse agonist acting at the BZD site of theGABAA receptor, RO4938581 had no direct effect on PSwhen applied alone to the slices. This suggests that α5 aswell as other GABAA receptor subtypes play a role for theinhibition of hippocampal CA1 neurons.

Hippocampally mediated cognitive processes involvelong-term changes in synaptic efficacy such as LTP.Nonselective GABAA inverse agonists like DMCM in-crease LTP (Seabrook et al. 1997), whereas nonselectiveGABAA agonists, e.g., diazepam impair LTP (del Cerro etal. 1992). In the present study, an α5 selective inverseagonist induced early LTP from a subthreshold tetanicstimulation paradigm in mouse hippocampal slices. Thisresult is in agreement with previous reports which showedenhancement of LTP by other α5 selective inverse agonists

(Atack et al. 2006b; Dawson et al. 2006). Therefore, weconclude that the α5 subtype plays a major role in themodulation of LTP as it was previously observed withnonselective compounds. GABAA α5-containing receptorsare predominantly localized extrasynaptically and mediatetonic inhibition (Glykys et al. 2008). A recent studysuggested that GABAA α5 receptors regulate the excitabil-ity of mouse hippocampal pyramidal neurons by influenc-ing the strength of depolarization required to generate anaction potential (Bonin et al. 2007). This way, it has beenexplained how extrasynaptic receptors can influence theexpression of LTP as it is shown in the present study.

Since nonselective inverse agonists are known to have apoor side effect profile (e.g., Dorow et al. 1983), wedetermined the effect of RO4938581 in a battery ofnoncognitive tests. RO4938581 had no effect on neurolog-ical signs, motor performance, and muscle strength.Moreover, the compound did not impair performance inthe chain pull (VI20) test, indicating no effect onmotivation or motor function. In the DMTP task,RO4938581 induced a slight increase in latency to pressthe correct lever and to collect the food reward at 10 mg/kg.The same dose decreased horizontal activity during the first15 min of a 1-h test session, but had no effect on swimspeed in the Morris water maze (in combination withdiazepam). Moreover, no sedative-like effects were ob-served in the monkeys during performance of the objectretrieval task. A similar effect on locomotor activity hasbeen reported by Savic et al. (2008) for the bindingselective partial inverse agonist PWZ-029 at doses slightlyhigher than the cognition-enhancing dose (10–20 mg/kg vs.5 mg/kg). This effect was shown to be a consequence ofinverse agonism at α5, since it was reversed by flumazeniland an α5 selective agonist, but not by an α1 selectiveantagonist (Savic et al. 2008). Since very few selectiveinverse agonists have been tested for locomotor activity, werefer to the detailed discussion of these effects by Savic etal. (2008) suggesting that this behavioral change requiresfurther investigation with compounds of different profiles.

We evaluated the anxiogenic potential of RO4938581 inthe SAA and EPM tasks. RO4938581 did not have aneffect on time spent in the hidden area of the SAA test anddid not have an effect on time spent on the open arms of theelevated plus maze indicating no anxiogenic-like profile. Incontrast, the positive control RO0194603, which is anonselective inverse agonist, exhibited an anxiogenic-likeprofile (Nicolas and Prinssen 2006).

RO4938581 did not have any proconvulsant activity upto 100 mg/kg p.o. in DBA2/J mice following an auditorystimulus. The audiogenic seizure model was used since adeficit in GABA-mediated inhibition in inferior colliculusneurons has been shown to be critical in this model and tobe sensitive to GABAA antagonists or inverse agonists

Table 4 Lack of pro-convulsant activity of RO4938581 following anauditory stimulus (70 dB, 14 kHz, 60 s) in 19–21 days DBA2/J mice

Dose (mg/kg) No. of mice

Vehicle 0/83 0/810 2/830 0/8100 0/8RO0194603 8/8***

Data are presented as number of mice with tonic convulsions, out of atotal group of 8. RO0194603 was used as a positive control within thesame experiment. Statistics: ***p<0.001 vs. vehicle.


