- Email: [email protected]
Hyperekplexia mutation R271L of α1 glycine receptors potentiates allosteric interactions of nortropeines, propofol and glycine with [3H]strychnine binding
Neurochemistry International 52 (2008) 235–240 www.elsevier.com/locate/neuint
Hyperekplexia mutation R271L of a1 glycine receptors potentiates allosteric interactions of nortropeines, propofol and glycine with [3H]strychnine binding Ga´bor Maksay a,*, Tı´mea Bı´ro´ a, Bodo Laube b,1, Pe´ter Nemes c a
Department of Molecular Pharmacology, Institute of Biomolecular Chemistry, Chemical Research Center, Hungarian Academy of Sciences, H-1525 Budapest, POB 17, Hungary b Department of Neurochemistry, Max-Planck-Institute for Brain Research, Deutschordenstraße 46, 60528 Frankfurt-M, Germany c Department of Chemistry, Szent Istva´n University, H-1400 Budapest, POB 2, Hungary Received 26 February 2007; received in revised form 11 June 2007; accepted 14 June 2007 Available online 21 June 2007
Abstract Human a1 and hyperekplexia mutant a1(R271L) glycine receptors (GlyRs) were transiently expressed in human embryonic kidney 293 cells for [3H]strychnine binding. Binding parameters were determined using a ternary allosteric model. The hyperekplexia mutation increased the positive cooperativity of 0.3 mM propofol and glycine binding by about six times: the cooperativity factor b was 0.26 for a1 GlyRs and 0.04 for a1(R271L) GlyRs. Thus, propofol restored the potency of glycine impaired by the mutation. Five nortropeines, i.e. substituted benzoates of nortropine and a new compound, nortropisetron were prepared and also examined on [3H]strychnine binding. They showed nanomolar displacing potencies amplified by the hyperekplexia mutation. The affinity of nor-O-zatosetron (2.6 nM) is one of the highest reported for GlyRs. This binding test offers an in vitro method to evaluate agents against neurological disorders associated with inherited mutations of GlyRs. # 2007 Elsevier Ltd. All rights reserved. Keywords: Recombinant a1 glycine receptors; [3H]strychnine binding; Hyperekplexia mutation; Tropeines; Propofol; Ternary allosteric model
1. Introduction The Cys-loop superfamily of ionotropic receptors comprises important neurotransmitter receptors such as glycine receptors (GlyRs), GABAA, nicotinic acetylcholine and 5-HT3-type serotonin receptors. Allosteric modulation of these ionotropic receptors can lead to the fine-tuning of the underlying neurotransmission. Allosteric agents can increase not only the potency but also the efficacy of partial agonists as demonstrated for GlyRs (Bı´ro´ and Maksay, 2004; O’Shea et al., 2004). Glycine is the major inhibitory neurotransmitter in spinal cord. Postsynaptic GlyRs play predominant roles in motor and sensory processing, neuronal development and inflammatory pain sensitization (Betz and Laube, 2006). Inherited mutations * Corresponding author. Tel.: +36 1 438 0414x282; fax: +36 1 325 7554. E-mail address: [email protected] (G. Maksay). 1 Current address: AG Zellula¨re Neurophysiologie, Technical University of Darmstadt, Schnittpahnstrasse 3, 64287 Darmstadt, Germany. 0197-0186/$ – see front matter # 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuint.2007.06.009
of GlyRs lead to neurological disorders of glycinergic transmission (Lynch, 2004). GlyRs are still ‘‘therapeutic orphans’’ although they might become therapeutic targets of analgesic, muscle relaxant and sedative agents (Laube et al., 2000). Four a (a1–a4) and b GlyR subunits have been cloned. Adult spinal GlyRs are composed of two a1 and three b subunits (Grudzinska et al., 2005). The major a1 subunits form pentamers with chloride channel function. One major point mutation of a1 GlyRs leading to hyperekplexia is R271L (Lynch, 2004; Shiang et al., 1993; Zhou et al., 2002). R271L impairs gating of a1 GlyRs and decreases the potency and efficacy of glycine (Langosch et al., 1994), while it turns partial agonists taurine and b-alanine into antagonists (Rajendra et al., 1995). Allosteric modulation of partial agonists is more effective: positive binding cooperativities of propofol and neuroactive steroids are stronger with taurine, b-alanine and GABA than with glycine (Bı´ro´ and Maksay, 2004), while Zn2+ increases the gating efficacy and potency of taurine (Laube et al., 2000). These observations justify the application of allosteric modulators on hyperekplexia mutations of GlyRs.
236
G. Maksay et al. / Neurochemistry International 52 (2008) 235–240
Even though gating and binding are strongly interrelated, Colquhoun (1998) has differentiated gating and binding mutations, and considered R271L as a ‘‘gating mutation’’ of a1 GlyRs. Subsequent homology modelling of a1 GlyRs (Fig. 1) supports this conclusion. Arginine 271 is located at the end of the second transmembrane (TM2) helix (Betz and Laube, 2006; Lynch, 2004). The binding site of agonists and the antagonist strychnine is in the extracellular interface of subunits (Grudzinska et al., 2005). Binding pockets of ethanol and intravenous anesthetic alcohols such as propofol have been localized between TM2 and TM3 of GABAA receptors and GlyRs (Bali and Akabas, 2004; Hemmings et al., 2005; Lobo et al., 2004). The three distinct locations are demonstrated in Fig. 1. Pharmacological fine-tuning of glycinergic transmission via allosteric modulation of GlyRs would require selective agents with high affinity. Although anaesthetic alcohols potentiate GlyR function but with low affinity and without selectivity: propofol is an intravenous anaesthetic acting via GABAA receptors rather than GlyRs (Belelli et al., 1999; Hemmings et al., 2005). However, subanaesthetic doses of propofol restores the functional and behavioural defects caused by mutations of R271 to Q, K and L in a1 GlyRs (O’Shea et al., 2004). On the other hand, submicromolar
Fig. 1. Homology model of a1 GlyRs. Two subunits are depicted with key residues in black indicating the different location of R271 from the binding sites of agonists and strychnine as well as that of anaesthetic alcohols. The modified figure is based on the model in Betz and Laube (2006).
concentrations of some 5-HT3 receptor antagonists such as tropisetron and zatosetron potentiate (Chesnoy-Marchais, 1996), while micromolar concentrations inhibit GlyR function (Maksay et al., 1999) via changing the gating potency of glycine. The structure required for ‘‘glycine-positive’’, potentiating effects has been attributed to tropeines, specifically, to aromatic esters and amides of 3a-hydroxy-tropane, i.e. tropine (Maksay, 1998). We have started from tropeines to develop high-affinity GlyR-selective allosteric modulators and observed that N-demethyl or nortropeines: substituted benzoates of nortropine displayed nanomolar affinities for native spinal (a1b) GlyRs of adult rats (Maksay et al., 2004). The potentiation of GlyR function by nanomolar tropisetron has shown remarkable subunit-selectivity for heteromeric ab GlyRs (Supplisson and Chesnoy-Marchais, 2000). Therefore we also prepared and examined the N-demethyl derivative, nortropisetron here. The aim of the present study was to develop an in vitro method to study hyperekplexia mutant GlyRs and to characterize their allosteric modulators quantitatively. Recombinant human a1 and hyperekplexia mutant a1(R271L) GlyRs were used. Displacement of [3H]strychnine binding was analysed via a ternary allosteric model (Maksay and Bı´ro´, 2002). We examined nortropisetron and four further nortropeines which were the most potent displacers of [3H]strychnine binding to native GlyRs (Maksay et al., 2004) as well as propofol (Fig. 2) as the most potent anaesthetic alcohol. One of the nortropeines is the N-demethyl derivative of zatosetron, an antiemetic 5-HT3 receptor antagonist, with an ester instead of amide moiety (nor-O-zatosetron, Fig. 2). Since binding and gating of ionotropic receptors are interrelated, binding constants including thermodynamic parameters of binding can inform us about the functional states of GlyRs (Maksay et al., 2002). The b value of binding cooperativity between glycine agonists and allosteric agents reflects the direction of allosteric modulation (potentiation or inhibition) of ionophore function. For potentiating agents of GlyRs b < 1, while for inhibitory ones b > 1 (Maksay and Bı´ro´, 2002). Here, we report that the hyperekplexia mutation R271L of a1 GlyRs which impairs gating and binding, facilitated both propofol potentiation of glycine displacement and nortropeine inhibition of [3H]strychnine binding. The displacing potency of nor-Ozatosetron at hyperekplexia mutant human a1 GlyRs was one of the highest affinities reported for GlyRs. In conclusion, this binding test offers an in vitro binding method to evaluate agents
Fig. 2. Chemical structures of nor-O-zatosetron and propofol.
