Introduction

Previous studies have suggested that the thiadiazole compound, SCH-202676 (N-(2,3-diphenyl-1,2,4-thiadiazol-5-(2H)-ylidene)methanamine), acts as an allosteric modulator of various G protein-coupled receptors (GPCRs). By definition, an allosteric modulator is a ligand that although silent per se, is capable of modulating (increase or decrease) the action of the 'orthosteric' ligand (agonist or antagonist) by binding to a distinct, that is, allosteric site on the receptor molecule (Kenakin, 2004). In the pioneering work by Fawzi and co-workers, it was demonstrated that low micromolar concentrations of SCH-202676 inhibited both agonist and antagonist radioligand binding to a number of structurally distinct GPCRs including human -, -, and -opioid, - and -adrenergic, muscarinic M₁ and M₂ acetylcholine and dopaminergic D₁ and D₂ receptors (Fawzi et al., 2001). It was further shown that SCH-202676 had no direct effect on heterotrimeric G protein activity but at similar concentrations, the compound totally abolished _2A-adrenoceptor-dependent [³⁵S]GTPS binding responses (Fawzi et al., 2001). These data were interpreted to suggest that SCH-202676 might directly bind a structural motif in the receptor molecule common to divergent members of the GPCR family.

Further studies on the muscarinic M₁ acetylcholine receptor led the authors to conclude that SCH-202676 may use a dual mode of ligand–receptor interaction involving both extra- and intracellular attachment points on the M₁ receptor that are distinct from the allosteric binding site recognized by prototypical modulators (Lanzafame & Christopoulos, 2004). Two additional studies that have been conducted using this compound concentrated on the interaction of SCH-202676 with adenosine and P2Y receptors and showed its divergent effects on the purinergic receptors. SCH-202676 inhibited radioligand binding to adenosine A₁, A_2A and A₃ receptors and affected their dissociation kinetics, but interestingly, it had no effect on the P2Y₁ receptor (Gao et al., 2004; van den Nieuwendijk et al., 2004).

From the viewpoint of drug discovery, allosteric modulators could present an alternative approach with potential benefits, as compared to the more traditional scheme aiming at development of drugs targeting the orthosteric site. Allosteric modulators might have a natural limitation of their effect, as the endogenous agonist needs to be present and would also show tissue selectivity, as the endogenous ligand concentrations may vary over different tissues. Although receptor-subtype specific allosteric modulators have been identified for several GPCRs (Kenakin, 2004), the promiscuous activity of SCH-202676 among diverse GPCRs raises an obvious question as to its molecular mechanism of action. In an effort to address this important question, we have used the functional approach of [³⁵S]GTPS autoradiography with brain cryostat sections to systematically study how SCH-202676 affects GPCR function, as this technique allows selective detection of receptor-dependent G protein activity simultaneously for several GPCRs in multiple brain regions (Laitinen, 2004). To complement the autoradiography approach, we used classical membrane [³⁵S]GTPS binding assays to provide a quantitative measure for the effects of the thiadiazole on the signalling of a panel of receptors coupling to the G_i-family of G proteins. This study demonstrates that SCH-202676 is a thiol-reactive compound rather than a true allosteric modulator, and disrupts GPCR signalling in a dithiotreitol (DTT)-sensitive manner in [³⁵S]GTPS-based functional assays. While this manuscript was in preparation, another study reached a similar conclusion by demonstrating that SCH-202676 and related thiadiazole compounds elicit their modulatory actions on adenosine A₁ receptor via sulphydryl modification (Göblyös et al., 2005).

Methods

Animals and preparation of cryostat sections

All animal experiments were approved by the local ethics committee and were conducted in vitro using brain sections from 4-week-old male Wistar rats. The animals lived in a 12-h light, 12-h dark cycle (lights on 0700) with water and food ad libitum. The rats were killed by decapitation, 8 h after lights on, whole brains were dissected out, dipped in isopentane on dry ice and stored at -80°C. Coronal or sagittal sections (20 m thick) were cut at -14°C using a Leica cryostat. In some experiments (illustrated in Figure 2), spleen sections (20 m thick) from four-week-old animals were also used. Tissue was thaw-mounted onto SuperFrost^* Plus slides (Menzel-Gläser, Germany) at 20°C. The sections were stored at -80°C for up to 8 months without any apparent loss of binding responses.

