European Journal of Pharmacology 740 (2014) 570–577

Contents lists available at ScienceDirect

European Journal of Pharmacology journal homepage: www.elsevier.com/locate/ejphar

Molecular and cellular pharmacology

Positive allosteric modulation of the GHB high-affinity binding site by the GABAA receptor modulator monastrol and the flavonoid catechin Laura F. Eghorn a, Kirsten Hoestgaard-Jensen a, Kenneth T. Kongstad a, Tina Bay a, David Higgins b, Bente Frølund a, Petrine Wellendorph a,n a Department of Drug Design and Pharmacology, Faculty of Health and Medical Sciences, University of Copenhagen, Fruebjergvej 3, 2100 Copenhagen, Denmark b Department of Histopathology, University of Dublin, Trinity College, Dublin, Ireland

art ic l e i nf o

a b s t r a c t

Article history: Received 2 January 2014 Received in revised form 16 June 2014 Accepted 18 June 2014 Available online 25 June 2014

γ-Hydroxybutyric

Keywords: γ-Hydroxybutyric acid GHB receptor NCS-382 Probe dependence Flavonoid Dihydropyrimidinone

acid (GHB) is a metabolite of γ-aminobutyric acid (GABA) and a proposed neurotransmitter in the mammalian brain. We recently identified α4βδ GABAA receptors as possible highaffinity GHB targets. GABAA receptors are highly sensitive to allosteric modulation. Thus to investigate whether GHB high-affinity binding sites are also sensitive to allosteric modulation, we screened both known GABAA receptor ligands and a library of natural compounds in the rat cortical membrane GHB specific high-affinity [3H]NCS-382 binding assay. Two hits were identified: Monastrol, a positive allosteric modulator of GABA function at δ-containing GABAA receptors, and the naturally occurring flavonoid catechin. These compounds increased [3H]NCS-382 binding to 185–272% in high micromolar concentrations. Monastrol and ( þ )-catechin significantly reduced [3H]NCS-382 dissociation rates and induced conformational changes in the binding site, demonstrating a positive allosteric modulation of radioligand binding. Surprisingly, binding of [3H]GHB and the GHB high-affinity site-specific radioligands [125I]BnOPh-GHB and [3H]HOCPCA was either decreased or only weakly increased, indicating that the observed modulation was critically probe-dependent. Both monastrol and ( þ)-catechin were agonists at recombinant α4β3δ receptors expressed in Xenopus laevis oocytes. When monastrol and GHB were co-applied no changes were seen compared to the individual responses. In summary, we have identified the compounds monastrol and catechin as the first allosteric modulators of GHB high-affinity binding sites. Despite their relatively weak affinity, these compounds may aid in further characterization of the GHB high-affinity sites that are likely to represent certain GABAA receptors. & 2014 Elsevier B.V. All rights reserved.

1. Introduction

γ-Hydroxybutyric acid (GHB) is a metabolite of γ-aminobutyric acid (GABA) and present in micromolar concentrations in the mammalian brain (Maitre, 1997). GHB is a proposed neurotransmitter, but its physiological role is yet unknown (Bay et al., 2014; Bernasconi et al., 1999; Crunelli et al., 2006) This versatile compound is a registered drug for the treatment of narcolepsy (Robinson and Keating, 2007) and alcoholism (Keating, 2014) as well as an illicit recreational drug known as Fantasy or liquid ecstacy, also used as a ‘date-rape drug’ in cases of drug-facilitated sexual assault (Drasbek et al., 2006). GHB has been speculated to Abbreviations: [125I]BnOPh-GHB, [125I]4-hydroxy-4-[4-(2-iodobenzyloxy)phenyl] butyric acid; GHB, γ-hydroxybutyric acid, [3H]HOCPCA, [3H]3-hydroxycyclopent-1enecarboxylic acid; [3H]NCS-382, [3H](E,RS)-(6,7,8,9-tetrahydro-5-hydroxy-5Hbenzocyclohept-6-ylidene)acetic acid n Corresponding author. Tel.: þ 45 3917 9811; fax: þ45 3533 6041. E-mail address: [email protected] (P. Wellendorph). http://dx.doi.org/10.1016/j.ejphar.2014.06.028 0014-2999/& 2014 Elsevier B.V. All rights reserved.

have therapeutic potential in several neurological disorders associated with sleep abnormalities (Bosch et al., 2012; Kantrowitz et al., 2009; Ondo et al., 2008), and clarification of its physiological mechanisms is of broad scientific interest. When administered exogenously, GHB readily enters the CNS and induces sedation, hypothermia and motor in-coordination at millimolar concentrations (Carter et al., 2009). These effects are mediated via metabotropic GABAB receptors and are completely absent in GABAB1 knockout mice (Kaupmann et al., 2003). Besides the low-affinity GABAB receptor binding sites, a distinct population of high-affinity binding sites is preserved in GABAB1 knockout mice (Bay et al., 2014; Benavides et al., 1982; Kaupmann et al., 2003). These elusive high-affinity binding sites can be selectively probed with the radioligands [3H](E,RS)-(6,7,8,9-tetrahydro-5hydroxy-5H-benzocyclohept-6-ylidene)acetic acid ([3H]NCS-382) (Mehta et al., 2001), [125I]4-hydroxy-4-[4-(2-iodobenzyloxy)phenyl]butyric acid ([125I]BnOPh-GHB) (Wellendorph et al., 2010) and [3H]3-hydroxycyclopent-1-enecarboxylic acid ([3H]HOCPCA) (Vogensen et al., 2013) (Fig. 1).

L.F. Eghorn et al. / European Journal of Pharmacology 740 (2014) 570–577

571 OH

HO

O

3

H

OH

O

GHB

H

[3H]NCS-382

125

O

OH

O

(+)-Catechin

N H

OH

OH

O

(-)-Catechin

OH

HO

OH

O

(+)-Taxifolin

OH

O

OH

OH OH

HO

OH HO

S

Monastrol

[125I]BnOPh-GHB OH

OH OH

OH

[3H]HOCPCA

HO

NH

O

I

OH

OH HO

O

O

3

HO

OH

HO

O

O

OH OH

O

Morin

Fig. 1. Structures of ligands and radioligands used in the present study.