(Faingold 2002). Furthermore, the underlying mechanismshave been proposed to be similar to other models ofepilepsy (Ross Coleman 2000). In order to fully explorepotential proconvulsant effects, additional methods whichdetermine seizure threshold could be used. However, itshould be noted that RO4938581 does not have convulsantactivity up to 300 mg/kg p.o in rats (Ballard 2005). Thelack of an anxiogenic or proconvulsant effect is consistentwith the behavioral phenotype of α5−/− mice (Collinson etal. 2002) and observations with the functionally α5selective compound, α5IA (Dawson et al. 2006).

In vivo binding experiments showed that receptoroccupancy at 10 mg/kg in Lister Hooded rats wasapproximately 90%. The SAA and EPM tests were run inFischer and Sprague–Dawley rats, respectively, so one canonly assume that at the dose-range tested, high levels ofGABAA α5 receptor occupancy were achieved withRO4938581, i.e., more than 70% at the maximal dosetested. It is also assumed that full receptor occupancyoccurs at the maximum dose (100 mg/kg) tested in themouse audiogenic seizures test. These data further supportthe notion that the multiple effects of nonselective BZDcompounds are mediated through distinct GABAA receptorpopulations (Rudolph and Möhler 2004) and confirms thatsufficient selectivity is important for excluding side effectsassociated with inverse agonistic activity at the α1, α2, andα3 receptor subtypes. In addition, other studies demon-strated that high-binding selectivity is required to avoid sideeffects of BZD inverse agonists with little functionalselectivity (Atack et al. 2006a; Atack et al. 2006b). It isprobably safest to avoid effects on other GABAA receptorsubtypes by combining both binding and functionalselectivity at the α5 receptor. RO4938581 displays such aprofile; it binds with higher affinity to GABAA α5receptors and displays close to full inverse agonistic effectat GABAA α5 receptors (−41%) with no or only a veryweak effect at the other GABAA receptor subtypes (±10%at 3 μM concentration). RO4938581 demonstrated essen-tially the same profile against human GABAA receptorstransiently expressed in oocytes and has negligible activity(<50% activity at 10 μM) in 71 receptor binding and 16enzyme assays (CEREP, Paris, France; data not shown).

The GABAA α5 inverse agonist, α5IA, enhancedperformance in the delayed-matching-to-position versionof the Morris water maze in unimpaired rats (Dawson et al.2006). In contrast, RO4938581 did not improve workingmemory of unimpaired rats in the DMTP task. However,the rats used for the latter experiment were adults pretrainedto asymptotic performance, and so, it would not be possibleto detect a cognitive improvement in this particular test. Inthe object retrieval task, young adult monkeys wereexposed to the task infrequently to prevent asymptoticperformance, thus allowing a window for improvement.

The positive effect of RO4938581 suggests that α5selective inverse agonists can enhance cognition in theabsence of pharmacological impairment.

The object retrieval (or detour reaching task) is a task ofprefrontal cognition in monkeys which involves attention,response inhibition, and planning, i.e., executive function.Lesions of the frontal cortex, but not of the hippocampus,impaired performance of this task (Diamond et al. 1989;Wilkinson et al. 1997). As mentioned above, there isevidence for improvement in a hippocampal-dependentmemory task following administration of a GABAA α5selective inverse agonist (Dawson et al. 2006). However, sofar, there has been little evidence for the involvement ofGABAA α5 receptors in prefrontal cortex-dependentexecutive function. GABAA α5 receptors have beenshown to be located in the cortex in rat, monkey, andman, but to a lesser degree than in the hippocampus(Akbarian et al., 1995; Fritschy and Möhler 1995; Maedaet al. 2003). The current results illustrate the importantrole of this receptor subtype on prefrontal cortex-mediatedexecutive function.