Figure 2.

In the absence of DTT, SCH-202676 elicits nonspecific binding responses in [³⁵S]GTPS-based G protein activation assays. [³⁵S]GTPS autoradiography of rat brain (sagittal plane) and spleen sections was conducted using a three-step protocol with ADA (1 U ml^-1) present throughout steps 2 and 3, as detailed in the Methods section. SCH-202676 was present at the indicated concentrations during the [³⁵S]GTPS labeling (step 3). The GPCR agonists 2-chloroadenosine (2ClAdo), lysophosphatidic acid (LPA), and CP-55,940 were present during step 3 to stimulate adenosine A₁, lysophosphatidic acid LPA₁ and cannabinoid CB₁ receptors, respectively. In the control panel (left), the anatomical brain loci where the receptor agonists typically activate G proteins at this sagittal plane are indicated. Note SCH-202676-evoked nonspecific and dose-dependent increase in [³⁵S]GTPS binding that is not restricted to the tissue sections but is apparent to glass slides (background) as well. Note also severe blunting of agonist-stimulated responses in the presence of 5 10^-6 M SCH-202676 and their total absence with 10^-5 M thiadiazole concentration. Abbreviations: cc, corpus callosum; Cbm, cerebellar molecular layer; Cbw, cerebellar white matter; Cx, cerebral cortex; fi, fimbria of the hippocampus; Hip, hippocampus; GP, globus pallidus; Str, striatum. Scale bars=2 mm.

Full figure and legend (134K)

Chemicals

Adenosine deaminase (ADA) was obtained from Roche Molecular Biochemicals (Ingelheim, Germany) whereas BSA (fatty acid free), DTT, DPCPX, GDP, GTPS, lysophosphatidic acid (LPA), and 2-methylthio ADP (2MeSADP) were from Sigma-RBI (St Louis, MO, U.S.A.). CP-55,940 and SCH-202676 were purchased from Tocris Cookson Ltd. (Bristol, U.K.). [³⁵S]GTPS (initial specific activity 1250 Ci mmol^-1) was from NEN (Boston, MA, U.S.A.). The stock solution was diluted 1 : 50 in 10 mM Tris-HCl, pH 7.4 containing 10 mM DTT and stored as single-use aliquots at -80°C. All other chemicals were of highest purity available.

[³⁵S]GTPS autoradiography

The assay was conducted under optimized conditions, where noise due to tonic adenosine A₁ receptor activity has been eliminated (Laitinen, 1999; Laitinen et al., 2001). Experiments were conducted in light-protected chambers. Briefly, the assay consisted of preincubation for 20 min at 20°C in buffer A (50 mM Tris-HCl, pH 7.4, 1 mM EDTA, 100 mM NaCl, 5 mM MgCl₂), followed by GDP loading for 1 h at 20°C in buffer A, containing additionally 2 mM GDP and DPCPX (10^-6 M) or ADA (1 U ml^-1) to eliminate tonic adenosine A₁ receptor activity. For agonist-stimulated [³⁵S]GTPS binding, sections were incubated for 90 min at 20°C in buffer A, containing additionally 80 pM [³⁵S]GTPS, 2 mM GDP, DPCPX (10^-6 M) or ADA (1 U ml^-1), plus the receptor agonists in alone or in combination with SCH-202676 (10^-7–10^-5 M) and DTT (1 mM), as detailed in the Results section. Nonspecific binding (Nsb) was determined in the presence of 10 M GTPS. The sections were washed twice at 0°C for 5 min in washing buffer (50 mM Tris-HCl, 5 mM MgCl₂, pH 7.4), rinsed in ice-cold deionised water for 30 s, air dried and apposed to Biomax™ MR film (Kodak) for 6–11 days. Autoradiography films were scanned at 150 dpi resolution to generate digital images.

[³⁵S]GTPS membrane binding assays

The incubations were carried out for 90 min at 25°C under constant shaking essentially as previously described (Kurkinen et al., 1997; Savinainen et al., 2003). The final concentrations of the components in the binding reaction were 50 mM Tris-HCl (pH 7.4), 1 mM EDTA, 100 mM NaCl, 5 mM MgCl₂, 0.5% BSA, 10 M GDP, 0.5 U ml^-1 ADA, 150 pM [³⁵S]GTPS and varying concentrations of SCH-202676 (10^-6, 5 10^-6 or 10^-5 M) and DTT (1 mM), as detailed in the Results sections. Nsb was defined using 10 M GTPS. Reaction was quenched by the addition of 4 ml ice-cold wash buffer (50 mM Tris-HCl, pH 7.4 and 5 mM MgCl₂) followed by rapid filtration through Whatman GF/B glass fibre filters (Whatman, Maidstone, U.K.) and two additional 4 ml washes with the buffer. Radioactivity in filters was counted with Wallac Rackbeta liquid scintillation counter (Wallac, Turku, Finland).