G. Maksay et al. / Neurochemistry International 52 (2008) 235–240
against neurological disorders associated with inherited mutations of GlyRs.
237
3. Results 3.1. Propofol restored the coupling of glycine binding impaired by the hyperekplexia mutation
2. Materials and methods 2.1. Materials Four benzoyl nortropeines: nor-O-zatosetron (nortropine 20 ,30 -benzofuran50 ,50 -dimethyl-80 -chloro-60 -carboxylate), nortropine benzoate, nortropine 3chloro-benzoate and nortropine 3-methoxy-benzoate were synthesized as described (Maksay et al., 2004). Nortropisetron (nortropine indol-3-carboxylate) was prepared through the demethylation of tropisetron with trichloroethyl chloroformate and subsequent removal of the trichloroethoxycarbonyl group with zinc in acetic acid. [3H]strychnine (10 mCi/mmol) was purchased from Dupont-NEN. Propofol originates from Tocris Cookson (Bristol, UK). Human embryonic kidney (HEK) 293 cells (ATCC CRL 1573) were transfected with the human GlyR a1 subunit cDNA and the mutant a1R271L inserted into the mammalian expression vectors pCIS2 as described (Langosch et al., 1994; Laube and Betz, 1998).
2.2. Binding studies Transfected cells were harvested after 48 h of expression, suspended in 50 mM Tris–HCl buffer (pH 7.4), homogenized with Ultra-Turrax for 10 s and centrifuged at 48,000 g for 15 min. The pellets were washed by a similar centrifugation and frozen. Thawed membrane suspensions in 50 mM Tris–HCl buffer containing 100 mM KSCN (pH 7.4) were centrifuged at 40,000 g for 15 min, resuspended and incubated with 5 nM [3H]strychnine at 0 8C for 45 min. Triplicate 0.2 ml samples were filtered on Whatman GF/C filters under vacuum and washed by 3 3-ml ice cold 50 mM Tris–HCl buffer containing 100 mM KCl. Radioactivity of the filters was measured in a scintillation spectrophotometer. Displacing effects of glycine and taurine were examined in the absence and presence of a constant concentration of propofol. Nonspecific binding was determined in the presence of 10 mM strychnine.
2.3. Data analysis Nonlinear regression via GraphPad Prism 4 (San Diego, CA) was used for curve fitting. The ternary allosteric model (Jakubı´k et al., 1997) was modified with a slope factor (n) of the displacement by glycine as described previously (Bı´ro´ and Maksay, 2004). It contains three dissociation constants (KS, KL and KA) for the binding of [3H]strychnine (S), glycine (L) and the allosteric agents (A) as well as the cooperativity factors of A with [3H]strychnine (a0 ) and with glycine (b). The lower are a0 and b values than unity, the stronger is positive cooperativity. Labelling of a0 is meant to distinguish it from a subunits. Eq. (1) expresses the ratio of specific [3H]strychnine binding in the presence of three ligands (BSAL) over control (BS, in the presence of [3H]strychnine):
[3H]strychnine binding was studied to human a1 GlyRs transiently expressed in HEK cells. Glycine is competitive with [3H]strychnine for GlyRs and resulted in concentrationdependent, full displacement of its specific binding (Fig. 3). In contrast, propofol is an allosteric agent and a partial displacer of [3H]strychnine binding (Maksay and Bı´ro´, 2002). Displacement of [3H]strychnine binding to a1(R271L) GlyRs was also partial and it reached saturation around 0.3 mM of propofol (data not shown). Propofol strongly increased the displacing potency of glycine and decreased the slope of displacement (Fig. 3). Therefore a ternary allosteric model was used modified with a slope factor (n) introduced for [3H]strychnine binding to native GlyRs (Bı´ro´ and Maksay, 2004). Table 1 shows that propofol reduced significantly the slope factor n. Further, propofol had weak (high micromolar) affinity to a1 GlyRs and weak negative cooperativity with [3H]strychnine binding (a0 = 2.1) similar to native spinal (a1b) GlyRs of adult rats (Maksay et al., 2004). The positive cooperativity between propofol and glycine binding was characterized with b = 0.26 (Table 1). Saturation analysis of [3H]strychnine binding at 0 8C resulted in KS = 4.3 nM for a1 GlyRs and KS = 3.1 nM for a1(R271L) GlyRs (Maksay et al., 2002). Consequently, the hyperekplexia mutation R271L of a1 GlyRs did not significantly affect either the affinity of [3H]strychnine binding or its thermodynamic parameters (Maksay et al., 2002). In contrast, this mutation strongly decreased the displacing potency of glycine (Fig. 3) (Langosch et al., 1994; Rajendra et al., 1995). Table 1 shows that propofol binding was characterized with KA and a0 values being similar for wild and
BSAL ½S þ K S ¼ BS ½S þ K S fð½Ln ðK A þ ½A=bÞ þ KLn ðK A þ ½AÞÞ=KLn ðK A þ ½Aa0 Þg (1) KS values of [3H]strychnine binding were determined via saturation analysis in identical conditions (Maksay et al., 2004). Parallel displacement experiments by glycine were performed in the absence and presence of 0.3 mM propofol. Control KL and n values were determined via Eq. (1) where A = 0. Concentration-dependent displacement by the nortropeines was fitted to Eq. (2): BSA ½S þ K S ¼ BS ½S þ fðK S ðKAm þ ½Am ÞÞ=ðKAm þ ð½Am =a0 m ÞÞg
(2)
where m is the slope factor of displacement by nortropeines. However, when m was not significantly different from unity, m = 1 was applied.
Fig. 3. Effects of propofol on the displacement of [3H]strychnine binding by glycine for a1 and a1(R271L) GlyRs transiently expressed in HEK 293 cells. Points are means S.E.M. of three experiments. Propofol (0.3 mM) decreased [3H]strychnine binding of a1 GlyRs to 66.6 1.1% and that of a1(R271L) GlyRs to 77.6 2.5% (means S.D. of three experiments). Ratios of specific [3H]strychnine binding are plotted in the presence of three ligands (BSAL) over control (BS, in the presence of [3H]strychnine alone). Displacement curves of glycine for a1 (&), in the presence of propofol (*); and a1(R271L) GlyRs (&), in the presence of propofol (*).
238
G. Maksay et al. / Neurochemistry International 52 (2008) 235–240
Table 1 Binding and cooperativity parameters of propofol with [3H]strychnine and glycine for a1 and a1(R271L) GlyRs Parameters Control KL (mM) n +0.3 mM propofol KA (mM) n a0 b
a1 GlyR
a1(R271L) GlyR
91 24 0.85 0.08
1140 150a 1.19 0.19
61 9 0.57 0.06b 2.1 0.1 0.26 0.03
49 5 0.98 0.03 2.0 0.1 0.04 0.01a
Membrane suspensions in 50 mM Tris–HCl buffer (pH 7.4) containing 100 mM KSCN were incubated with 5 nM [3H]strychnine for 45 min at 0 8C. Displacement data were fitted to Eq. (1) according to the ternary allosteric model with variable slopes, where KS = 4.3 nM for a1 GlyRs and KS = 3.1 nM for a1(R271L) GlyRs. Data are means S.E.M. of three independently fitted experiments. a p < 0.05, significantly different from a1 GlyRs in Mann–Whitney nonparametric test. b p < 0.05, significantly lower than control when the curves for propofol were normalized for their control pairs and compared in Wilcoxon matched pairs test.