We have shown that the dual binding and functionalselectivity of RO4938581 offers a unique profile forenhancing cognition without side effects associated withother GABAA receptor subtypes. Importantly, it wasrecently reported in a clinical study that a GABAA α5functionally selective inverse agonist reduced the amnesiceffect of alcohol (Nutt et al. 2007). In the hippocampus ofAlzheimer’s disease patients, the density and pharmacologyof the GABAA α5 subunit-containing receptors is preserved(Howell et al. 2000). Therefore, selective GABAA α5inverse agonists have the potential to improve cognition inpatients suffering from Alzheimer’s disease and other typesof dementias.

Acknowledgements We would like to acknowledge the experttechnical assistance of Sebastian Debilly, Michel Enderlin, PatriciaGlaentzlin, Jasmin Graehler, Cecile Guizani, Rachel Haab, MarieHaman, Marie-Laurence Harle-Yge, Maria Karg, Yeter Kolb, RolandMory, Marie Claire Pflimlin, Pascal Pflimlin, Stefanie Saenger,Severine Weil-Bandinelli, Michael Weber, Roger Wyler. We wouldlike to thank Dr. Joachim Manns for formulation support.

References

Aggleton JP, Keith AB, Rawlins JN, Hunt PR, Sahgal A (1992)Removal of the hippocampus and transection of the fornixproduce comparable deficits on delayed non-matching to positionby rats. Behav Brain Res 52:61–71

Akbarian S, Huntsman MM, Kim JJ, Tafazzoli A, Potkin SG, BunneyWE Jr, Jones EG (1995) GABAA receptor subunit geneexpression in human prefrontal cortex: comparison of schizo-phrenics and controls. Cereb Cortex 5(6):550–560

Atack JR, Bayley PJ, Fletcher SR, McKernan RM, Wafford KA,Dawson GR (2006a) The proconvulsant effects of the GABAA


α5 subtype-selective compound RY-080 may not be α5-mediated. Eur J Pharmacol 548:77–82

Atack JR, Bayley PJ, Seabrook GR, Wafford KA, McKernan RM,Dawson GR (2006b) L-655,708 enhances cognition in rats but isnot proconvulsant at a dose selective for α5-containing GABAA

receptors. Neuropharmacology 51:1023–1029Ballard TM, Woolley ML, Prinssen E, Huwyler J, Porter R, Spooren W

(2005) The effect of the mGlu5 receptor antagonist MPEP inrodent tests of anxiety and cognition: a comparison. Psychophar-macology (Berl) 179:218–229

Birks J, Harvey RJ (2006) Donepezil for dementia due to Alzheimer’sdisease. Cochrane Database Syst Rev:CD001190

Bonin RP, Martin LJ, Macdonald JF, Orser BA (2007) α5 GABAA

receptors regulate the intrinsic excitability of mouse hippocampalpyramidal neurons. J Neurophysiol

Cheng Y, Prusoff WH (1973) Relationship between the inhibitionconstant (K1) and the concentration of inhibitor which causes 50per cent inhibition (I50) of an enzymatic reaction. BiochemPharmacol 22:3099–3108

Collinson N, Atack JR, Laughton P, Dawson GR, Stephens DN (2006)An inverse agonist selective for α5 subunit-containing GABAA

receptors improves encoding and recall but not consolidation inthe Morris water maze. Psychopharmacology (Berl) 188:619–628

Collinson N, Kuenzi FM, Jarolimek W, Maubach KA, Cothliff R, Sur C,Smith A, Otu FM, Howell O, Atack JR, McKernan RM, SeabrookGR, Dawson GR, Whiting PJ, Rosahl TW (2002) Enhancedlearning and memory and altered GABAergic synaptic transmissionin mice lacking the α5 subunit of the GABAA receptor. J Neurosci22:5572–5580