We recently reported that GHB is an agonist at recombinant ionotropic α4βδ GABAA receptors and showed that up to 40% of high-affinity binding sites correspond to α4-containing GABAA receptors (Absalom et al., 2012). At present, the identity of the remaining 60% of high-affinity binding sites remains elusive (Bay et al., 2014). So far, a direct GHB response at native α4βδ receptors remains to be shown (Connelly et al., 2013). GABAA receptors are ligand-gated chloride ion channels, and a large number of pentameric subtypes can be formed from the 19 known subunits [α(1-6), β (1-3), γ(1-3), δ, ε, θ, π and ρ(1-3)] (Whiting et al., 1999). The δcontaining GABAA receptors are located extrasynaptically and mediate tonic inhibition in response to low GABA concentrations generated by spillover from the synaptic cleft (Farrant and Nusser, 2005). GABAA receptors contain a variety of allosteric sites for both endogenous ligands such as neurosteroids (Belelli and Lambert, 2005), for clinically important drugs including benzodiazepines and barbiturates (Sieghart, 1992) and several distinct sites for naturally occurring compounds such as flavonoids (Hanrahan et al., 2011; Karim et al., 2011). Allosteric ligands act by inducing conformational changes that affect the orthosteric binding site and hence the receptor–ligand interaction (Christopoulos, 2002; Sigel and Buhr, 1997). Most GABAA receptor modulators non-selectively affect a variety of subtypes (Sieghart, 1992), but the synthetic compounds DS1, DS2 (Wafford et al., 2009) and AA29504 (Hoestgaard-Jensen et al., 2010) as well as the dihydropyrimidinones monastrol and JM-II-43A (Lewis et al., 2010) were shown to be relatively selective for δ-containing GABAA receptors. Relevant structures used in the present study are shown in Fig. 1. Based on our previous finding that GHB high-affinity binding sites in part correspond to α4δ-containing GABAA receptors, we hypothesized that allosteric modulators of GABAA receptors (α4βδ receptors in particular) that induce conformational changes to increase GABA affinity or function may also affect the interaction of GHB. The aim of the present study was therefore to identify modulators of highaffinity GHB binding and function from a selection of known GABAA receptor modulators and a library of 160 natural compounds, which contained mainly alkaloids but also several flavonoids.

2. Materials and methods 2.1. Chemical compounds GHB, GABA, ( þ)- and (  )-catechin and all tested GABAA modulators were purchased from Sigma-Aldrich (St. Louis, MO, USA).

Racemic monastrol (1,2,3,4-tetrahydro-4-(3-hydroxyphenyl)-6methyl-2-thioxo-5-pyrimidinecarboxylic acid ethyl ester) was purchased from Tocris BioScience (Bristol, UK) or synthesized using the microwave-assisted Biginelli multicomponent reaction by cyclocondensation of 3-hydroxybenzaldehyde, ethyl acetoacetate, thiourea and ytterbium(III) trifluoromethanesulfonate (Yb(OTf)3) as a catalyst as described earlier (Dallinger and Kappe, 2007). [3H]NCS-382 was purchased from Biotrend (Köln, Germany), [3H]flunitrazepam and [3H]muscimol from PerkinElmer (Waltham, MA, USA) and [3H]GHB from American Radiolabeled Chemicals (Saint Louis, MO, USA). [3H]HOCPCA and [125I]BnOPhGHB were synthesized and radiolabeled as previously described (Sabbatini et al., 2010; Vogensen et al., 2013). 2.2. Library of natural compounds A library of 160 natural compounds, mainly alkaloids and flavonoids, was created in-house (Table S1). All compounds were dissolved in dimethyl sulfoxide to 50 mM and diluted with incubation buffer to a test concentration of 50 mM. 2.3. Radioligand binding studies 2.3.1. Tissue and membrane preparation Cortical tissue was dissected from adult male Sprague–Dawley rats (approx. 250 g), and cortical membranes were prepared as previously described (Ransom and Stec, 1988). 2.3.2. Equilibrium binding assays All binding assays were performed in 96-well format modified from the original report (Mehta et al., 2001). Test compounds were dissolved in incubation buffer (A: 50 mM potassium phosphate buffer, pH 6.0 or B: 50 mM Tris–HCl buffer, pH 7.4) and incubated for 1 h at 0–4 1C with membranes and radioligand in a total volume of 200–250 ml (triplicates, quadruplicates for [3H]GHB assays). The following combinations of radioligand concentration, protein amount and incubation buffer were used: 16 nM [3H]NCS382, 35–50 mg protein/well, buffer A; 50 nM [3H]GHB, 50–70 mg protein/well, buffer A; 10 nM [3H]HOCPCA, 15–35 mg protein/well, buffer A; 0.2 nM [125I]BnOPh-GHB, 5–10 mg protein/well, buffer A; 5 nM [3H]flunitrazepam, 50–70 mg protein/well, tested in both buffer A and B; 5 nM [3H]muscimol, 70–90 mg protein/well, tested in both buffer A and B. Nonspecific binding was determined in the presence of 1 mM GHB ([3H]NCS-382, [125I]BnOPh-GHB, [3H]