point-mutated a1 GlyRs. Interestingly, however, propofol has seven times stronger positive cooperativity with glycine (b = 0.04) on mutant than on wild GlyRs. Thus, propofol practically restored the displacing potency of glycine on [3H]strychnine impaired by the hyperekplexia mutation. Taurine, an endogenous partial agonist of extrasynaptic GlyRs (Flint et al., 1998; Betz and Laube, 2006) was also examined. [3H]Strychnine binding to recombinant a1 GlyRs was displaced by taurine with KL = 43 6 mM. Propofol (50 mM) displaced [3H]strychnine binding to 83.4 8.7% and its cooperativity with taurine in displacement of [3H]strychnine binding to a1(R271L) GlyRs was around neutral: b = 1.1 0.4 (mean S.E.M. of three experiments). 3.2. Nanomolar displacing potencies of nortropeines are increased by mutation R271L Four nortropeines, substituted benzoates of nortropine were the most potent displacers of [3H]strychnine binding to native GlyRs of rat spinal cord (Maksay et al., 2004). Therefore they were also examined here with recombinant GlyRs. The nortropeines exerted concentration-dependent, full displacement of specific [3H]strychnine binding to both a1 and a1(R271L) GlyRs as demonstrated for nor-O-zatosetron in Fig. 4. This can be attributed to strong negative cooperativity with [3H]strychnine (a0 10). Alternatively, this can be reconciled with competitive displacement as well. Fitting according to a competitive model (Maksay et al., 2005) resulted in KA values which were not significantly different from those in Table 2 (data not shown). All esters displayed nanomolar affinities to a1 GlyRs (Table 2) similar to those observed for native spinal GlyRs (Maksay et al., 2004). It is remarkable that the mutation R271L enhanced the potency of the nortropeines (Table 2) as demonstrated by nor-O-zatosetron displacement in Fig. 4. Further, nortropisetron was more than two orders of magnitude stronger displacer than tropisetron (Maksay and
Fig. 4. Displacement of [3H]strychnine binding by nor-O-zatosetron to recombinant human a1 and a1(R271L) GlyRs transiently expressed in HEK 293 cells. Points are means S.E.M. of three to five experiments. Ratios of specific [3H]strychnine binding are plotted in the presence of nor-O-zatosetron (BSA) over control (BS).
Bı´ro´, 2002). It should be noted that fitting to Eq. (2) resulted in slope values of displacement m = 0.64 0.10 for nortropisetron and m = 0.70 0.18 for nor-O-zatosetron (means S.E.M. of three experiments) for a1(R271L) GlyRs (Table 2). As shown in Table 2, nor-O-zatosetron displayed the highest affinity with KA = 2.6 nM. Moreover, it was selective for a1 GlyRs as compared to its lower displacing potency on native 5-HT3Rs (Table 2). 4. Discussion 4.1. Propofol restores the coupling of glycine binding and gating impaired by the hyperekplexia mutation Propofol binding to a1 GlyRs showed low affinity and weak negative cooperativity to displace [3H]strychnine binding (a0 = 2.1). However, 10 mM glycine increased the affinity of Table 2 Displacing potencies of nortropeines on [3H]strychnine binding to recombinant human a1 and a1(R271L) GlyRs expressed in HEK 293 cells Esters of nortropine
Nor-O-Zatosetron 3-Methoxy-benzoate Benzoate 3-Chloro-benzoate Nortropisetron
KA (nM)a a1
a1(R271L)
a1bb
8.6 1.9 16.3 1.7 13.5 1.2 ND ND
2.6 0.3c,d 12.2 4.6 7.4 1.2c 46 21 9.7 2.0f
18.1 (5-HT3R: 56 nM)e 7.7e 10.4e 22.6e
Data are mean S.E.M. of three to five experiments. Slope values not shown were not significantly different from unity. ND: not determined. a Dissociation constants of nortropeine binding (KA) were determined via Eq. (2), with negative cooperativities of a0 10. For comparison, the displacing potency of nor-O-zatosetron is indicated for [3H]granisetron binding to rat cortical 5-HT3 receptors. b Native GlyRs of rat spinal cord. c p < 0.05, significantly lower than for a1 GlyRs in Mann–Whitney test. d m = 0.70 0.18. e Taken from Maksay et al. (2004). f m = 0.64 0.10.
G. Maksay et al. / Neurochemistry International 52 (2008) 235–240
propofol to KA 20 mM to native GlyRs (Maksay, 1998). Propofol was studied in a high, saturating concentration (0.3 mM) strongly enhancing agonist gating of recombinant a1 GlyRs expressed in Xenopus oocytes (O’Shea et al., 2004). It exerted positive cooperativity with glycine displacement of [3H]strychnine binding to a1 GlyRs (b = 0.26, Table 1), in agreement with potentiation of a1 GlyR-ionophore function (O’Shea et al., 2004). Propofol also decreased the slope factor n of displacement by glycine significantly for a1 GlyRs. The slope of displacement by taurine was also decreased by 0.1 mM propofol for native spinal GlyRs (Bı´ro´ and Maksay, 2004). Decreased slope values might be attributed to increased binding heterogeneity of GlyR agonists (i.e. decreased overlap of two displacement phases) due to propofol. The hyperekplexia mutation R271L did not change significantly propofol binding and its cooperativity with [3H]strychnine (Fig. 3 and Table 1). This means that this mutation did not affect the binding sites of propofol and [3H]strychnine, each located elsewhere (Fig. 1). In contrast, R271L decreased the displacing potency of glycine strongly via impaired gating, although the agonist sites are also far from R271. Remarkably, the positive cooperativity between glycine and propofol (b = 0.04) was much stronger with a1(R271L) than with wild GlyRs. This means that propofol restored the displacing potency of glycine impaired by the hyperekplexia mutation. Accordingly, propofol restored gating of hyperekplexia mutants by glycine, a partial agonist of a1(R271L) GlyRs (O’Shea et al., 2004). In contrast, the binding cooperativity was about neutral between propofol and taurine, that becomes an antagonist of a1(R271L) GlyRs (Rajendra et al., 1995). In conclusion, even though propofol potentiates GlyRs with low affinity and without selectivity, it might become a therapeutic lead since subanaesthetic doses of propofol alleviated the spontaneous and induced tremor episodes of transgenic mice carrying the hyperekplexia mutation R271Q of a1 GlyRs (O’Shea et al., 2004). 4.2. Nanomolar displacing potencies of nortropeines are increased by the hyperekplexia mutation Four representative nortropeines displayed nanomolar affinities to recombinant human a1 GlyRs and strong negative cooperativity to displace [3H]strychnine binding. Their binding parameters were similar to those for native spinal (a1b) GlyRs of adult rats (Maksay et al., 2004). It should be noted that glycine (10 mM) increased the displacing potencies of tropisetron and zatosetron on [3H]strychnine binding to native GlyRs (Maksay, 1998). The glycine-elicited binding changes correlate with the potentiating effects of zatosetron and tropisetron on GlyR function (Maksay and Bı´ro´, 2002). Further, the slope values of these two tropeines for native spinal GlyRs (Maksay, 1998) as well as those of their nor derivatives for a1(R271L) GlyRs here were below unity. Slope values lower than unity can be reconciled with binding heterogeneity. Most allosteric modulators of GlyRs such as anaesthetic alcohols, steroids, dihydropyridines and Zn2+ have bidirectional
239