Nuclear magnetic resonance

¹H-NMR spectra were recorded at 300 K on a Bruker Avance DRX 500 spectrometer equipped with a normal 2.5 mm BBI probe. The samples were evaporated to dryness and dissolved in 130 l of D₂O and ¹H-NMR spectra were measured in a 2.5 mm microtube.

Data analysis

[³⁵S]GTPS-membrane binding data were analysed with GraphPad Prism software (GraphPad, San Diego, CA, U.S.A.). Assays were run with duplicate samples and repeated independently at least three times. Statistical analyses were made with one-way analysis of variance (ANOVA) followed by Tukey's multiple comparison test with P<0.05 considered as statistically significant.

Results

Previous studied indicate that SCH-202676 exhibits an extraordinary steep and narrow dose–response range (typical effective concentration range 10^-7–10^-5 M) in modulating radioligand binding to GPCRs (Fawzi et al., 2001; Gao et al., 2004; Lanzafame & Christopoulos, 2004; van den Nieuwendijk et al., 2004). Therefore, our initial studies were set up to clarify whether SCH-202676 at this concentration range affects receptor-dependent G protein activation, as assessed in rat brain sections using the optimised standard protocol for [³⁵S]GTPS autoradiography. We focused the initial experiments on four G_i-linked receptors, namely the purinergic P2Y₁₂, the lysophophatidic acid LPA₁, the cannabinoid CB₁ and the sphingosine 1-phosphate (S1P) receptors, as G protein activity upon stimulation of these receptors has been previously characterized using the autoradiography approach and each receptor shows a unique anatomical distribution pattern in the developing rat brain (Laitinen, 2004). As depicted in Figure 1, the P2Y receptor agonist 2MeSADP activated G proteins both in gray and white matter regions, producing a heterogeneous activity pattern as described earlier (Laitinen et al., 2001; Vasiljev et al., 2003). Our previous work has established that the 2MeSADP-stimulated [³⁵S]GTPS binding responses in rat brain sections are mediated by a P2Y receptor subtype that pharmacologically corresponds to P2Y₁₂ (Laitinen et al., 2001; Vasiljev et al., 2003). As further evident from Figure 1, LPA-stimulated G_i protein activity was largely restricted to the myelinating white matter tracts, such as the corpus callosum (cc), the fimbria of the hippocampus (fi) and the optic tract (opt), closely reflecting the anatomical distribution of the LPA₁ receptor subtype (Waeber & Chiu, 1999; Laitinen, 2004). In contrast, CB₁ and S1P receptor responses show widespread distribution mainly in the grey matter regions (Waeber & Chiu, 1999; Laitinen, 2004), such regions visible in Figure 1 include, for example, the cerebral cortex (Cx) and the hippocampus (Hip). In contrast to our expectations, SCH-202676 at concentrations previously shown to inhibit radioligand binding to various GPCRs (10^-7–10^-5 M) and to totally abolish _2A-adrenoceptor-mediated [³⁵S]GTPS binding responses in CHO cell membranes (Fawzi et al., 2001) had no apparent effect on the G protein activity mediated by the tested GPCRs (Figure 1). These experiments also indicated that nonspecific and basal [³⁵S]GTPS binding was unaffected by the thiadiazole (Figure 1). Additional studies (not shown) further indicated that SCH-202676 (10^-5 M) had no detectable effect on G protein activity mediated by an additional set of G_i-coupled receptors, including the ₂-adrenergic, adenosine A₁, and muscarinic M₂/M₄ receptors. This negative outcome was rather unexpected in light of previous findings demonstrating the promiscuous ability of SCH-202676 to modulate various GPCRs.

Figure 1.

