Emerging evidence has suggested that inhibitory glycine receptors (GlyRs) are an important molecular target in the treatment of numerous neurological disorders. Rhizoma curcumae is a medicinal plant with positive neurological effects. In this study, we showed that curcumol, a major bioactive component of R. curcumae, reversibly and concentration-dependently inhibited the glycine-activated current (IGly) in cultured rat hippocampal neurons. The inhibitory effect was neither voltage- nor agonist concentration-dependent. Moreover, curcumol selectively inhibited homomeric α2-containing, but not α1- or α3-containing, GlyRs. The addition of β subunit conferred the curcumol sensitivity of α3-containing, but not α1-containing, GlyRs. Site-directed mutagenesis analysis revealed that a threonine at position 59 of the α2 subunit is critical for the susceptibility of GlyRs to curcumol-mediated inhibition. Furthermore, paralleling a decline of α2 subunit expression during spinal cord development, the degree of IGly inhibition by curcumol decreased with prolonged culture of rat spinal dorsal horn neurons. Taken together, our results suggest that the GlyRs are novel molecular targets of curcumol, which may underlie its pharmaceutical effects in the central nervous system.
The strychnine-sensitive glycine receptors (GlyRs) are ligand-gated chloride channels that mediate prominent inhibition in spinal cord and brain stem as well as in other regions of the central nervous system (CNS) (Malosio et al., 1991; Lynch, 2004). Five different subunits of GlyRs have been identified in mammals, including four α subunits (α1-α4) and one β subunit, which form two different configurations of functional receptors, either homopentamers composed of five identical α subunits or heteropentamers composed of two α subunits and three β subunits (Grudzinska et al., 2005). The different GlyR subunits exhibit uneven regional and developmental distributions (Malosio et al., 1991) and form channels that differ subtly with respect to their basic functional and pharmacological properties. In the hippocampus, GlyRs generally play a tonic inhibitory role in accordance with their extrasynaptic distribution (Xu and Gong, 2010), where the levels of α2 and β subunits are relatively high and that of α3 subunit is low (Malosio et al., 1991; Eichler et al., 2009). In the spinal cord, GlyRs mediate fast synaptic inhibition and are implicated in the control of motor rhythm generation, coordination of spinal reflex responses, and processing of sensory signals (Legendre, 2001). It is noteworthy that there is a developmental switch from α2 to α1β of the GlyR subunit composition during spinal cord development (Betz and Laube, 2006). In addition, it has been shown that the α3-containing GlyRs are an essential target for spinal inflammatory pain sensitization in superficial layers of the spinal cord dorsal horn (Harvey et al., 2004). Furthermore, GlyRs have emerged as a potential pharmacological target for therapeutic interventions of neurological diseases because they have fewer side effects than other central receptors (Xu and Gong, 2010). Thus, agents that cause subunit-specific modulation of the GlyRs are of particular interest.
Among the natural plants, Rhizoma curcumae (rhizome of Curcuma; Ezhu) has been used as a condiment and home remedy in China for thousands of years. The chemical basis of R. curcumae is actively being investigated, and its main constituents are typically classified into the nonvolatile phenolic pigment and the volatile essential oil. The former includes curcumin [1,7-bis(4-hydroxy-3-methoxyphenyl)-1,6-heptadiene-3,5-dione] and its analogs, whereas the latter mainly includes two sesquiterpene compounds, curcumol [(3S,5S,6S,8aS)-3-methyl-8-methylidene-5-(propan-2-yl) octahydro-6H-3a,6-epoxyazulen-6-ol] and curdione [(3S,6E,10S)-6,10-dimethyl-3-propan-2-ylcyclodec-6-ene-1,4-dione] (Xia et al., 2005; Deng et al., 2006). Generally, the volatile essential oil is considered as the active constituent mediating many pharmacological properties of the plant such as cognition enhancement (Sun et al., 2008), neuroprotection (Dohare et al., 2008), and antiepilepsy efficacy (Wang and Zhao, 2004). Among all of the constituents, curcumol is commonly used as the quality control for the effectiveness of essential oil in R. curcumae (Deng et al., 2006). In contrast to extensive studies on the chemical essence of this traditional Chinese medicine, the molecular targets for its various constituents have not been identified. In this study, we characterized subunit-specific modulation of GlyRs by curcumol and further explored the underlying molecular mechanism for its action. Our findings may add a new dimension to further characterizations of the pharmacological properties of curcumol in the CNS.
Materials and Methods
Primary Neuronal Cultures.