activities (Laube et al., 2000; Hirzel et al., 2006). Similarly, submicromolar concentrations of several tropeines potentiate GlyRs, while micromolar tropeines inhibit them (ChesnoyMarchais, 1996). Accordingly, increasing tropeine concentrations shift the cooperativity with glycine from positive to negative (Maksay et al., 2004). Furthermore, (nor)tropeines having strong affinities usually have weak positive cooperativities with glycine (Maksay et al., 2004). Potentiation of GlyR function by nanomolar tropisetron requires low agonist occupancy (Supplisson and Chesnoy-Marchais, 2000). It can be concluded that high affinities and saturating concentrations of tropeines and agonists lead to displacement of [3H]strychnine and agonists. Consequently, inhibition of GlyRs by high-affinity nortropeines appears to mask potentiation. Differences in the acyl parts of nortropeines lead to structure–activity relationships of recombinant human a1 and native rat (a1b) GlyRs (Maksay et al., 2004) which are similar but not identical: nortropine 3-methoxy-benzoate was most active on native GlyRs, while least active on a1 GlyRs here. On the other hand, these differences in the acyl parts did not result in excessive changes in displacing potencies. Consequently, the high affinities can be attributed predominantly to binding of the alcohol part, the bicyclic nortropane skeleton, particularly to N-demethylation. It is remarkable that mutation R271L enhanced the affinity of the nortropeines, especially that of nor-O-zatosetron. Potentiation of the displacing potencies by R271L supports that the nortropeines interact with a1 GlyRs differently from the antagonist strychnine which is unaffected by R271L. It is probably not coincidental that mutation R271L potentiated both the inhibition by nortopeines as well as the positive cooperativity of propofol with glycine. Similarly, allosteric modulation of native GlyRs has been more efficacious for various partial agonists rather than for glycine and this has been observed not only for positive cooperativity with propofol and neuroactive steroids but also for negative modulators (Bı´ro´ and Maksay, 2004). These findings can be interpreted so that less efficacious channel opening, either due to improper agonist structure (partial agonists) or due to receptor function impaired by hyperekplexia mutations can be both potentiated and attenuated better by allosteric interactions. The KA value of 2.6 nM of nor-O-zatosetron is one of the strongest affinities reported for GlyRs including those of strychnine and R-5135, a steroidal antagonist of GABAA and GlyRs (Hunt and Clements-Jewery, 1981). Glycine ‘‘elicits’’ a similar high affinity, but minor displacement phase by tropisetron for [3H]strychnine binding to native spinal GlyRs (Maksay, 1998). Since nor-O-zatosetron is selective in comparison to 5-HT3 receptors, it might be used to develop a selective label, a radioligand of GlyRs and to localize the binding sites of GlyRs for tropeines. Granisetron is an antagonist of 5-HT3 -type serotonin receptors with a similar tropeine structure and high affinity. Homology modelling and point mutations have led to the characterization of two binding modes of granisetron in the interface of 5-HT3A subunits where 5-HT3 receptor antagonists are bound (Joshi et al., 2006). The extracellular interface of Cys-
240
G. Maksay et al. / Neurochemistry International 52 (2008) 235–240
loop receptors including GlyRs (see in Fig. 1) is a reasonable candidate for binding sites of tropeines of similar structure. Finally, the present binding test provides us with an in vitro binding method to evaluate drug candidates against neurological disorders associated with inherited mutations of GlyRs. Acknowledgements This work was supported by a grant K 62203 of the Hungarian Science Research Fund OTKA. Continuous support of professor Heinrich Betz (Frankfurt/M), stimulating discussions with Dr. Sean O’Shea and technical assistance of Mrs. Ilona Kawka (CRC, Budapest) are acknowledged. References Bali, M., Akabas, M.H., 2004. Defining the propofol binding site location on the GABAA receptor. Mol. Pharmcol. 65, 68–76. Belelli, D., Pistis, M., Peters, J.A., Lambert, J.J., 1999. The interaction of general anaesthetics and neurosteroids with GABAA and glycine receptors. Neurochem. Int. 34, 447–452. Betz, H., Laube, B., 2006. Glycine receptors: recent insights into their structural organisation and functional diversity. J. Neurochem. 97, 1600–1610. Bı´ro´, T., Maksay, G., 2004. Allosteric modulation of glycine receptors is more efficacious for partial rather than full agonists. Neurochem. Int. 44, 521– 527. Chesnoy-Marchais, D., 1996. Potentiation of chloride responses to glycine by three 5-HT3 antagonists in rat spinal cord. Br. J. Pharmacol. 118, 2115– 2125. Colquhoun, D., 1998. Binding, gating, affinity and efficacy: the interpretation of structure–activity relationships for agonists and the effects of mutating receptors. Br. J. Pharmacol. 125, 923–947. Flint, A.C., Liu, X., Kriegstein, A.R., 1998. Nonsynaptic glycine receptor activation during early neocortical development. Neuron 20, 43–53. Grudzinska, J., Schemm, R., Haeger, S., Nicke, A., Schmalzing, G., Betz, H., Laube, B., 2005. The beta subunit determines the ligand binding properties of synaptic glycine receptors. Neuron 45, 727–739. Hemmings, H.C., Akabas, M.H., Goldstein, P.A., Trudell, J.R., Orser, B.A., Harrison, N.L., 2005. Emerging molecular mechanisms of general anesthetic action. Trends Pharmacol. Sci. 26, 503–510. Hirzel, K., Mu¨ller, U., Latal, A.T., Hu¨lsmann, S., Grudzinska, J., Seeliger, M.W., Betz, H., Laube, B., 2006. Hyperekplexia phenotype of glycine receptor a1 subunit mutant mice identifies Zn2+ as an essential endogenous modulator of glycinergic neurotransmission. Neuron 52, 679–690. Hunt, P., Clements-Jewery, S., 1981. A steroid derivative, R5135, antagonizes the GABA–benzodiazepine receptor interaction. Neuropharmacology 20, 357–361.
Jakubı´k, J., Bacakova, L., El-Fakahany, E.E., Tucˇek, S., 1997. Positive cooperativity of acetylcholine and other agonist with allosteric ligands on muscarinic acetylcholine receptors. Mol. Pharmacol. 52, 172–179. Joshi, P.R., Suryanarayanan, A., Hazai, E., Schulte, M.K., Maksay, G., Bika´di, Zs., 2006. Interactions of granisetron with an agonist-free 5-HT3A receptor model. Biochemistry 45, 1099–1105. Langosch, D., Laube, B., Rundstro¨m, N., Schmieden, V., Bormann, J., Betz, H., 1994. Decreased agonist affinity and chloride conductance of mutant glycine receptors associated with human hereditary hyperekplexia. EMBO J. 13, 4223–4228. Laube, B., Betz, H., 1998. Purification and heterogenous expression of inhibitory glycine receptors. Methods Enzymol. 294, 260–273. Laube, B., Maksay, G., Schemm, R., Betz, H., 2000. Modulation of glycine receptor function: a novel approach for therapeutic intervention at inhibitory synapses? Trends Pharmacol. Sci. 23, 519–527. Lobo, I.A., Mascia, M.P., Trudell, J.R., Harris, R.A., 2004. Channel gating of the glycine receptor changes accessibility to residues implicated in receptor potentiation by alcohols and anesthetics. J. Biol. Chem. 279, 33919–33927. Lynch, J.W., 2004. Molecular structure and function of the glycine receptor chloride channel. Physiol. Rev. 84, 1051–1095. Maksay, G., 1998. Bidirectional allosteric modulation of strychnine-sensitive glycine receptors by tropeines and 5-HT3 receptor ligands. Neuropharmacology 37, 1633–1641. Maksay, G., Bı´ro´, T., 2002. Dual cooperative allosteric modulation of binding to ionotropic glycine receptors. Neuropharmacology 43, 1087–1098 (Erratum: 2003, 44, 843). Maksay, G., Laube, B., Betz, H., 1999. Selective blocking effects of tropisetron and atropine on recombinant glycine receptors. J. Neurochem. 73, 802–806. Maksay, G., Bı´ro´, T., Laube, B., 2002. Hyperekplexia mutation of glycine receptors: decreased gating efficacy with altered binding thermodynamics. Biochem. Pharmacol. 64, 285–288. Maksay, G., Nemes, P., Bı´ro´, T., 2004. Synthesis of tropeines and allosteric modulation of ionotropic glycine receptors. J. Med. Chem. 47, 6384–6391. Maksay, G., Bı´ro´, T., Bugovics, G., 2005. Allosteric modulation of 5-HT3 serotonin receptors. Eur. J. Pharmacol. 514, 17–24. O’Shea, S.M., Becker, L., Weiher, H., Betz, H., Laube, B., 2004. Propofol restores the function of ‘hyperekplexic’ mutant glycine receptors in Xenopus oocytes and mice. J. Neurosci. 24, 2322–2327. Rajendra, S., Lynch, J.W., Pierce, K.D., French, C.R., Barry, P.H., Schofield, P.R., 1995. Mutation of an arginine residue in the human glycine receptor transforms b-alanine and taurine from agonists into competitive antagonists. Neuron 14, 169–175. Shiang, R., Ryan, S.G., Zhu, Y.Z., Hahn, A.F., O’Connell, P., Wasmuth, J.J., 1993. Mutations in the a1 subunit of the inhibitory glycine receptor cause the dominant neurologic disorder, hyperekplexia. Nat. Genet. 5, 351–358. Supplisson, S., Chesnoy-Marchais, D., 2000. Glycine receptor b subunits play a critical role in potentiation of glycine responses by ICS-205,930. Mol. Pharmacol. 58, 763–770. Zhou, L., Chillag, K.L., Nigro, M.A., 2002. Hyperekplexia: a treatable neurogenetic disease. Brain Dev. 24, 669–674.