Under the routine conditions of [³⁵S]GTPS autoradiography, SCH-202676 has no effect on receptor-dependent G protein activity in rat brain cryostat sections. [³⁵S]GTPS autoradiography of coronal brain sections was conducted using a three-step protocol with 10^-6 M DPCPX present throughout steps 2 and 3, as detailed in the Methods section. SCH-202676 was present at the indicated concentrations together with 1 mM DTT during the [³⁵S]GTPS labeling (step 3). The GPCR agonists 2-methylthio ADP (2MeSADP), lysophosphatidic acid (LPA), CP-55,940 and sphingosine 1-phosphate (S1P) were present during step 3 to stimulate purinergic P2Y₁₂, lysophosphatidic acid LPA₁, cannabinoid CB₁ and sphingosine 1-phosphate S1P₁ receptors, respectively. In the control panel (left), the anatomical loci where the receptor agonists typically activate G proteins at this coronal plane are indicated. Note the apparent lack of effect of SCH-202676 on nonspecific, basal or agonist-stimulated responses. Abbreviations: cc, corpus callosum; Cx, cerebral cortex; EP, entopeduncular nucleus fi; fimbria of the hippocampus; Hip, hippocampus, Hy, hypothalamus; opt, optic tract. Scale bar=2 mm.

Full figure and legend (118K)

To resolve the reason behind this discrepant outcome, we modified the standard incubation protocol by omitting DTT, a reducing agent that is frequently included in [³⁵S]GTPS binding assays to reduce the levels of nonspecific and basal binding (Happe et al., 2001). As anticipated, this modification resulted in a dramatically different outcome with respect to the behaviour of SCH-202676. In the absence of DTT, the thiadiazole (5 10^-6 and 10^-5 M) caused an overall, nonspecific, and dose-dependent increase in [³⁵S]GTPS labelling, not only to brain sections but to glass slides as well (Figure 2). Such a behaviour was not restricted to rat brain sections, as the thiadiazole similarly compromised specific [³⁵S]GTPS labelling to sections of mouse brain (data not shown) and rat spleen (Figure 2, bottom row), as well as to rat brain membrane preparations (see below). It was notable that under the DTT-free conditions, all agonist-stimulated [³⁵S]GTPS binding responses were blunted with 5 10^-6 M SCH-202676 and practically absent with the higher (10^-5 M) thiadiazole concentration (Figure 2). Additional studies indicated that SCH-202676 at only slightly smaller concentrations (10^-6 and 2 10^-6 M) failed to affect basal or receptor-dependent responses in brain sections (data not shown). These data indicate, in agreement with previous radioligand binding studies (Fawzi et al., 2001; Gao et al., 2004; Lanzafame & Christopoulos, 2004; van den Nieuwendijk et al., 2004), that SCH-202676 exhibits an extraordinary narrow effective dose-range also in [³⁵S]GTPS autoradiography assessing GPCR function. Prolonging the incubation time of SCH-202676 (10^-5 M) with tissue further increased the nonspecific [³⁵S]GTPS labelling of brain sections (Figure 3), indicating that the compound likely reacted with tissue components as present during the binding assay. Importantly, the thiadiazole action was fully reversed upon addition of 1 mM DTT (Figure 3), revealing that sulphydryl-sensitive mechanisms were responsible for the [³⁵S]GTPS binding-interfering action of SCH-202676.

Figure 3.

Time-dependent and fully DTT-reversible effects of SCH-202676 on [³⁵S]GTPS-based G protein activation assays with brain cryostat sections. [³⁵S]GTPS autoradiography of rat brain sagittal sections was conducted using a three-step protocol with 10^-6 M DPCPX present throughout steps 2 and 3, as detailed in the Methods section. SCH-202676 was present at the indicated concentrations either throughout steps 2 and 3 (60+90 min) or only during the [³⁵S]GTPS labeling (step3, 90 min). Where indicated, DTT (1 mM) was included during step3. The GPCR agonists 2-methylthio ADP (2MeSADP), lysophosphatidic acid (LPA), and CP-55,940 were present during step 3 to stimulate purinergic P2Y₁₂, lysophosphatidic acid LPA₁, and cannabinoid CB₁ receptors. In the control panel (left), the anatomical loci where the receptor agonists typically activate G proteins at this sagittal plane are indicated. Note the time-dependent increase in the SCH-202676-evoked nonspecific [³⁵S]GTPS labeling. Note also complete reversal of the thiadiazole action in the presence of DTT. Abbreviations: cc, corpus callosum; Cbm, cerebellar molecular layer; Cbw, cerebellar white matter; Cx, cerebral cortex; fi, fimbria of the hippocampus; Hip, hippocampus; Str, striatum; SN, substantia nigra. Scale bar=2 mm.