Crestani F, Keist R, Fritschy JM, Benke D, Vogt K, Prut L, Blüthmann H,Möhler H, Rudolph U (2002) Trace fear conditioning involveshippocampal α5 GABAA receptors. Proc Natl Acad Sci U S A99:8980–8985

Dawson GR, Maubach KA, Collinson N, Cobain M, Everitt BJ,MacLeod AM, Choudhury HI, McDonald LM, Pillai G,Rycroft W, Smith AJ, Sternfeld F, Tattersall FD, Wafford KA,Reynolds DS, Seabrook GR, Atack JR (2006) An inverse agonistselective for α5 subunit-containing GABAA receptors enhancescognition. J Pharmacol Exp Ther 316:1335–1345

del Cerro S, Jung M, Lynch G (1992) Benzodiazepines block long-term potentiation in slices of hippocampus and piriform cortex.Neuroscience 49:1–6

Diamond A, Zola-Morgan S, Squire LR (1989) Successful perfor-mance by monkeys with lesions of the hippocampal formation onAB and object retrieval, two tasks that mark developmentalchanges in human infants. Behav Neurosci 103:526–537

Dorow R, Horowski R, Paschelke G, Amin M (1983) Severe anxietyinduced by FG 7142, a beta-carboline ligand for benzodiazepinereceptors. Lancet 2:98–99

Duka T, Ott H, Rohloff A, Voet B (1996) The effects of abenzodiazepine receptor antagonist beta-carboline ZK-93426 onscopolamine-induced impairment on attention, memory andpsychomotor skills. Psychopharmacology (Berl) 123:361–373

Faingold CL (2002) Role of GABA abnormalities in the inferiorcolliculus pathophysiology—audiogenic seizures. Hear Res 168(1-2):223–237

Fritschy JM, Möhler H (1995) GABAA-receptor heterogeneity in theadult rat brain: differential regional and cellular distribution ofseven major subunits. J Comp Neurol 359:154–194

Glykys J, Mann EO, Mody I (2008) Which GABA(A) receptorsubunits are necessary for tonic inhibition in the hippocampus. JNeurosci 28:1421–1426

Hadingham KL, Wingrove P, Le Bourdelles B, Palmer KJ, Ragan CI,Whiting PJ (1993) Cloning of cDNA sequences encoding humanα2 and α3 GABAA receptor subunits and characterization of thebenzodiazepine pharmacology of recombinant α1-, α2-, α3-, and

α5-containing human GABAA receptors. Mol Pharmacol43:970–975

Haefely WE (1989) Pharmacology of the benzodiazepine receptor. EurArch Psychiatry Neurol Sci 238:294–301

Hampson RE, Jarrard LE, Deadwyler SA (1999) Effects of ibotenatehippocampal and extrahippocampal destruction on delayed-match and -nonmatch-to-sample behavior in rats. J Neurosci19:1492–1507

Higgins GA, Grottick AJ, Ballard TM, Richards JG, Messer J,Takeshima H, Pauly-Evers M, Jenck F, Adam G, Wichmann J(2001) Influence of the selective ORL1 receptor agonist, Ro64-6198, on rodent neurological function. Neuropharmacology41:97–107

Higgins GA, Enderlin M, Fimbel R, HamanM, Grottick AJ, SorianoM,Richards JG, Kemp JA, Gill R (2002) Donepezil reverses amnemonic deficit produced by scopolamine but not by perforantpath lesion or transient cerebral ischaemia. Eur J Neurosci15:1827–1840

Higgins GA, Ballard TM, Kew JN, Richards JG, Kemp JA, Adam G,Woltering T, Nakanishi S, Mutel V (2004) Pharmacologicalmanipulation of mGlu2 receptors influences cognitive perfor-mance in the rodent. Neuropharmacology 46:907–917

Howell O, Atack JR, Dewar D, McKernan RM, Sur C (2000) Densityand pharmacology of α5 subunit-containing GABAA receptorsare preserved in hippocampus of Alzheimer’s disease patients.Neuroscience 98:669–675