572

L.F. Eghorn et al. / European Journal of Pharmacology 740 (2014) 570–577

HOCPCA), 1 mM NCS-382 ([3H]GHB), 1 mM GABA ([3H]muscimol) or 100 mM diazepam ([3H]flunitrazepam). After rapid filtration through GF/C unifilters, radioactivity was measured using a Packard TopCount NXT Microplate Scintillation Counter (PerkinElmer). All known GABAA receptor modulators were screened in the concentration 100 mM except for tracazolate, which was tested at 30 mM due to its solubility limit. Parts of these results were published previously (Absalom et al., 2012). 2.3.3. Dissociation kinetic assays Dissociation kinetic assays were performed in 96-well format with buffer A using rat cortical membranes incubated at 0–4 1C with 16 nM [3H]NCS-382 in the presence or absence of 400 mM monastrol or (þ)-catechin. Nonspecific binding was determined with 1 mM GHB. Dissociation was initiated at different time points by addition of a small volume of GHB to a final concentration of 1 mM. The reaction was terminated as described for equilibrium binding studies. 2.4. Electrophysiology 2.4.1. cDNA and cRNA preparation and injection The cDNA used for cRNA was in the vectors pcDNA3.1 (α4, δ) and pGEMHE (β3). The cDNA of human α4, β3 and δ was linearized with SmaI, NheI and StuI (Thermo Fisher, Waltham, MA, USA), respectively, and transcribed into cRNA using the T7 mMessage mMachine Transcription High Yield Capped RNA Transcription Kit (Ambion, Cambridgeshire, UK). Defollicated Xenopus laevis oocytes from adult female frogs (EcoCyte Bioscience, Germany) were injected with α4β3δ subunits in a ratio 10:1:10 (105 ng cRNA in total). Incorporation of the δ subunit was verified with the δ-selective modulator DS2 (Wafford et al., 2009). Injected oocytes were incubated at 18 1C in modified Barth's solution (88 mM NaCl/1 mM KCl/15 mM HEPES/2.4 mM NaHCO3/0.41 mM CaCl2/0.82 mM MgSO4/0.3 mM Ca(NO3)2/100 U/ml penicillin/ 100 μg/ml streptomycin, pH 7.5). 2.4.2. Two-electrode voltage clamp recordings 3–7 days after injection, two-electrode voltage clamp was performed at room temperature as previously described (HoestgaardJensen et al., 2013). The recording chamber was continuously perfused with Ringer's solution (115 mM NaCl/2.5 mM KCl/10 mM HEPES/1.8 mM CaCl2/0.1 mM MgCl2, pH 7.5). The electrodes were filled with 3 M KCl, and the oocytes were clamped at  70 to 80 mV by a GeneClamp 500B amplifier (Axon Instruments, Union City, CA). The concentration–response curve of monastrol was obtained by applying 6 different concentrations in the Ringer's solution. Monastrol was applied until saturation, and after returning to baseline, the next concentration was applied after a 3-min washout-out period. When testing the modulatory effect of monastrol, GHB and monastrol were first applied separately and then in combination. 2.5. Data analysis All data analysis was performed using GraphPad Prism 5.0 (GraphPad Software Inc, La Jolla, CA, USA). The binding data were fitted according to the Allosteric modulator titration equation based on the allosteric ternary complex model developed for G proteincoupled receptors (Christopoulos, 2002), but formally in agreement with the two-state allosteric model also used for ionotropic receptors (Leff, 1995). Y¼

½A ½A þ ðK A ð1 þ ½B=K B Þ=ð1 þ α½B=K B ÞÞ

where Y denotes the fractional specific binding, [A] denotes the concentration of [3H]NCS-382 (fixed as constant) and [B] the concentration of monastrol. KA and KB denote the equilibrium dissociation constants of [3H]NCS-382 and monastrol, respectively, and KA was fixed as constant (based on the previously published [3H]NCS-382 Kd value of 430 nM (Wellendorph et al., 2010)). α is the cooperativity factor reflecting the direction and magnitude of the modulation. An α value of 1.0 indicates no modulatory effect, α o1.0 indicates negative modulation and α 41.0 positive allosteric modulation. KB and α were fitted by non-linear regression. For the dissociation kinetic experiment, the data were fitted to an exponential two-phase decay equation Y ¼ Plateau þSpan1 x e  k1ðtÞ þ Span2 x e k2ðtÞ where Y denotes fractional specific binding, Span1 and Span2 denote the percentage of each dissociation phase, and k1 and k2 denote the corresponding rate constants. Plateau denotes the Y value at infinite time. An extra-sum-of-squares F-test showed that the two-phase decay model fitted the data significantly better than the simpler one-phase decay model. Data from the electrophysiological experiments were normalized to the maximum response of GABA at the individual oocyte. The monastrol concentration–response curve was fitted by nonlinear regression using the equation for sigmoidal concentration– response with variable slope. 2.6. Statistics All data were expressed as mean 7 S.E.M. Mean values were compared using the unpaired, two-tailed Student's t-test.

3. Results 3.1. Monastrol and catechin increased [3H]NCS-382 binding Since no high-throughput functional assay is available for testing GHB-specific effects, the previously reported [3H]NCS-382 rat cortical membrane binding assay was employed as an initial screening method to identify modulators (Høg et al., 2008; Wellendorph et al., 2009). Both a 160 natural compounds library (Table S1, Fig. S1) and a selection of known GABAA receptor modulators (Table S2) were tested. The GABAA modulator monastrol (100 mM) and the naturally occurring flavonoid stereoisomers (þ)-catechin and ( )-catechin (50 mM) increased [3H]NCS-382 binding to 4125% of control and were therefore selected for further investigation. Monastrol and the two catechin stereoisomers concentrationdependently increased [3H]NCS-382 binding to rat cortical membranes (Fig. 2A–B). Data were fitted to the allosteric modulator titration equation, and estimated values of pKB, α and maximal [3H]NCS-382 binding are listed in Table 1. The α value reflects the fold change in affinity of [3H]NCS-382 when both the allosteric and orthosteric ligand are bound to the receptor, thus α 41 indicates positive binding co-operativity of the radioligand. ( þ)-Catechin displayed a markedly stronger modulatory effect than its stereoisomer (  )-catechin, thus monastrol and ( þ)-catechin were selected for further characterization. Interestingly, other flavonoids tested in the library of natural compounds did not appear to enhance [3H]NCS-382 binding (Table S1, Fig. S1). Flavonoids with high structural similarity to ( þ)-catechin were further investigated, showing that [3H]NCS-382 binding was inhibited completely by morin (pKi 7S.E.M.¼ 4.26 70.02; n ¼3) and (þ)-taxifolin (pKi 7S.E.M.¼ 2.857 0.04; n ¼3) (Fig. 2A). Neither monastrol nor (þ )-catechin increased binding of the GABAA receptor radioligands [3H]muscimol and [3H]flunitrazepam

L.F. Eghorn et al. / European Journal of Pharmacology 740 (2014) 570–577

573

Fig. 2. Equilibrium binding of [3H]NCS-382 (16 nM) to rat cortical membranes in the presence of ( þ)-catechin, (  )-catechin, ( þ)-taxifolin, morin (A) and monastrol (B). Data are presented as mean percentage of specific binding 7 S.E.M (n¼3). Dissociation of [3H]NCS-382 (16 nM) in the absence and presence of ( þ )-catechin (n¼ 3) (C) and monastrol (n ¼6) (D). Data were fitted to a two-phase decay model due to biphasic dissociation (F-test, P o 0.0001).