Hyperekplexia mutation R271L of a1 glycine receptors potentiates allosteric interactions of nortropeines, propofol and glycine with [3H]strychnine binding Ga´bor Maksay a,*, Tı´mea Bı´ro´ a, Bodo Laube b,1, Pe´ter Nemes c a
Department of Molecular Pharmacology, Institute of Biomolecular Chemistry, Chemical Research Center, Hungarian Academy of Sciences, H-1525 Budapest, POB 17, Hungary b Department of Neurochemistry, Max-Planck-Institute for Brain Research, Deutschordenstraße 46, 60528 Frankfurt-M, Germany c Department of Chemistry, Szent Istva´n University, H-1400 Budapest, POB 2, Hungary Received 26 February 2007; received in revised form 11 June 2007; accepted 14 June 2007 Available online 21 June 2007
Abstract Human a1 and hyperekplexia mutant a1(R271L) glycine receptors (GlyRs) were transiently expressed in human embryonic kidney 293 cells for [3H]strychnine binding. Binding parameters were determined using a ternary allosteric model. The hyperekplexia mutation increased the positive cooperativity of 0.3 mM propofol and glycine binding by about six times: the cooperativity factor b was 0.26 for a1 GlyRs and 0.04 for a1(R271L) GlyRs. Thus, propofol restored the potency of glycine impaired by the mutation. Five nortropeines, i.e. substituted benzoates of nortropine and a new compound, nortropisetron were prepared and also examined on [3H]strychnine binding. They showed nanomolar displacing potencies amplified by the hyperekplexia mutation. The affinity of nor-O-zatosetron (2.6 nM) is one of the highest reported for GlyRs. This binding test offers an in vitro method to evaluate agents against neurological disorders associated with inherited mutations of GlyRs. # 2007 Elsevier Ltd. All rights reserved. Keywords: Recombinant a1 glycine receptors; [3H]strychnine binding; Hyperekplexia mutation; Tropeines; Propofol; Ternary allosteric model
1. Introduction The Cys-loop superfamily of ionotropic receptors comprises important neurotransmitter receptors such as glycine receptors (GlyRs), GABAA, nicotinic acetylcholine and 5-HT3-type serotonin receptors. Allosteric modulation of these ionotropic receptors can lead to the fine-tuning of the underlying neurotransmission. Allosteric agents can increase not only the potency but also the efficacy of partial agonists as demonstrated for GlyRs (Bı´ro´ and Maksay, 2004; O’Shea et al., 2004). Glycine is the major inhibitory neurotransmitter in spinal cord. Postsynaptic GlyRs play predominant roles in motor and sensory processing, neuronal development and inflammatory pain sensitization (Betz and Laube, 2006). Inherited mutations * Corresponding author. Tel.: +36 1 438 0414x282; fax: +36 1 325 7554. E-mail address: [email protected] (G. Maksay). 1 Current address: AG Zellula¨re Neurophysiologie, Technical University of Darmstadt, Schnittpahnstrasse 3, 64287 Darmstadt, Germany. 0197-0186/$ – see front matter # 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuint.2007.06.009
of GlyRs lead to neurological disorders of glycinergic transmission (Lynch, 2004). GlyRs are still ‘‘therapeutic orphans’’ although they might become therapeutic targets of analgesic, muscle relaxant and sedative agents (Laube et al., 2000). Four a (a1–a4) and b GlyR subunits have been cloned. Adult spinal GlyRs are composed of two a1 and three b subunits (Grudzinska et al., 2005). The major a1 subunits form pentamers with chloride channel function. One major point mutation of a1 GlyRs leading to hyperekplexia is R271L (Lynch, 2004; Shiang et al., 1993; Zhou et al., 2002). R271L impairs gating of a1 GlyRs and decreases the potency and efficacy of glycine (Langosch et al., 1994), while it turns partial agonists taurine and b-alanine into antagonists (Rajendra et al., 1995). Allosteric modulation of partial agonists is more effective: positive binding cooperativities of propofol and neuroactive steroids are stronger with taurine, b-alanine and GABA than with glycine (Bı´ro´ and Maksay, 2004), while Zn2+ increases the gating efficacy and potency of taurine (Laube et al., 2000). These observations justify the application of allosteric modulators on hyperekplexia mutations of GlyRs.
236
G. Maksay et al. / Neurochemistry International 52 (2008) 235–240
Even though gating and binding are strongly interrelated, Colquhoun (1998) has differentiated gating and binding mutations, and considered R271L as a ‘‘gating mutation’’ of a1 GlyRs. Subsequent homology modelling of a1 GlyRs (Fig. 1) supports this conclusion. Arginine 271 is located at the end of the second transmembrane (TM2) helix (Betz and Laube, 2006; Lynch, 2004). The binding site of agonists and the antagonist strychnine is in the extracellular interface of subunits (Grudzinska et al., 2005). Binding pockets of ethanol and intravenous anesthetic alcohols such as propofol have been localized between TM2 and TM3 of GABAA receptors and GlyRs (Bali and Akabas, 2004; Hemmings et al., 2005; Lobo et al., 2004). The three distinct locations are demonstrated in Fig. 1. Pharmacological fine-tuning of glycinergic transmission via allosteric modulation of GlyRs would require selective agents with high affinity. Although anaesthetic alcohols potentiate GlyR function but with low affinity and without selectivity: propofol is an intravenous anaesthetic acting via GABAA receptors rather than GlyRs (Belelli et al., 1999; Hemmings et al., 2005). However, subanaesthetic doses of propofol restores the functional and behavioural defects caused by mutations of R271 to Q, K and L in a1 GlyRs (O’Shea et al., 2004). On the other hand, submicromolar
Fig. 1. Homology model of a1 GlyRs. Two subunits are depicted with key residues in black indicating the different location of R271 from the binding sites of agonists and strychnine as well as that of anaesthetic alcohols. The modified figure is based on the model in Betz and Laube (2006).
concentrations of some 5-HT3 receptor antagonists such as tropisetron and zatosetron potentiate (Chesnoy-Marchais, 1996), while micromolar concentrations inhibit GlyR function (Maksay et al., 1999) via changing the gating potency of glycine. The structure required for ‘‘glycine-positive’’, potentiating effects has been attributed to tropeines, specifically, to aromatic esters and amides of 3a-hydroxy-tropane, i.e. tropine (Maksay, 1998). We have started from tropeines to develop high-affinity GlyR-selective allosteric modulators and observed that N-demethyl or nortropeines: substituted benzoates of nortropine displayed nanomolar affinities for native spinal (a1b) GlyRs of adult rats (Maksay et al., 2004). The potentiation of GlyR function by nanomolar tropisetron has shown remarkable subunit-selectivity for heteromeric ab GlyRs (Supplisson and Chesnoy-Marchais, 2000). Therefore we also prepared and examined the N-demethyl derivative, nortropisetron here. The aim of the present study was to develop an in vitro method to study hyperekplexia mutant GlyRs and to characterize their allosteric modulators quantitatively. Recombinant human a1 and hyperekplexia mutant a1(R271L) GlyRs were used. Displacement of [3H]strychnine binding was analysed via a ternary allosteric model (Maksay and Bı´ro´, 2002). We examined nortropisetron and four further nortropeines which were the most potent displacers of [3H]strychnine binding to native GlyRs (Maksay et al., 2004) as well as propofol (Fig. 2) as the most potent anaesthetic alcohol. One of the nortropeines is the N-demethyl derivative of zatosetron, an antiemetic 5-HT3 receptor antagonist, with an ester instead of amide moiety (nor-O-zatosetron, Fig. 2). Since binding and gating of ionotropic receptors are interrelated, binding constants including thermodynamic parameters of binding can inform us about the functional states of GlyRs (Maksay et al., 2002). The b value of binding cooperativity between glycine agonists and allosteric agents reflects the direction of allosteric modulation (potentiation or inhibition) of ionophore function. For potentiating agents of GlyRs b < 1, while for inhibitory ones b > 1 (Maksay and Bı´ro´, 2002). Here, we report that the hyperekplexia mutation R271L of a1 GlyRs which impairs gating and binding, facilitated both propofol potentiation of glycine displacement and nortropeine inhibition of [3H]strychnine binding. The displacing potency of nor-Ozatosetron at hyperekplexia mutant human a1 GlyRs was one of the highest affinities reported for GlyRs. In conclusion, this binding test offers an in vitro binding method to evaluate agents
Fig. 2. Chemical structures of nor-O-zatosetron and propofol.