Jensen LH, Stephens DN, Sarter M, Petersen EN (1987) Bidirectionaleffects of beta-carbolines and benzodiazepines on cognitiveprocesses. Brain Res Bull 19:359–364

Kawasaki K, Eigyo M, Ikeda M, Kihara T, Koike K, Matsushita A,Murata S, Shiomi T, Takada S, Yasui M (1996) A novelbenzodiazepine inverse agonist, S-8510, as a cognitive enhancer.Prog Neuropsychopharmacol Biol Psychiatry 20:1413–1425

Kemp JA, Marshall GR, Wong EH, Woodruff GN (1987) Theaffinities, potencies and efficacies of some benzodiazepine-receptor agonists, antagonists and inverse-agonists at rat hippo-campal GABAA-receptors. Br J Pharmacol 91:601–608

Lingford-Hughes A, Hume SP, Feeney A, Hirani E, Osman S,Cunningham VJ, Pike VW, Brooks DJ, Nutt DJ (2002) Imagingthe GABA-benzodiazepine receptor subtype containing the α5-subunit in vivo with [11C]Ro15 4513 positron emissiontomography. J Cereb Blood Flow Metab 22:878–889

Little HJ, Nutt DJ, Taylor SC (1984) Acute and chronic effects of thebenzodiazepine receptor ligand FG 7142: proconvulsant proper-ties and kindling. Br J Pharmacol 83:951–958

Liu R, Hu RJ, Zhang P, Skolnick P, Cook JM (1996) Synthesis andpharmacological properties of novel 8-substituted imidazobenzo-diazepines: high-affinity, selective probes for α5-containingGABAA receptors. J Med Chem 39:1928–1934

Löw K, Crestani F, Keist R, Benke D, Brünig I, Benson JA, Fritschy JM,Rülicke T, Bluethmann H, Möhler H, Rudolph U (2000) Molecularand neuronal substrate for the selective attenuation of anxiety.Science 290:131–134

Maeda J, Suhara T, Kawabe K, Okauchi T, Obayashi S, Hojo J,Suzuki K (2003) Visualization of α5 subunit of GABAA/benzodiazepine receptor by 11C Ro15-4513 using positronemission tomography. Synapse 47:200–208

Martin JR, Ballard TM, Higgins GA (2002) Influence of the 5-HT2Creceptor antagonist, SB-242084, in tests of anxiety. PharmacolBiochem Behav 71:615–625

Maubach K (2003) GABAA receptor subtype selective cognitionenhancers. Curr Drug Targets CNS Neurol Disord 2:233–239

McKernan RM, Rosahl TW, Reynolds DS, Sur C, Wafford KA,Atack JR, Farrar S, Myers J, Cook G, Ferris P, Garrett L,Bristow L, Marshall G, Macaulay A, Brown N, Howell O,Moore KW, Carling RW, Street LJ, Castro JL, Ragan CI,


Dawson GR, Whiting PJ (2000) Sedative but not anxiolyticproperties of benzodiazepines are mediated by the GABAA

receptor α1 subtype. Nat Neurosci 3:587–592McNamara RK, Skelton RW (1993) Benzodiazepine receptor antag-

onists flumazenil and CGS 8216 and inverseagonist beta-CCMenhance spatial learning in the rat: dissociation from anxiogenicactions. Psychobiology 21:101–108

McNaughton N, Morris RG (1987) Chlordiazepoxide, an anxiolyticbenzodiazepine, impairs place navigation in rats. Behav BrainRes 24:39–46

McShane R, Areosa Sastre A, Minakaran N (2006) Memantine fordementia. Cochrane Database Syst Rev:CD003154

Möhler H (2006) GABAA receptor diversity and pharmacology. CellTissue Res 326:505–516

Morris R (1984) Developments of a water-maze procedure forstudying spatial learning in the rat. J Neurosci Methods 11:47–60