Table 1 Allosteric model parameter estimates for ( þ)-catechin, (  )-catechin and monastrol. [3H]NCS-382 binding data in the presence of modulators were fitted according to the allosteric modulator titration equation, providing estimates of pKb (the negative logarithm to the dissociation constant for the allosteric modulator at the free receptor) and α (the cooperativity factor). Data are presented as means 7 S.E.M. (n¼ 3).

α Log α pKb Maximal binding (% of control)

( þ)-Catechin

(  )-Catechin

Monastrol

3.35 0.52 7 0.04 3.52 7 0.10 2727 16.5

2.31 0.36 7 0.06 3.32 7 0.13 185 7 14.1

2.00 0.317 0.05 3.96 7 0.16 1857 7.1

(data not shown), demonstrating that the observed modulation was specific for the GHB high-affinity binding site.

reduced [3H]GHB binding to 5479% of control. Also [3H]HOCPCA binding was significantly increased by (þ)-catechin (12273% of control), but to a much lower degree than [3H]NCS-382 and [3H] GHB binding (Fig. 3B). Similar to data obtained for [3H]NCS-382, the 4-ketone-substituted flavonoid, morin, completely inhibited [3H] HOCPCA radioligand binding (pIC50 7S.E.M.¼4.6270.07; n¼3), indicative of a competitive interaction. This inhibition was still complete and the IC50 unchanged (pIC50 7S.E.M.¼4.4670.06; n¼ 3) when the [3H]HOCPCA radioligand concentration was increased to 80 nM, a concentration much closer to the estimated Kd value (Vogensen et al., 2013) (Fig. S2). As observed for [3H]GHB, monastrol concentrationdependently decreased [3H]HOCPCA binding to 64.678% of control (1 mM monastrol) (Fig. 3B). [125I]BnOPh-GHB binding was significantly reduced by both monastrol (80.370.1% of control) and (þ)-catechin (89.270.6%) (Fig. 3C).

3.2. Monastrol and ( þ)-catechin reduced [3H]NCS-382 dissociation 3.4. Electrophysiology Dissociation kinetics were investigated to confirm that monastrol and (þ )-catechin were indeed allosteric modulators of [3H] NCS-382 binding. In the absence and presence of either ligand, dissociation of [3H]NCS-382 by GHB was fast and biphasic (P o0.0001; F test) (Fig. 2C–D), and data were therefore fitted to a two-phase decay model. ( þ)-Catechin (400 mM) significantly decreased the slow dissociation rate, whereas monastrol (400 mM) significantly decreased both the slow and fast dissociation rate constants (Table 2). Both compounds significantly reduced the percentage of the fast phase with  18–22% of control. 3.3. Effects of monastrol and (þ )-catechin on binding of other specific radioligands (þ)-Catechin significantly increased [3H]GHB binding to 163711% of control (Fig. 3A). In contrast, monastrol concentration-dependently

Functional effects of (þ)-catechin and monastrol were investigated at α4β3δ receptors expressed in X. laevis oocytes. This subtype was chosen, because it gives robust responses to GHB with less variation than α4β1δ receptors (Absalom et al., 2012). (þ)-Catechin (100 mM) displayed a marked direct agonist effect, inducing currents of 68 75% of the maximal GABA response (Fig. 4A), and co-application with GHB was therefore not tested. Also monastrol directly activated α4β3δ receptors with a pEC50 value of 4.4 70.05 and a maximal effect of 207 3% of the GABA-induced response (Fig. 4B–C). Monastrol did not induce any response in control oocytes (no RNA injected; n ¼3). Co-application of 1 mM GHB (EC50 at α4β3δ (Absalom et al., 2012)) and 100 mM monastrol did not induce significant alterations compared to the monastrol-induced response alone (P ¼0.120; Student's paired t-test, n ¼8) (Fig. 4D).

574

L.F. Eghorn et al. / European Journal of Pharmacology 740 (2014) 570–577

Table 2 Dissociation rate constants for [3H]NCS-382 at rat cortical membranes in the absence or presence of ( þ )-catechin and monastrol.

k (slow) (min  1) Half-life (slow) (min) k (fast) (min  1) Half-life (fast) (min) Span (fast) (%)

Control

( þ)-Catechin

Control

Monastrol

0.229 7 0.01 3.03 7 0.08 4.02 7 0.44 0.1777 0.02 75.0 7 1.7

0.100 70.01c 7.21 71.05a 3.46 70.69 0.214 70.04 52.9 72.0b

0.2007 0.02 3.617 0.31 3.87 7 0.12 0.1807 0.01 80.4 7 1.3

0.150 70.01a 4.65 70.20a 2.84 70.26b 0.254 70.02a 62.1 72.1c

Mean values were estimated by fitting data to a two-phase decay model and compared using Student's two-tailed, unpaired t-test. Data are expressed as means7 S.E.M. (n¼6 for monastrol, n¼ 3 for ( þ )-catechin) and significance levels compared to control. a b c

P o0.05. Po 0.01. Po 0.001.

Fig. 3. Equilibrium binding of 50 nM [3H]GHB (A), 10 nM [3H]HOCPCA (B) and 0.2 nM [125I]BnOPh-GHB (C) in the presence of ( þ)-catechin and monastrol. Competitive inhibition by GHB is shown for [3H]GHB and [125I]BnOPh-GHB.