Full figure and legend (167K)

To rule out the possibility that the above findings were an artefact inherent to the autoradiography approach, classical membrane [³⁵S]GTPS binding assays were conducted in parallel. In accordance with the autoradiography results, SCH-202676 when tested under DTT-free conditions and at concentrations of 10^-6, 5 10^-6 and 10^-5 M, caused a nonspecific and dose-dependent increase in [³⁵S]GTPS labelling to rat forebrain membranes (Figure 4). Notably, Nsb, which under the control conditions represents approximately 0.3% of total added radioactivity, was increased up to 10-fold in the presence of 10^-5 M SCH-202676 (Figure 4, upper left). Similarly, total membrane-associated [³⁵S]GTPS binding under the basal condition was significantly increased with the highest thiadiazole concentration (Figure 4, upper middle), whereas specific binding, obtained after subtracting nonspecific from total binding, was not significantly affected (Figure 4, upper right). When DTT (1 mM) was included in the binding cocktail, 10^-5 M SCH-202676 had no effect on the above-mentioned parameters. Furthermore and in full accordance with the autoradiography results, agonist-stimulated [³⁵S]GTPS binding responses under the DTT-free assay conditions were variably blunted in the presence of 5 10^-5 M thiadiazole, and totally abolished at 10^-5 M concentration of the compound (Figure 4, bottom panel). In the presence of DTT (1 mM), SCH-202676 (10^-5 M) had no effect on the agonist potency or efficacy at the four studied receptors (Figure 4 and Table 1).

Figure 4.

SCH-202676 elicits nonspecific and fully DTT-sensitive [³⁵S]GTPS binding responses in filtration-based G protein activation assays. Rat forebrain membranes (5 g per tube) were incubated for 90 min at 25°C in a final volume of 0.4 ml with the indicated combinations of compounds in the presence of 0.5 U ml^-1 ADA, as detailed in the Methods section. Where indicated, incubations contained 1 mM DTT. The GPCR agonists 2-chloroadenosine (2ClAdo), CP-55,940, carbachol and lysophosphatidic acid (LPA) were used to stimulate adenosine A₁, cannabinoid CB₁, muscarinic M₂/M₄ and lysophosphatidic acid LPA₁ receptors, respectively. The data represent the means.e. from two to three independent experiments performed in duplicate. An asterix denotes a statistically significant difference (P<0.05; ANOVA followed by Tukey's multiple comparison) as compared to the respective control condition.

Full figure and legend (93K)

Table 1 - In the presence of DTT SCH-202676 has no effect on agonist potency of efficacy in rat forebrain [³⁵S]GTPS binding assays of various G_i-coupled receptors.

Full table

To shed further light on the DTT-sensitive behaviour of SCH-202676, we evaluated the compound stability using ¹H-NMR (Figure 5). The reference spectrum (spectrum f, SCH-202676 in DMSO) showed the expected chemical shifts at 7.6–7.3 p.p.m. due to aromatic hydrogens, indicating chemical integrity of the compound. Identical spectrum was recorded after storage for several days at room temperature (data not shown), or after incubations of the compound for 90 min in the buffer system of our study (spectrum b). This analysis indicated that the thiadiazole was chemically stable in the buffer system used in [³⁵S]GTPS binding assays. In contrast, addition of DTT to the buffer resulted in dramatic changes in the aromatic hydrogen signals at 7.6–7.3 p.p.m. (compare spectrum c against that of b and f), indicating that the presence of thiol groups per se induced chemical decomposition of the thiadiazole molecule. Similarly, incubation of SCH-202676 together with brain sections (Figure 5a) resulted in notable changes in the chemical shifts at 7.6–7.3 p.p.m. Interestingly, no further modification was evident when incubations with brain sections contained additionally DTT (compare spectrum d against a). This structural analysis indicates that SCH-202676 undergoes chemical decomposition after reacting with the reducing agent DTT or with brain tissue.

Figure 5.