G. Maksay et al. / Neurochemistry International 52 (2008) 235–240
against neurological disorders associated with inherited mutations of GlyRs.
237
3. Results 3.1. Propofol restored the coupling of glycine binding impaired by the hyperekplexia mutation
2. Materials and methods 2.1. Materials Four benzoyl nortropeines: nor-O-zatosetron (nortropine 20 ,30 -benzofuran50 ,50 -dimethyl-80 -chloro-60 -carboxylate), nortropine benzoate, nortropine 3chloro-benzoate and nortropine 3-methoxy-benzoate were synthesized as described (Maksay et al., 2004). Nortropisetron (nortropine indol-3-carboxylate) was prepared through the demethylation of tropisetron with trichloroethyl chloroformate and subsequent removal of the trichloroethoxycarbonyl group with zinc in acetic acid. [3H]strychnine (10 mCi/mmol) was purchased from Dupont-NEN. Propofol originates from Tocris Cookson (Bristol, UK). Human embryonic kidney (HEK) 293 cells (ATCC CRL 1573) were transfected with the human GlyR a1 subunit cDNA and the mutant a1R271L inserted into the mammalian expression vectors pCIS2 as described (Langosch et al., 1994; Laube and Betz, 1998).
2.2. Binding studies Transfected cells were harvested after 48 h of expression, suspended in 50 mM Tris–HCl buffer (pH 7.4), homogenized with Ultra-Turrax for 10 s and centrifuged at 48,000 g for 15 min. The pellets were washed by a similar centrifugation and frozen. Thawed membrane suspensions in 50 mM Tris–HCl buffer containing 100 mM KSCN (pH 7.4) were centrifuged at 40,000 g for 15 min, resuspended and incubated with 5 nM [3H]strychnine at 0 8C for 45 min. Triplicate 0.2 ml samples were filtered on Whatman GF/C filters under vacuum and washed by 3 3-ml ice cold 50 mM Tris–HCl buffer containing 100 mM KCl. Radioactivity of the filters was measured in a scintillation spectrophotometer. Displacing effects of glycine and taurine were examined in the absence and presence of a constant concentration of propofol. Nonspecific binding was determined in the presence of 10 mM strychnine.
2.3. Data analysis Nonlinear regression via GraphPad Prism 4 (San Diego, CA) was used for curve fitting. The ternary allosteric model (Jakubı´k et al., 1997) was modified with a slope factor (n) of the displacement by glycine as described previously (Bı´ro´ and Maksay, 2004). It contains three dissociation constants (KS, KL and KA) for the binding of [3H]strychnine (S), glycine (L) and the allosteric agents (A) as well as the cooperativity factors of A with [3H]strychnine (a0 ) and with glycine (b). The lower are a0 and b values than unity, the stronger is positive cooperativity. Labelling of a0 is meant to distinguish it from a subunits. Eq. (1) expresses the ratio of specific [3H]strychnine binding in the presence of three ligands (BSAL) over control (BS, in the presence of [3H]strychnine):
[3H]strychnine binding was studied to human a1 GlyRs transiently expressed in HEK cells. Glycine is competitive with [3H]strychnine for GlyRs and resulted in concentrationdependent, full displacement of its specific binding (Fig. 3). In contrast, propofol is an allosteric agent and a partial displacer of [3H]strychnine binding (Maksay and Bı´ro´, 2002). Displacement of [3H]strychnine binding to a1(R271L) GlyRs was also partial and it reached saturation around 0.3 mM of propofol (data not shown). Propofol strongly increased the displacing potency of glycine and decreased the slope of displacement (Fig. 3). Therefore a ternary allosteric model was used modified with a slope factor (n) introduced for [3H]strychnine binding to native GlyRs (Bı´ro´ and Maksay, 2004). Table 1 shows that propofol reduced significantly the slope factor n. Further, propofol had weak (high micromolar) affinity to a1 GlyRs and weak negative cooperativity with [3H]strychnine binding (a0 = 2.1) similar to native spinal (a1b) GlyRs of adult rats (Maksay et al., 2004). The positive cooperativity between propofol and glycine binding was characterized with b = 0.26 (Table 1). Saturation analysis of [3H]strychnine binding at 0 8C resulted in KS = 4.3 nM for a1 GlyRs and KS = 3.1 nM for a1(R271L) GlyRs (Maksay et al., 2002). Consequently, the hyperekplexia mutation R271L of a1 GlyRs did not significantly affect either the affinity of [3H]strychnine binding or its thermodynamic parameters (Maksay et al., 2002). In contrast, this mutation strongly decreased the displacing potency of glycine (Fig. 3) (Langosch et al., 1994; Rajendra et al., 1995). Table 1 shows that propofol binding was characterized with KA and a0 values being similar for wild and
BSAL ½S þ K S ¼ BS ½S þ K S fð½Ln ðK A þ ½A=bÞ þ KLn ðK A þ ½AÞÞ=KLn ðK A þ ½Aa0 Þg (1) KS values of [3H]strychnine binding were determined via saturation analysis in identical conditions (Maksay et al., 2004). Parallel displacement experiments by glycine were performed in the absence and presence of 0.3 mM propofol. Control KL and n values were determined via Eq. (1) where A = 0. Concentration-dependent displacement by the nortropeines was fitted to Eq. (2): BSA ½S þ K S ¼ BS ½S þ fðK S ðKAm þ ½Am ÞÞ=ðKAm þ ð½Am =a0 m ÞÞg
(2)
where m is the slope factor of displacement by nortropeines. However, when m was not significantly different from unity, m = 1 was applied.
Fig. 3. Effects of propofol on the displacement of [3H]strychnine binding by glycine for a1 and a1(R271L) GlyRs transiently expressed in HEK 293 cells. Points are means S.E.M. of three experiments. Propofol (0.3 mM) decreased [3H]strychnine binding of a1 GlyRs to 66.6 1.1% and that of a1(R271L) GlyRs to 77.6 2.5% (means S.D. of three experiments). Ratios of specific [3H]strychnine binding are plotted in the presence of three ligands (BSAL) over control (BS, in the presence of [3H]strychnine alone). Displacement curves of glycine for a1 (&), in the presence of propofol (*); and a1(R271L) GlyRs (&), in the presence of propofol (*).
238
G. Maksay et al. / Neurochemistry International 52 (2008) 235–240
Table 1 Binding and cooperativity parameters of propofol with [3H]strychnine and glycine for a1 and a1(R271L) GlyRs Parameters Control KL (mM) n +0.3 mM propofol KA (mM) n a0 b
a1 GlyR
a1(R271L) GlyR
91 24 0.85 0.08
1140 150a 1.19 0.19
61 9 0.57 0.06b 2.1 0.1 0.26 0.03
49 5 0.98 0.03 2.0 0.1 0.04 0.01a
Membrane suspensions in 50 mM Tris–HCl buffer (pH 7.4) containing 100 mM KSCN were incubated with 5 nM [3H]strychnine for 45 min at 0 8C. Displacement data were fitted to Eq. (1) according to the ternary allosteric model with variable slopes, where KS = 4.3 nM for a1 GlyRs and KS = 3.1 nM for a1(R271L) GlyRs. Data are means S.E.M. of three independently fitted experiments. a p < 0.05, significantly different from a1 GlyRs in Mann–Whitney nonparametric test. b p < 0.05, significantly lower than control when the curves for propofol were normalized for their control pairs and compared in Wilcoxon matched pairs test.