4. Discussion The dihydropyrimidinone monastrol and the flavonoid (þ )-catechin display several characteristics demonstrating an allosteric modulation of the [3H]NCS-382 binding site.

Besides the well-established antioxidant and free-radicalscavenging properties of flavonoids (Hanrahan et al., 2011), catechin derivatives possess anxiolytic, sedative and anticonvulsant effects, which are believed to be mediated via GABAA receptors (Adachi et al., 2006; Campbell et al., 2004). Only few studies investigated the compound (þ)-catechin in isolation; one reporting non-competitive inhibition of GABA-induced currents at recombinant α1β1 GABAA receptors (Hossain et al., 2002), whereas Campbell et al. saw no effect of 1–100 mM (þ)-catechin on the GABA-elicited response in α1β2γ2L receptors (Campbell et al., 2004). The latter study may reflect that catechins are less likely to interact with the benzodiazepine site than other flavonoids due to their saturated C2–C3 bond, rendering the catechins chiral and less planar molecules (Campbell et al., 2004; Fernandez et al., 2008). The synthetic molecule monastrol was initially described as a kinesin inhibitor causing arrest of mitosis in mammalian cells (Mayer et al., 1999). Lewis et al. showed that monastrol potentiated GABA currents 3-fold in recombinant α1β2δ GABAA receptors expressed in HEK293T cells (Lewis et al., 2010). As the current literature describing the pharmacology of monastrol and (þ)-catechin at other δ-containing GABAA receptor subtypes is limited, it is difficult to make relevant comparisons to our data at this point. The observed positive modulation of [3H]NCS-382 binding to rat cortical membranes by monastrol and the catechin stereoisomers was relatively weak in terms of affinity (Fig. 2, Table 1). A characteristic feature of allosteric modulation is the alteration of dissociation kinetics of the orthosteric ligand (Christopoulos, 2002), and [3H]NCS-382 dissociation studies thus validate the allosteric nature of (þ )-catechin and monastrol (Fig. 2, Table 2). The biphasic dissociation of the radioligand indicates the presence of two populations of binding sites. [3H]NCS-382 was previously reported to display binding to at least two distinct classes of binding sites in the mammalian brain (Mehta et al., 2001). The increased span of the slow dissociation phase of [3H]NCS382 suggests that monastrol and ( þ)-catechin exert their modulatory actions by inducing a shift towards a higher-affinity receptor conformation more favorable for [3H]NCS-382 binding. A similar mechanism of action was previously reported for the allosteric GABAA modulator pentobarbital, which increased the number of high-affinity sites at the expense of low-affinity sites for [3H]GABA and [3H]muscimol at 0 1C in bovine and mouse brain membranes, respectively (Olsen et al., 1984; Yang and Olsen, 1987). The opposing effects of monastrol in the different GHB radioligand binding assays underscore how detection of allosteric phenomena critically depends on the choice of radioactive orthosteric probe, and the direction and magnitude of an allosteric interaction can be quite different between probes binding to the same site (Christopoulos, 2002; Valant et al., 2012). Since the intrinsic activities of HOCPCA and BnOPh-GHB are unknown, and that of NCS-382 seems rather ambiguous (Crunelli et al., 2006), no conclusions can be easily inferred. Theoretically, the ideal probe

L.F. Eghorn et al. / European Journal of Pharmacology 740 (2014) 570–577

575

Fig. 4. Representative current traces showing functional response of 100 mM ( þ )-catechin alone (consistent in 4 oocytes from 4 different injection batches) (A) and 100 mM monastrol alone and in combination with 1 mM GHB (consistent in 8 out of 10 oocytes from 4 different injection batches) (B) at recombinant α4β3δ GABAA receptors expressed in X. laevis oocytes. (C) Concentration–response curve of monastrol normalized to maximum GABA response (n¼ 4). (D) Summarized data displaying current responses induced by 1 mM GHB, 100 mM monastrol and co-application of 1 mM GHB and 100 mM monastrol. The response to co-application did not differ significantly from that induced by monastrol alone (n¼ 8, means 7S.E.M.). The difference in monastrol effect between (C) and (D) is due to general variation between injection batches.

for screening for allosteric modulators of GHB function would be the endogenous orthosteric agonist [3H]GHB, but [3H]NCS-382 was preferred due to its higher amount of specific binding, higher sensitivity and lower affinity for the GABAB receptor compared to [3H]GHB (Mehta et al., 2001). Evidently, in our hands [3H]NCS-382 appears to be a more sensitive radioligand for detecting conformational changes involving GHB high-affinity binding sites. However, some allosteric interactions may not have been detected with [3H] NCS-382 in the present study. For example, an earlier study found that pentobarbital and diazepam significantly increased [3H]GHB binding (Snead and Nichols, 1987), whereas these compounds did not affect [3H]NCS-382 binding in the present study. The observed probe dependence reflects that each radioligand adopts different molecular orientations in the binding pocket, or prefers certain states of the receptor (Leff, 1995), and therefore is differently affected by the conformational changes induced by (þ )-catechin and monastrol. This oppositely directed effects of monastrol and (þ )-catechin could suggest that they interact with distinct allosteric binding sites. The effect of (þ )-catechin was unique among the tested flavonoids, indicating very specific interactions at the allosteric (þ )-catechin binding site. The difference in maximal effect of (þ )and ( )-catechin indicates that stereocenter configurations play a role, and the inhibitory effect of ( þ)-taxifolin and morin on [3H] NCS-382 binding suggests that presence of the 4-ketone group is detrimental for the positive modulation of binding. Since the 4-ketone substituted compounds completely inhibited both [3H] NCS-382 and [3H]HOCPCA binding and this ability was clearly unchanged after increasing the radioligand concentration of [3H] HOCPCA to values closer to the Kd value (Vogensen et al., 2013),