SCH-202676 undergoes structural changes after incubation with the reducing agent DTT with brain tissue. The thiadiazole (6 10^-5 M) was incubated at 20°C for 90 min in Tris-based assay buffer (50 mM Tris-HCl, pH 7.4, 1 mM EDTA, 100 mM NaCl, 5 mM MgCl₂) together with DTT (1 mM) or with brain sections, as indicated in traces a–e. The postincubation supernatant was collected and ¹H-NMR spectra determined as detailed in the Methods section. Trace f represents the ¹H-NMR spectrum of SCH-202676 in DMSO. Note chemical stability of SCH-202676 in the aqueous buffer (b) and structural changes (altered ¹H NMR spectra) after incubation with DTT (c) or brain sections (a).

Full figure and legend (50K)

Discussion

The present study demonstrated that in incubations containing no DTT, the 'allosteric modulator' SCH-202676 is a thiol-reactive substance and elicits nonspecific effects in [³⁵S]GTPS binding-based G protein activation assays, thereby challenging previous conclusions regarding the compound's ability to allosterically modulate GPCR function. The nonspecific behaviour was fully reversed upon addition of DTT, indicating the involvement of a thiol-based mechanism of action. Indeed, in routine incubations containing DTT, SCH-202676 had no detectable effect on agonist-stimulated G protein activity, as determined for various GPCRs, arguing against true allosteric modulatory action. Structural analysis using ¹H NMR revealed that SCH-202676 was prone to decomposition after reacting with the reducing agent DTT or with brain tissue. While our study was in progress, another group independently demonstrated that SCH-202676 and related thiadiazoles elicit their modulatory action on adenosine A₁ receptor via sulphydryl modification rather than via true allosteric mechanism (Göblyös et al., 2005). Using several bioanalytical approaches, this study elegantly demonstrated that the thiadiazoles were reduced into their thiourea precursors by thiol groups of cysteine residues, thus explaining the promiscuous activity towards divergent members of the GPCR family. The present findings are in good agreement with the observations of Göblyös and co-workers.

Collectively, the seminal observations made in these two studies can likely explain the promiscuous action of SCH-202676 towards various GPCRs reported in previous studies (Fawzi et al., 2001; Gao et al., 2004; Lanzafame & Christopoulos, 2004; van den Nieuwendijk et al., 2004). The pioneering work demonstrated that SCH-202676 had no direct effect on heterotrimeric G protein activity as assessed by [³⁵S]GTPS binding to purified recombinant G_o subunit or G-stimulated ADP-ribosylation of G_o by pertussis toxin (Fawzi et al., 2001). Quite opposite, SCH-202676 at similar concentrations, totally inhibited _2A-adrenoceptor-dependent [³⁵S]GTPS binding responses in the same study. Curiously, the experiments assessing direct action on G proteins were conducted in the presence of millimolar concentrations of DTT whereas the [³⁵S]GTPS binding assay measuring _2A-adrenoceptor function contained no DTT. In retrospect, this methodological deviation was actually the first to catch our attention, leading to modify the standard assay protocol so that DDT was omitted. Indeed, such a modification was already successfully applied in our recent study characterizing the highly receptor-specific modulation of GPCR signalling by S-nitrosothiols (Kokkola et al., 2005). As presently shown, attempts to address the allosteric action of SCH-202676 with [³⁵S]GTPS-based G protein activation assays not containing DTT are severely compromised due to the nonspecific, sulphydryl-involving action of the compound. On the other hand, the reported lack of direct activity of SCH-202676 on G proteins (Fawzi et al., 2001) can be fully explained by the inclusion of DTT in those assays.

Previous studies indicate that thiol-reactive agents can have profound effects on G_i-mediated receptor signalling, as assesses using [³⁵S]GTPS autoradiography. For example, treatment of brain sections with N-ethylmaleimide (NEM), an irreversible thiol-alkylating reagent targeting the C-terminal cysteine residue in the subunits of G_i family members, results in total loss of receptor-stimulated G protein activity (Laitinen, 2004). On the other hand, S-nitrosothiols were shown to modulate GPCR signalling via reversible mechanisms probably involving S-nitrosylation (Kokkola et al., 2005). In contrast to the inhibitory action of NEM, S-nitrosothiols were capable of very targeted regulation of GPCR signalling, as they potentiated muscarinic M₂/M₄ receptor responses, inhibited P2Y₁₂, LPA₁, and CB₁ receptors, and only marginally affected several other receptors (Kokkola et al., 2005). The behaviour of SCH-202676 is clearly different from those of the above-mentioned thiol agents, as the thiadiazole elicited a generalized nonspecific inhibitory action in [³⁵S]GTPS binding assays (evident in autoradiography experiments throughout the glass slides and tissue sections, see Figures 2 and 3). Importantly, DTT could fully reverse this effect. NMR analysis indicated that the presence of sulphydryl groups per se (as was the case with DTT) was sufficient to induce chemical decomposition of SCH-202676. We made no further attempts to identify the chemical structure of the thiol-modified product but Göblyös and co-workers (2005) demonstrated that SCH-202676 and related thiadiazoles were reduced into their thiourea precursors by thiol groups of cysteine residues. This study further showed that the thiadiazoles underwent structural changes only after reaction with cysteine, but not other amino acids.