point-mutated a1 GlyRs. Interestingly, however, propofol has seven times stronger positive cooperativity with glycine (b = 0.04) on mutant than on wild GlyRs. Thus, propofol practically restored the displacing potency of glycine on [3H]strychnine impaired by the hyperekplexia mutation. Taurine, an endogenous partial agonist of extrasynaptic GlyRs (Flint et al., 1998; Betz and Laube, 2006) was also examined. [3H]Strychnine binding to recombinant a1 GlyRs was displaced by taurine with KL = 43 6 mM. Propofol (50 mM) displaced [3H]strychnine binding to 83.4 8.7% and its cooperativity with taurine in displacement of [3H]strychnine binding to a1(R271L) GlyRs was around neutral: b = 1.1 0.4 (mean S.E.M. of three experiments). 3.2. Nanomolar displacing potencies of nortropeines are increased by mutation R271L Four nortropeines, substituted benzoates of nortropine were the most potent displacers of [3H]strychnine binding to native GlyRs of rat spinal cord (Maksay et al., 2004). Therefore they were also examined here with recombinant GlyRs. The nortropeines exerted concentration-dependent, full displacement of specific [3H]strychnine binding to both a1 and a1(R271L) GlyRs as demonstrated for nor-O-zatosetron in Fig. 4. This can be attributed to strong negative cooperativity with [3H]strychnine (a0 10). Alternatively, this can be reconciled with competitive displacement as well. Fitting according to a competitive model (Maksay et al., 2005) resulted in KA values which were not significantly different from those in Table 2 (data not shown). All esters displayed nanomolar affinities to a1 GlyRs (Table 2) similar to those observed for native spinal GlyRs (Maksay et al., 2004). It is remarkable that the mutation R271L enhanced the potency of the nortropeines (Table 2) as demonstrated by nor-O-zatosetron displacement in Fig. 4. Further, nortropisetron was more than two orders of magnitude stronger displacer than tropisetron (Maksay and
Fig. 4. Displacement of [3H]strychnine binding by nor-O-zatosetron to recombinant human a1 and a1(R271L) GlyRs transiently expressed in HEK 293 cells. Points are means S.E.M. of three to five experiments. Ratios of specific [3H]strychnine binding are plotted in the presence of nor-O-zatosetron (BSA) over control (BS).
Bı´ro´, 2002). It should be noted that fitting to Eq. (2) resulted in slope values of displacement m = 0.64 0.10 for nortropisetron and m = 0.70 0.18 for nor-O-zatosetron (means S.E.M. of three experiments) for a1(R271L) GlyRs (Table 2). As shown in Table 2, nor-O-zatosetron displayed the highest affinity with KA = 2.6 nM. Moreover, it was selective for a1 GlyRs as compared to its lower displacing potency on native 5-HT3Rs (Table 2). 4. Discussion 4.1. Propofol restores the coupling of glycine binding and gating impaired by the hyperekplexia mutation Propofol binding to a1 GlyRs showed low affinity and weak negative cooperativity to displace [3H]strychnine binding (a0 = 2.1). However, 10 mM glycine increased the affinity of Table 2 Displacing potencies of nortropeines on [3H]strychnine binding to recombinant human a1 and a1(R271L) GlyRs expressed in HEK 293 cells Esters of nortropine
Nor-O-Zatosetron 3-Methoxy-benzoate Benzoate 3-Chloro-benzoate Nortropisetron
KA (nM)a a1
a1(R271L)
a1bb
8.6 1.9 16.3 1.7 13.5 1.2 ND ND
2.6 0.3c,d 12.2 4.6 7.4 1.2c 46 21 9.7 2.0f
18.1 (5-HT3R: 56 nM)e 7.7e 10.4e 22.6e
Data are mean S.E.M. of three to five experiments. Slope values not shown were not significantly different from unity. ND: not determined. a Dissociation constants of nortropeine binding (KA) were determined via Eq. (2), with negative cooperativities of a0 10. For comparison, the displacing potency of nor-O-zatosetron is indicated for [3H]granisetron binding to rat cortical 5-HT3 receptors. b Native GlyRs of rat spinal cord. c p < 0.05, significantly lower than for a1 GlyRs in Mann–Whitney test. d m = 0.70 0.18. e Taken from Maksay et al. (2004). f m = 0.64 0.10.
G. Maksay et al. / Neurochemistry International 52 (2008) 235–240
propofol to KA 20 mM to native GlyRs (Maksay, 1998). Propofol was studied in a high, saturating concentration (0.3 mM) strongly enhancing agonist gating of recombinant a1 GlyRs expressed in Xenopus oocytes (O’Shea et al., 2004). It exerted positive cooperativity with glycine displacement of [3H]strychnine binding to a1 GlyRs (b = 0.26, Table 1), in agreement with potentiation of a1 GlyR-ionophore function (O’Shea et al., 2004). Propofol also decreased the slope factor n of displacement by glycine significantly for a1 GlyRs. The slope of displacement by taurine was also decreased by 0.1 mM propofol for native spinal GlyRs (Bı´ro´ and Maksay, 2004). Decreased slope values might be attributed to increased binding heterogeneity of GlyR agonists (i.e. decreased overlap of two displacement phases) due to propofol. The hyperekplexia mutation R271L did not change significantly propofol binding and its cooperativity with [3H]strychnine (Fig. 3 and Table 1). This means that this mutation did not affect the binding sites of propofol and [3H]strychnine, each located elsewhere (Fig. 1). In contrast, R271L decreased the displacing potency of glycine strongly via impaired gating, although the agonist sites are also far from R271. Remarkably, the positive cooperativity between glycine and propofol (b = 0.04) was much stronger with a1(R271L) than with wild GlyRs. This means that propofol restored the displacing potency of glycine impaired by the hyperekplexia mutation. Accordingly, propofol restored gating of hyperekplexia mutants by glycine, a partial agonist of a1(R271L) GlyRs (O’Shea et al., 2004). In contrast, the binding cooperativity was about neutral between propofol and taurine, that becomes an antagonist of a1(R271L) GlyRs (Rajendra et al., 1995). In conclusion, even though propofol potentiates GlyRs with low affinity and without selectivity, it might become a therapeutic lead since subanaesthetic doses of propofol alleviated the spontaneous and induced tremor episodes of transgenic mice carrying the hyperekplexia mutation R271Q of a1 GlyRs (O’Shea et al., 2004). 4.2. Nanomolar displacing potencies of nortropeines are increased by the hyperekplexia mutation Four representative nortropeines displayed nanomolar affinities to recombinant human a1 GlyRs and strong negative cooperativity to displace [3H]strychnine binding. Their binding parameters were similar to those for native spinal (a1b) GlyRs of adult rats (Maksay et al., 2004). It should be noted that glycine (10 mM) increased the displacing potencies of tropisetron and zatosetron on [3H]strychnine binding to native GlyRs (Maksay, 1998). The glycine-elicited binding changes correlate with the potentiating effects of zatosetron and tropisetron on GlyR function (Maksay and Bı´ro´, 2002). Further, the slope values of these two tropeines for native spinal GlyRs (Maksay, 1998) as well as those of their nor derivatives for a1(R271L) GlyRs here were below unity. Slope values lower than unity can be reconciled with binding heterogeneity. Most allosteric modulators of GlyRs such as anaesthetic alcohols, steroids, dihydropyridines and Zn2+ have bidirectional
239