morin most likely interacts with the orthosteric binding site in a competitive manner. We earlier proposed that the GHB high-affinity binding site in α4βδ receptors is overlapping, but non-identical with the binding site of GABA in the β þ α  interface (Absalom et al., 2012). In addition to αβδ it is possible that other GABAA subtype interfaces also contribute to the large density of GHB high-affinity binding sites, and an interaction of monastrol at a non-GABAA receptor binding site cannot be excluded (Bay et al., 2014). Most likely, the monastrol-induced increase in [3H]NCS-382 binding is an average effect of modulation at several different GABAA receptor subtypes in native tissue, possibly displaying positive as well as negative or neutral modulation. Indeed the close monastrol analog JM-II-43A was reported to potentiate GABA-induced peak currents in both α4β2 and several αβδ subtypes (Lewis et al., 2010). An allosteric modulator may display different affinity and degree of modulation at different GABAA receptor subtypes (Karim et al., 2011), and our observations in α4β3δ receptors can therefore not necessarily be generalized to other GABAA receptors. A systematic investigation of different subtypes is relevant to fully characterize the pharmacological effect of monastrol. Direct activation of GABAA receptors by monastrol was not previously observed in HEK293T cells expressing recombinant α1β2δ GABAA receptors (Lewis et al., 2010). This difference may simply be due to the fact that we used α4β3δ receptors or may relate to differences in receptor expression level (Noetzel et al., 2012) or stoichiometry. Also the marked direct activation of α4β3δ receptors by ( þ)-catechin is interesting, since only few reports of flavonoid-induced allosteric agonism exist (Hanrahan et al., 2011; Karim et al., 2011). A full characterization of catechin on a number of subtype combinations

576

L.F. Eghorn et al. / European Journal of Pharmacology 740 (2014) 570–577

would be interesting. The observed direct agonist effects complicated our electrophysiological experiments. However, the absence of modulation when GHB and monastrol were co-applied could indicate differences in the receptor populations studied in binding and functional assays. Despite their relatively low affinity and non-specific effects, monastrol and (þ )-catechin are considered potential tool compounds or lead compound structures for investigating the molecular pharmacology of the GABAA receptor GHB high-affinity site in recombinant systems. Although the antimitotic effect of monastrol (Mayer et al., 1999) is retained by many monastrol analogs (Kamal et al., 2010), this effect is stereospecific and largely confined to the (S)-enantiomer (Maliga et al., 2002). Thus, (R)monastrol could be an interesting lead structure (Rashid et al., 2013) for development of analogs showing higher affinity in the [3H]NCS-382 binding assay. Also (þ)-catechin might be a useful lead structure, since there are several examples of flavonoids displaying high GABAA receptor subtype selectivity (Hanrahan et al., 2011; Karim et al., 2011, 2012). In conclusion, we have identified the first allosteric modulators of the GHB high-affinity binding site. Our data show that monastrol and (þ )-catechin bind allosterically and induce conformational changes that enhance [3H]NCS-382 affinity. Thus our study provides proof-of-concept that GHB activity – probably via certain GABAA receptors – can be affected through allosteric binding sites. Future structure-activity studies of monastrol and catechin may be used to elucidate the molecular and structural determinants for the allosteric effects. This may guide towards localization and delineation of these sites and potentially the development of selective and potent allosteric modulators of the GHB highaffinity site present on α4βδ or other GABAA receptors. Such compounds would be of value in relation to both in vitro and in vivo applications.

Acknowledgments Late professor Jerzy W. Jaroszewski († October 2011) is acknowledged for the initial design of the natural product library and Dorte Brix for technical assistance. The authors also wish to thank professor Hans Bräuner-Osborne for fruitful discussions. This work was supported by a Ph.D. stipend to LFE from the Faculty of Pharmaceutical Sciences (now Faculty of Health and Medical Sciences), and by the AP Møller Foundation, the Novo Nordisk Foundation, the Poul & Agnes Friis Foundation, ‘Carl og Ellen Hertz Legat for Dansk Lægevidenskab’, the Lundbeck Foundation, the Drug Research Academy and the ERASMUS student mobility program (DH).

Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.ejphar.2014.06.028. References Absalom, N., Eghorn, L.F., Villumsen, I.S., Karim, N., Bay, T., Olsen, J.V., Knudsen, G. M., Bräuner-Osborne, H., Frølund, B., Clausen, R.P., Chebib, M., Wellendorph, P., 2012. α4βδ GABAA receptors are high-affinity targets for γ-hydroxybutyric acid (GHB). Proc. Natl. Acad. Sci. USA 109, 13404–13409. Adachi, N., Tomonaga, S., Tachibana, T., Denbow, D.M., Furuse, M., 2006. (  )-Epigallocatechin gallate attenuates acute stress responses through GABAergic system in the brain. Eur. J. Pharmacol. 531, 171–175. Bay, T., Eghorn, L.F., Klein, A.B., Wellendorph, P., 2014. GHB receptor targets in the CNS: focus on high-affinity binding sites. Biochem. Pharmacol. 87, 220–228. Belelli, D., Lambert, J.J., 2005. Neurosteroids: endogenous regulators of the GABAA receptor. Nat. Rev. Neurosci. 6, 565–575.