Although not always evident from the original methods description, there is a good reason to assume that the previous radioligand binding assays characterizing the allosteric action of SCH-202676 on various GPCRs were actually conducted in the absence of DTT (Fawzi et al., 2001; Lanzafame & Christopoulos, 2004; Gao et al., 2004, van den Nieuwendijk et al., 2004). As radioligand binding to most GPCRs appears to be sensitive to sulphydryl modification, the allosteric effect of SCH-202676 in these assays is thus also likely best explained by mechanisms involving thiol groups. It is noteworthy that among the P2Y receptors, sensitivity to sulphydryl modification is characteristic for the P2Y₁₂ receptors but not the P2Y₁ receptors (Macfarlane et al., 1983, Mills, 1996; Savi et al., 2001; Ding et al., 2003). Interestingly, P2Y₁ is so far also the only receptor that in radioligand binding studies was fully resistant to the actions of SCH-202676 (Gao et al., 2004). Thus, SCH-202676 is best described as a sulphydryl-reactive compound rather than a true allosteric modulator.

References

1. DING, Z., KIM, S., DORSAM, R.T., JIN, J. & KUNAPULI, S.P. (2003). Inactivation of the human P2Y₁₂ receptor by thiol reagents requires interaction with both extracellular cysteine residues, Cys17 and Cys270. Blood, 101, 3908–3914. | Article | PubMed | ISI | ChemPort |
2. FAWZI, A.B., MACDONALD, D., BENBOW, L.L., SMITH-TORHAN, A., ZHANG, H., WEIG, B.C., HO, G., TULSHIAN, D., LINDER, M.E. & GRAZIANO, M.P. (2001). SCH-202676: an allosteric modulator of both agonist and antagonist binding to G protein-coupled receptors. Mol. Pharmacol., 59, 30–37. | PubMed | ISI | ChemPort |
3. GAO, Z.G., GROSS, A.S. & JACOBSON, K.A. (2004). Effects of the allosteric modulator SCH-202676 on adenosine and P2Y receptors. Life Sci., 74, 3173–3180. | Article | PubMed | ISI | ChemPort |
4. GOBLYOS, A., DE VRIES, H., BRUSSEE, J. & IJZERMAN, A.P. (2005). Synthesis and biological evaluation of a new series of 2,3,5-substituted [1,2,4]-thiadiazoles as modulators of adenosine A1 receptors and their molecular mechanism of action. J. Med. Chem., 48, 1145–1151. | Article | PubMed | ISI | ChemPort |
5. HAPPE, H.K., BYLUND, D.B. & MURRIN, L.C. (2001). Agonist-stimulated [³⁵S]GTPS autoradiography: optimization for high sensitivity. Eur. J. Pharmacol., 422, 1–13. | Article | PubMed | ISI | ChemPort |
6. KENAKIN, T. (2004). Allosteric modulators: the new generation of receptor antagonist. Mol. Interv., 4, 222–229. | Article | PubMed | ISI | ChemPort |
7. KOKKOLA, T., SAVINAINEN, J.R., MÖNKKÖNEN, K.S., DURÁN RETAMAL, M. & LAITINEN, J.T. (2005). S-Nitrosothiols modulate G protein-coupled receptor signaling in a reversible and highly receptor-specific manner. BMC Cell Biol., 6, 21. | Article | PubMed | ChemPort |
8. KURKINEN, K.M.A., KOISTINAHO, J. & LAITINEN, J.T. (1997). [-³⁵S]GTP autoradiography allows region-specific detection of muscarinic receptor-dependent G protein activation in the chick optic tectum. Brain Res., 769, 21–28. | Article | PubMed | ISI | ChemPort |