activities (Laube et al., 2000; Hirzel et al., 2006). Similarly, submicromolar concentrations of several tropeines potentiate GlyRs, while micromolar tropeines inhibit them (ChesnoyMarchais, 1996). Accordingly, increasing tropeine concentrations shift the cooperativity with glycine from positive to negative (Maksay et al., 2004). Furthermore, (nor)tropeines having strong affinities usually have weak positive cooperativities with glycine (Maksay et al., 2004). Potentiation of GlyR function by nanomolar tropisetron requires low agonist occupancy (Supplisson and Chesnoy-Marchais, 2000). It can be concluded that high affinities and saturating concentrations of tropeines and agonists lead to displacement of [3H]strychnine and agonists. Consequently, inhibition of GlyRs by high-affinity nortropeines appears to mask potentiation. Differences in the acyl parts of nortropeines lead to structure–activity relationships of recombinant human a1 and native rat (a1b) GlyRs (Maksay et al., 2004) which are similar but not identical: nortropine 3-methoxy-benzoate was most active on native GlyRs, while least active on a1 GlyRs here. On the other hand, these differences in the acyl parts did not result in excessive changes in displacing potencies. Consequently, the high affinities can be attributed predominantly to binding of the alcohol part, the bicyclic nortropane skeleton, particularly to N-demethylation. It is remarkable that mutation R271L enhanced the affinity of the nortropeines, especially that of nor-O-zatosetron. Potentiation of the displacing potencies by R271L supports that the nortropeines interact with a1 GlyRs differently from the antagonist strychnine which is unaffected by R271L. It is probably not coincidental that mutation R271L potentiated both the inhibition by nortopeines as well as the positive cooperativity of propofol with glycine. Similarly, allosteric modulation of native GlyRs has been more efficacious for various partial agonists rather than for glycine and this has been observed not only for positive cooperativity with propofol and neuroactive steroids but also for negative modulators (Bı´ro´ and Maksay, 2004). These findings can be interpreted so that less efficacious channel opening, either due to improper agonist structure (partial agonists) or due to receptor function impaired by hyperekplexia mutations can be both potentiated and attenuated better by allosteric interactions. The KA value of 2.6 nM of nor-O-zatosetron is one of the strongest affinities reported for GlyRs including those of strychnine and R-5135, a steroidal antagonist of GABAA and GlyRs (Hunt and Clements-Jewery, 1981). Glycine ‘‘elicits’’ a similar high affinity, but minor displacement phase by tropisetron for [3H]strychnine binding to native spinal GlyRs (Maksay, 1998). Since nor-O-zatosetron is selective in comparison to 5-HT3 receptors, it might be used to develop a selective label, a radioligand of GlyRs and to localize the binding sites of GlyRs for tropeines. Granisetron is an antagonist of 5-HT3 -type serotonin receptors with a similar tropeine structure and high affinity. Homology modelling and point mutations have led to the characterization of two binding modes of granisetron in the interface of 5-HT3A subunits where 5-HT3 receptor antagonists are bound (Joshi et al., 2006). The extracellular interface of Cys-
240
G. Maksay et al. / Neurochemistry International 52 (2008) 235–240
loop receptors including GlyRs (see in Fig. 1) is a reasonable candidate for binding sites of tropeines of similar structure. Finally, the present binding test provides us with an in vitro binding method to evaluate drug candidates against neurological disorders associated with inherited mutations of GlyRs. Acknowledgements This work was supported by a grant K 62203 of the Hungarian Science Research Fund OTKA. Continuous support of professor Heinrich Betz (Frankfurt/M), stimulating discussions with Dr. Sean O’Shea and technical assistance of Mrs. Ilona Kawka (CRC, Budapest) are acknowledged. References Bali, M., Akabas, M.H., 2004. Defining the propofol binding site location on the GABAA receptor. Mol. Pharmcol. 65, 68–76. Belelli, D., Pistis, M., Peters, J.A., Lambert, J.J., 1999. The interaction of general anaesthetics and neurosteroids with GABAA and glycine receptors. Neurochem. Int. 34, 447–452. Betz, H., Laube, B., 2006. Glycine receptors: recent insights into their structural organisation and functional diversity. J. Neurochem. 97, 1600–1610. Bı´ro´, T., Maksay, G., 2004. Allosteric modulation of glycine receptors is more efficacious for partial rather than full agonists. Neurochem. Int. 44, 521– 527. Chesnoy-Marchais, D., 1996. Potentiation of chloride responses to glycine by three 5-HT3 antagonists in rat spinal cord. Br. J. Pharmacol. 118, 2115– 2125. Colquhoun, D., 1998. Binding, gating, affinity and efficacy: the interpretation of structure–activity relationships for agonists and the effects of mutating receptors. Br. J. Pharmacol. 125, 923–947. Flint, A.C., Liu, X., Kriegstein, A.R., 1998. Nonsynaptic glycine receptor activation during early neocortical development. Neuron 20, 43–53. Grudzinska, J., Schemm, R., Haeger, S., Nicke, A., Schmalzing, G., Betz, H., Laube, B., 2005. The beta subunit determines the ligand binding properties of synaptic glycine receptors. Neuron 45, 727–739. Hemmings, H.C., Akabas, M.H., Goldstein, P.A., Trudell, J.R., Orser, B.A., Harrison, N.L., 2005. Emerging molecular mechanisms of general anesthetic action. Trends Pharmacol. Sci. 26, 503–510. Hirzel, K., Mu¨ller, U., Latal, A.T., Hu¨lsmann, S., Grudzinska, J., Seeliger, M.W., Betz, H., Laube, B., 2006. Hyperekplexia phenotype of glycine receptor a1 subunit mutant mice identifies Zn2+ as an essential endogenous modulator of glycinergic neurotransmission. Neuron 52, 679–690. Hunt, P., Clements-Jewery, S., 1981. A steroid derivative, R5135, antagonizes the GABA–benzodiazepine receptor interaction. Neuropharmacology 20, 357–361.
Jakubı´k, J., Bacakova, L., El-Fakahany, E.E., Tucˇek, S., 1997. Positive cooperativity of acetylcholine and other agonist with allosteric ligands on muscarinic acetylcholine receptors. Mol. Pharmacol. 52, 172–179. Joshi, P.R., Suryanarayanan, A., Hazai, E., Schulte, M.K., Maksay, G., Bika´di, Zs., 2006. Interactions of granisetron with an agonist-free 5-HT3A receptor model. Biochemistry 45, 1099–1105. Langosch, D., Laube, B., Rundstro¨m, N., Schmieden, V., Bormann, J., Betz, H., 1994. Decreased agonist affinity and chloride conductance of mutant glycine receptors associated with human hereditary hyperekplexia. EMBO J. 13, 4223–4228. Laube, B., Betz, H., 1998. Purification and heterogenous expression of inhibitory glycine receptors. Methods Enzymol. 294, 260–273. Laube, B., Maksay, G., Schemm, R., Betz, H., 2000. Modulation of glycine receptor function: a novel approach for therapeutic intervention at inhibitory synapses? Trends Pharmacol. Sci. 23, 519–527. Lobo, I.A., Mascia, M.P., Trudell, J.R., Harris, R.A., 2004. Channel gating of the glycine receptor changes accessibility to residues implicated in receptor potentiation by alcohols and anesthetics. J. Biol. Chem. 279, 33919–33927. Lynch, J.W., 2004. Molecular structure and function of the glycine receptor chloride channel. Physiol. Rev. 84, 1051–1095. Maksay, G., 1998. Bidirectional allosteric modulation of strychnine-sensitive glycine receptors by tropeines and 5-HT3 receptor ligands. Neuropharmacology 37, 1633–1641. Maksay, G., Bı´ro´, T., 2002. Dual cooperative allosteric modulation of binding to ionotropic glycine receptors. Neuropharmacology 43, 1087–1098 (Erratum: 2003, 44, 843). Maksay, G., Laube, B., Betz, H., 1999. Selective blocking effects of tropisetron and atropine on recombinant glycine receptors. J. Neurochem. 73, 802–806. Maksay, G., Bı´ro´, T., Laube, B., 2002. Hyperekplexia mutation of glycine receptors: decreased gating efficacy with altered binding thermodynamics. Biochem. Pharmacol. 64, 285–288. Maksay, G., Nemes, P., Bı´ro´, T., 2004. Synthesis of tropeines and allosteric modulation of ionotropic glycine receptors. J. Med. Chem. 47, 6384–6391. Maksay, G., Bı´ro´, T., Bugovics, G., 2005. Allosteric modulation of 5-HT3 serotonin receptors. Eur. J. Pharmacol. 514, 17–24. O’Shea, S.M., Becker, L., Weiher, H., Betz, H., Laube, B., 2004. Propofol restores the function of ‘hyperekplexic’ mutant glycine receptors in Xenopus oocytes and mice. J. Neurosci. 24, 2322–2327. Rajendra, S., Lynch, J.W., Pierce, K.D., French, C.R., Barry, P.H., Schofield, P.R., 1995. Mutation of an arginine residue in the human glycine receptor transforms b-alanine and taurine from agonists into competitive antagonists. Neuron 14, 169–175. Shiang, R., Ryan, S.G., Zhu, Y.Z., Hahn, A.F., O’Connell, P., Wasmuth, J.J., 1993. Mutations in the a1 subunit of the inhibitory glycine receptor cause the dominant neurologic disorder, hyperekplexia. Nat. Genet. 5, 351–358. Supplisson, S., Chesnoy-Marchais, D., 2000. Glycine receptor b subunits play a critical role in potentiation of glycine responses by ICS-205,930. Mol. Pharmacol. 58, 763–770. Zhou, L., Chillag, K.L., Nigro, M.A., 2002. Hyperekplexia: a treatable neurogenetic disease. Brain Dev. 24, 669–674.