Benavides, J., Rumigny, J.F., Bourguignon, J.J., Cash, C., Wermuth, C.G., Mandel, P., Vincendon, G., Maitre, M., 1982. High affinity binding sites for γ-hydroxybutyric acid in rat brain. Life Sci. 30, 953–961. Bernasconi, R., Mathivet, P., Bischoff, S., Marescaux, C., 1999. Gammahydroxybutyric acid: an endogenous neuromodulator with abuse potential. Trends Pharmacol. Sci. 20, 135–141. Bosch, O.G., Quednow, B.B., Seifritz, E., Wetter, T.C., 2012. Reconsidering GHB: orphan drug or new model antidepressant. J. Psychopharmacol. 26, 618–628. Campbell, E.L., Chebib, M., Johnston, G.A., 2004. The dietary flavonoids apigenin and (  )-epigallocatechin gallate enhance the positive modulation by diazepam of the activation by GABA of recombinant GABAA receptors. Biochem. Pharmacol. 68, 1631–1638. Carter, L.P., Koek, W., France, C.P., 2009. Behavioral analyses of GHB: receptor mechanisms. Pharmacol. Ther. 121, 100–114. Christopoulos, A., 2002. Allosteric binding sites on cell-surface receptors: novel targets for drug discovery. Nat. Rev. Drug Discov. 1, 198–210. Connelly, W.M., Errington, A.C., Crunelli, V., 2013. γ-Hydroxybutyric acid (GHB) is not an agonist of extrasynaptic GABAA receptors. PLoS One 8, e79062. Crunelli, V., Emri, Z., Leresche, N., 2006. Unravelling the brain targets of γ-hydroxybutyric acid. Curr. Opin. Pharmacol. 6, 44–52. Dallinger, D., Kappe, C.O., 2007. Rapid preparation of the mitotic kinesin Eg5 inhibitor monastrol using controlled microwave-assisted synthesis. Nat. Protoc. 2, 317–321. Drasbek, K.R., Christensen, J., Jensen, K., 2006. Gamma-hydroxybutyrate – a drug of abuse. Acta Neurol. Scand. 114, 145–156. Farrant, M., Nusser, Z., 2005. Variations on an inhibitory theme: phasic and tonic activation of GABAA receptors. Nat. Rev. Neurosci. 6, 215–229. Fernandez, S.P., Mewett, K.N., Hanrahan, J.R., Chebib, M., Johnston, G.A., 2008. Flavan-3-ol derivatives are positive modulators of GABAA receptors with higher efficacy for the α2 subtype and anxiolytic action in mice. Neuropharmacology 55, 900–907. Hanrahan, J.R., Chebib, M., Johnston, G.A., 2011. Flavonoid modulation of GABAA receptors. Br. J. Pharmacol. 163, 234–245. Hoestgaard-Jensen, K., Dalby, N.O., Wolinsky, T.D., Murphey, C., Jones, K.A., Rottlander, M., Frederiksen, K., Watson, W.P., Jensen, K., Ebert, B., 2010. Pharmacological characterization of a novel positive modulator at α4β3δ-containing extrasynaptic GABAA receptors. Neuropharmacology 58, 702–711. Hoestgaard-Jensen, K., O'Connor, R.M., Dalby, N.O., Simonsen, C., Finger, B.C., Golubeva, A., Hammer, H., Bergmann, M.L., Kristiansen, U., Krogsgaard-Larsen, P., Bräuner-Osborne, H., Ebert, B., Frølund, B., Cryan, J.F., Jensen, A.A., 2013. The orthosteric GABAA receptor ligand Thio-4-PIOL displays distinctly different functional properties at synaptic and extrasynaptic receptors. Br. J. Pharmacol. 170, 919–932. Høg, S., Wellendorph, P., Nielsen, B., Frydenvang, K., Dahl, I.F., Bräuner-Osborne, H., Brehm, L., Frølund, B., Clausen, R.P., 2008. Novel high-affinity and selective biaromatic 4-substituted γ-hydroxybutyric acid (GHB) analogues as GHB ligands: design, synthesis, and binding studies. J. Med. Chem. 51, 8088–8095. Hossain, S.J., Hamamoto, K., Aoshima, H., Hara, Y., 2002. Effects of tea components on the response of GABAA receptors expressed in Xenopus oocytes. J. Agric. Food Chem. 50, 3954–3960. Kamal, A., Azeeza, S., Bharathi, E.V., Malik, M.S., Shetti, R.V., 2010. Search for new and novel chemotherapeutics for the treatment of human malignancies. Mini Rev. Med. Chem. 10, 405–435. Kantrowitz, J.T., Citrome, L., Javitt, D.C., 2009. A review of tolerability and abuse liability of γ-hydroxybutyric acid for insomnia in patients with schizophrenia. Clin. Ther. 31 (Pt 1), 1360–1373. Karim, N., Gavande, N., Wellendorph, P., Johnston, G.A., Hanrahan, J.R., Chebib, M., 2011. 3-Hydroxy-2'-methoxy-6-methylflavone: a potent anxiolytic with a unique selectivity profile at GABAA receptor subtypes. Biochem. Pharmacol. 82, 1971–1983. Karim, N., Wellendorph, P., Absalom, N., Bang, L.H., Jensen, M.L., Hansen, M.M., Lee, H.J., Johnston, G.A., Hanrahan, J.R., Chebib, M., 2012. Low nanomolar GABA effects at extrasynaptic α4β1/β3δ GABAA receptor subtypes indicate a different binding mode for GABA at these receptors. Biochem. Pharmacol. 84, 549–557. Kaupmann, K., Cryan, J.F., Wellendorph, P., Mombereau, C., Sansig, G., Klebs, K., Schmutz, M., Froestl, W., van der Putten, H., Mosbacher, J., Bräuner-Osborne, H., Waldmeier, P., Bettler, B., 2003. Specific γ-hydroxybutyrate-binding sites but loss of pharmacological effects of γ-hydroxybutyrate in GABAB(1)-deficient mice. Eur. J. Neurosci. 18, 2722–2730. Keating, G.M., 2014. Sodium oxybate: a review of its use in alcohol withdrawal syndrome and in the maintenance of abstinence in alcohol dependence. Clin. Drug Investig. 34, 63–80. Leff, P., 1995. The two-state model of receptor activation. Trends Pharmacol. Sci. 16, 89–97. Lewis, R.W., Mabry, J., Polisar, J.G., Eagen, K.P., Ganem, B., Hess, G.P., 2010. Dihydropyrimidinone positive modulation of δ-subunit-containing γaminobutyric acid type A receptors, including an epilepsy-linked mutant variant. Biochemistry 49, 4841–4851. Maitre, M., 1997. The γ-hydroxybutyrate signalling system in brain: organization and functional implications. Prog. Neurobiol. 51, 337–361. Maliga, Z., Kapoor, T.M., Mitchison, T.J., 2002. Evidence that monastrol is an allosteric inhibitor of the mitotic kinesin Eg5. Chem. Biol. 9, 989–996. Mayer, T.U., Kapoor, T.M., Haggarty, S.J., King, R.W., Schreiber, S.L., Mitchison, T.J., 1999. Small molecule inhibitor of mitotic spindle bipolarity identified in a phenotype-based screen. Science 286, 971–974.