9. LAITINEN, J.T. (1999). Selective detection of adenosine A₁ receptor-dependent G-protein activity in basal and stimulated conditions of rat brain [³⁵S]guanosine 5'-(-thio)triphosphate autoradiography. Neuroscience, 90, 1265–1279. | Article | PubMed | ISI | ChemPort |
10. LAITINEN, J.T. (2004). [³⁵S]GTPS autoradiography: a powerful functional approach with expanding potential for neuropharmacological studies on receptors coupled to G_i family of G proteins. Curr. Neuropharmacol., 2, 191–206. | ChemPort |
11. LAITINEN, J.T., URI, A., RAIDARU, G. & MIETTINEN, R. (2001). [³⁵S]GTPS autoradiography reveals a wide distribution of G_i/o-linked ADP receptors in the nervous system: close similarities with the platelet P2Y_ADP receptor. J. Neurochem., 77, 505–518. | Article | PubMed | ISI | ChemPort |
12. LANZAFAME, A. & CHRISTOPOULOS, A. (2004). Investigation of the interaction of a putative allosteric modulator, N-(2,3-diphenyl-1,2,4-thiadiazole-5-(2H)-ylidene) methanamine hydrobromide (SCH-202676), with M₁ muscarinic acetylcholine receptors. J. Pharmacol. Exp. Ther., 308, 830–837. | PubMed | ISI | ChemPort |
13. MACFARLANE, D.E., SRIVASTAVA, P.C. & MILLS, D.C.B. (1983). 2-Methylthio-adenosine[-³²P]diphosphate. An agonist and radioligand for the receptor that inhibits the accumulation of cyclic AMP in intact blood platelets. J. Clin. Invest., 71, 420–428. | PubMed | ISI | ChemPort |
14. MILLS, D.C.B. (1996). ADP receptors on platelets. Thromb. Haemost., 76, 835–856. | PubMed | ISI | ChemPort |
15. SAVI, P., LABOURET, C., DELESQUE, N., GUETTE, F., LUPKER, J. & HERBERT, J.M. (2001). P2Y₁₂, a new platelet ADP receptor, target of clopidogrel. Biochem. Biophys. Res. Commun., 283, 379–383. | Article | PubMed | ISI | ChemPort |
16. SAVINAINEN, J.R., SAARIO, S.M., NIEMI, R., JÄRVINEN, T. & LAITINEN, J.T. (2003). An optimised approach to study endocannabinoid signalling: evidence against constitutive activity of rat brain adenosine A₁ and cannabinoid CB₁ receptors. Br. J. Pharmacol., 140, 1451–1459. | Article | PubMed | ISI | ChemPort |
17. VAN DEN NIEUWENDIJK, A.M., PIETRA, D., HEITMAN, L., GOBLYOS, A. & IJZERMAN, A.P. (2004). Synthesis and biological evaluation of 2,3,5-substituted [1,2,4]thiadiazoles as allosteric modulators of adenosine receptors. J. Med. Chem., 47, 663–672. | PubMed | ISI | ChemPort |
18. VASILJEV, K.S., URI, A. & LAITINEN, J.T. (2003). 2-Alkylthio-substituted platelet P2Y₁₂ receptor antagonists reveal pharmacological identity between the rat brain G_i-linked ADP receptor and P2Y₁₂. Neuropharmacology, 45, 145–154. | Article | PubMed | ISI | ChemPort |
19. WAEBER, C. & CHIU, M.L. (1999). In vitro autoradiographic visualization of guanosine-5'-O-(3-[³⁵S]thio)triphosphate binding stimulated by sphingosine 1-phosphate and lysophosphatidic acid. J. Neurochem., 73, 1212–1221. | Article | PubMed | ISI | ChemPort |

Acknowledgements

Anna M. Lewandowicz is a medical student from Lodz, Poland and participated in this study as an IFMSA (International Federation of Medical Students' Association) exchange student. We are thankful to Mrs Taina Vihavainen and Mrs Taija Vaarala for skillful technical assistance and to Mrs Sandra Perez for conducting some of the membrane [³⁵S]GTPS binding assays. Mr Pasi Soininen is acknowledged for his help in generating the NMR figure.