Nicolas LB, Prinssen EP (2006) Social approach-avoidance behaviorof a high-anxiety strain of rats: effects of benzodiazepine receptorligands. Psychopharmacology (Berl) 184:65–74

Nutt DJ, Besson M, Wilson SJ, Dawson GR, Lingford-Hughes AR(2007) Blockade of alcohol’s amnestic activity in humans by anα5 subtype benzodiazepine receptor inverse agonist. Neurophar-macology 53:810–820

Petersen EN, Jensen LH, Honore T, Braestrup C (1983) Differentialpharmacological effects of benzodiazepine receptor inverseagonists. Adv Biochem Psychopharmacol 38:57–64

Pirker S, Schwarzer C, Wieselthaler A, Sieghart W, Sperk G (2000)GABAA receptors: immunocytochemical distribution of 13subunits in the adult rat brain. Neuroscience 101:815–850

Quirk K, Blurton P, Fletcher S, Leeson P, Tang F, Mellilo D, Ragan CI,McKernan RM (1996) [3H]L-655,708, a novel ligand selective forthe benzodiazepine site of GABAA receptors which contain theα5 subunit. Neuropharmacology 35:1331–1335

Ross KC, Coleman JR (2000) Developmental and genetic audiogenicseizure models: behavior and biological substrates. NeurosciBiobehav Rev 24(6):639–653

Rudolph U, Möhler H (2004) Analysis of GABAA receptor functionand dissection of the pharmacology of benzodiazepines andgeneral anesthetics through mouse genetics. Annu Rev Pharma-col Toxicol 44:475–498

Rudolph U, Crestani F, Benke D, Brünig I, Benson JA, Fritschy JM,Martin JR, Bluethmann H, Möhler H (1999) Benzodiazepineactions mediated by specific GABAA receptor subtypes. Nature401:796–800

Rutten K, Basile JL, Prickaerts J, Blokland A, Vivian JA (2008)Selective PDE inhibitors rolipram and sildenafil improve objectretrieval performance in adult cynomolgus macaques. Psycho-pharmacology (Berl) 196:643–648

Savic MM, Clayton T, Furtmuller R, Gavrilovic I, Samardzic J, Savic S,Huck S, Sieghart W, Cook JM (2008) PWZ-029, a compound withmoderate inverse agonist functional selectivity at GABAA

receptors containing α5 subunits, improves passive, but notactive, avoidance learning in rats. Brain Res 1208:150–159

Seabrook GR, Easter A, Dawson GR, Bowery BJ (1997) Modulationof long-term potentiation in CA1 region of mouse hippocampalbrain slices by GABAA receptor benzodiazepine site ligands.Neuropharmacology 36:823–830

Sternfeld F, Carling RW, Jelley RA, Ladduwahetty T, Merchant KJ,Moore KW, Reeve AJ, Street LJ, O’Connor D, Sohal B, Atack JR,Cook S, Seabrook G, Wafford K, Tattersall FD, Collinson N,Dawson GR, Castro JL, MacLeod AM (2004) Selective, orallyactive GABAA α5 receptor inverse agonists as cognitionenhancers. J Med Chem 47:2176–2179

Venault P, Chapouthier G, de Carvalho LP, Simiand J, Morre M,Dodd RH, Rossier J (1986) Benzodiazepine impairs and beta-carboline enhances performance in learning and memorytasks. Nature 321:864–866

Wilkinson LS, Dias R, Thomas KL, Augood SJ, Everitt BJ, Robbins TW,Roberts AC (1997) Contrasting effects of excitotoxic lesions of theprefrontal cortex on the behavioural response to D-amphetamine andpresynaptic and postsynaptic measures of striatal dopamine functionin monkeys. Neuroscience 80:717–730


RO4938581, a novel cognitive enhancer acting at GABAA α5 subunit-containing receptors

Documents

Transcript of RO4938581, a novel cognitive enhancer acting at GABAA α5 subunit-containing receptors