L.F. Eghorn et al. / European Journal of Pharmacology 740 (2014) 570–577

Mehta, A.K., Muschaweck, N.M., Maeda, D.Y., Coop, A., Ticku, M.K., 2001. Binding characteristics of the γ-hydroxybutyric acid receptor antagonist [3H](2E)-(5hydroxy-5,7,8,9-tetrahydro-6H-benzo[a][7]annulen-6-ylidene) ethanoic acid in the rat brain. J. Pharmacol. Exp. Ther. 299, 1148–1153. Noetzel, M.J., Rook, J.M., Vinson, P.N., Cho, H.P., Days, E., Zhou, Y., Rodriguez, A.L., Lavreysen, H., Stauffer, S.R., Niswender, C.M., Xiang, Z., Daniels, J.S., Jones, C.K., Lindsley, C.W., Weaver, C.D., Conn, P.J., 2012. Functional impact of allosteric agonist activity of selective positive allosteric modulators of metabotropic glutamate receptor subtype 5 in regulating central nervous system function. Mol. Pharmacol. 81, 120–133. Olsen, R.W., Wong, E.H., Stauber, G.B., King, R.G., 1984. Biochemical pharmacology of the γ-aminobutyric acid receptor/ionophore protein. Fed. Proc. 43, 2773–2778. Ondo, W.G., Perkins, T., Swick, T., Hull Jr., K.L., Jimenez, J.E., Garris, T.S., Pardi, D., 2008. Sodium oxybate for excessive daytime sleepiness in Parkinson disease: an open-label polysomnographic study. Arch. Neurol. 65, 1337–1340. Ransom, R.W., Stec, N.L., 1988. Cooperative modulation of [3H]MK-801 binding to the N-methyl-D-aspartate receptor-ion channel complex by L-glutamate, glycine, and polyamines. J. Neurochem. 51, 830–836. Rashid, U., Batool, I., Wadood, A., Khan, A., ul-Haq, Z., Chaudhary, M.I., Ansari, F.L., 2013. Structure based virtual screening-driven identification of monastrol as a potent urease inhibitor. J. Mol. Graph. Model. 43, 47–57. Robinson, D.M., Keating, G.M., 2007. Sodium oxybate: a review of its use in the management of narcolepsy. CNS Drugs 21, 337–354. Sabbatini, P., Wellendorph, P., Høg, S., Pedersen, M.H.F., Bräuner-Osborne, H., Martiny, L., Frølund, B., Clausen, R.P., 2010. Design, synthesis and in vitro pharmacology of new radiolabelled GHB analogues including photolabile analogues with irreversible binding to the high-affinity GHB binding sites. J. Med. Chem. 53, 6506–6510. Sieghart, W., 1992. GABAA receptors: ligand-gated Cl  ion channels modulated by multiple drug-binding sites. Trends Pharmacol. Sci. 13, 446–450.

577

Sigel, E., Buhr, A., 1997. The benzodiazepine binding site of GABAA receptors. Trends Pharmacol. Sci. 18, 425–429. Snead, O.C.I., Nichols, A.C., 1987. γ-Hydroxybutyric acid binding sites: evidence for coupling to a chloride anion channel. Neuropharmacology 26, 1519–1523. Valant, C., Felder, C.C., Sexton, P.M., Christopoulos, A., 2012. Probe dependence in the allosteric modulation of a G protein-coupled receptor: implications for detection and validation of allosteric ligand effects. Mol. Pharmacol. 81, 41–52. Vogensen, S.B., Marek, A., Bay, T., Wellendorph, P., Kehler, J., Bundgaard, C., Frølund, B., Pedersen, M.H., Clausen, R.P., 2013. New synthesis and tritium labeling of a selective ligand for studying high-affinity γ-hydroxybutyrate (GHB) binding sites. J. Med. Chem. 56, 8201–8205. Wafford, K.A., van Niel, M.B., Ma, Q.P., Horridge, E., Herd, M.B., Peden, D.R., Belelli, D., Lambert, J.J., 2009. Novel compounds selectively enhance δ subunit containing GABAA receptors and increase tonic currents in thalamus. Neuropharmacology 56, 182–189. Wellendorph, P., Høg, S., Sabbatini, P., Pedersen, M.H., Martiny, L., Knudsen, G.M., Frølund, B., Clausen, R.P., Bräuner-Osborne, H., 2010. Novel radioiodinated GHB analogues for radiolabeling and photolinking of high-affinity GHB binding sites. J. Pharmacol. Exp. Ther. 335, 458–464. Wellendorph, P., Høg, S., Skonberg, C., Bräuner-Osborne, H., 2009. Phenylacetic acids and the structurally related non-steroidal anti-inflammatory drug (NSAID) diclofenac bind to specific γ-hydroxybutyric acid (GHB) sites in rat brain. Fundam. Clin. Pharmacol. 23, 207–213. Whiting, P.J., Bonnert, T.P., McKernan, R.M., Farrar, S., Le Bourdelles, B., Heavens, R.P., Smith, D.W., Hewson, L., Rigby, M.R., Sirinathsinghji, D.J., Thompson, S.A., Wafford, K.A., 1999. Molecular and functional diversity of the expanding GABAA receptor gene family. Ann. N. Y. Acad. Sci. 868, 645–653. Yang, J.S., Olsen, R.W., 1987. γ-Aminobutyric acid receptor binding in fresh mouse brain membranes at 22 degrees C: ligand-induced changes in affinity. Mol. Pharmacol. 32, 266–277.

Positive allosteric modulation of the GHB high-affinity binding site by the GABAA receptor modulator monastrol and the flavonoid catechin.

γ-Hydroxybutyric acid (GHB) is a metabolite of γ-aminobutyric acid (GABA) and a proposed neurotransmitter in the mammalian brain. We recently identifi...
800KB Sizes 0 Downloads 4 Views