Animals used in the present study were obtained from Shanghai Slac Laboratory Animal Company Limited (Shanghai, China) and maintained under a 12 hour light/12 hour dark cycle at 22–25°C. All experiments were performed in accordance with the guidelines provided by the Care and Use of Animals Committee of the Institute of Medical Sciences, Shanghai Jiao Tong University School of Medicine, China. Spinal dorsal horn (SDH) and hippocampal neurons were prepared from 15- to 18-day-old embryonic Sprague-Dawley rats as described previously (Zhang et al., 2008). In brief, rat SDH and hippocampi were dissociated in calcium-free saline with sucrose (20 mM) and seeded (1–5 × 105 cell/ml) on poly-d-lysine (Sigma-Aldrich, St. Louis, MO)-coated glass coverslips. The neurons were grown in Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA) with l-glutamine plus 10% fetal bovine serum (Invitrogen) and 10% F12 Nutrient mixtures (Invitrogen) at 37°C in a 5% CO2 humidified atmosphere. After 1 day, the medium was changed to Neurobasal medium (1.5 ml; Invitrogen) supplemented with 2% B27 and replaced every 3 to 4 days. Starting on the third day after plating, the cultures were treated with 5-fluoro-5′-deoxyuridine (20 μg/ml; Sigma-Aldrich), which blocks cell division of non-neuronal cells and helps stabilize the cell population. Cells were used for electrophysiological recordings 5 to 21 days after plating.
Mutations of receptor cDNA were generated with the QuikChange mutagenesis kit (Agilent Technologies, Santa Clara, CA) in accordance with the manufacturer's protocol using high-pressure liquid chromatography-purified or polyacrylamide gel electrophoresis-purified oligonucleotide primers (Sigma-Genosys, The Woodlands, TX). All mutants were verified by DNA sequence analysis.
Functional Expression of the Recombinant GlyRs.
The human α1 and β subunits were kindly provided by Dr. Yu-tian Wang (University of British Columbia, Vancouver, BC, Canada). The human α2 subunit was provided by Dr. Heinrich Betz (Department of Neurochemistry, Max Planck Institute for Brain Research, Frankfurt, Germany). The rat α3 subunit was obtained from Dr. Jochen Meier (Department of Developmental Physiology, Johannes-Mueller Center of Physiology, Charite University Medicine, Berlin, Germany). All constructs were expressed in Chinese hamster ovary (CHO) cells as described previously (Zhang et al., 2008). In brief, CHO cells were cultured at 37°C in a humidified atmosphere of 5% CO2 and 95% air. The cells were maintained in F12 medium supplemented with 1 mM l-glutamine, 10% fetal bovine serum, and 100 units/ml penicillin/streptomycin (Invitrogen). Transient transfection of CHO cells was carried out by using Lipofectamine 2000 (Invitrogen). The ratio of cDNAs encoding GlyR α and β subunits was 1:3 to ensure the formation of functional heteroligomers. After exposure to transfection solution for 4 h, cells were washed twice with the culture medium. Electrophysiological measurements were performed 24 to 48 h after transfection. Green fluorescent protein was used for the identification of the transfected cells.
The electrophysiological recordings were performed in conventional whole-cell configuration under voltage-clamp conditions. The cells were perfused by the standard external solution that contained 150 mM NaCl, 5 mM KCl, 1 mM MgCl2, 2 mM CaCl2, 10 mM glucose, and 10 mM HEPES, with the pH adjusted to 7.4 with Tris-base. The osmolarity of all bath solutions was adjusted to 325 to 330 mOsM with sucrose (3300; Advanced Instruments, Norwood, MA). The pipette solution for whole-cell patch recording was 120 mM KCl, 30 mM NaCl, 1 mM MgCl2, 0.5 mM CaCl2, 5 mM EGTA, 2 mM Mg-ATP, and 10 mM HEPES. The internal solution was adjusted to pH 7.2 with Tris-base. When the current-voltage relationships for glycine-activated current (IGly) were examined, 300 nM tetrodotoxin and 100 μM CdCl2 were added to exclude any potential interference factors such as the increase in intracellular Ca2+ (Xu et al., 1999; Fucile et al., 2000) induced by the activation of voltage-gated calcium channels in the standard external solution, and K+ was replaced with Cs+ in the pipette solution. Patch pipettes were pulled from glass capillaries with an o.d. of 1.5 mm on a two-stage puller (PP-830; Narishige, Tokyo, Japan). The resistance between the recording electrode filled with the pipette solution and the reference electrode was 3 to 5 MΩ. Membrane currents were measured by using a Multiclamp 700A (Molecular Devices, Sunnyvale, CA) and sampled and analyzed by using a Digidata 1320A interface and a computer running Clampex and Clamp-fit software (version 8.0.1; Molecular Devices). Unless otherwise noted, the membrane potential was held at −60 mV throughout the experiment under voltage-clamp conditions. The average access resistances for neurons and CHO cells used in the experiments all were approximately 10 MΩ. The average whole-cell capacitance for the spinal cord neurons, hippocampal neurons, and CHO cells was approximately 20, 25, and 30 pF, respectively. All experiments were carried out at room temperature (23 ± 2°C).
Chemicals and Drugs.
The chemicals used in the present study, including curcumol, curdione, and curcumin were purchased from Sigma-Aldrich. Curcumol, curdione, curcumin, and bicuculline methiodide (BMI) were initially dissolved as concentrated stock solutions in dimethyl sulfoxide and subsequently diluted to the desired concentration in the standard external solution. The final concentration of dimethyl sulfoxide was lower than 0.1% and was verified ineffective alone at the same concentration in control experiments (data not shown). Other drugs were either first dissolved in deionized water and then diluted to the final concentrations in the standard external solution just before use or dissolved directly in the standard external solution. Drugs were applied by using a rapid application technique termed the “Y-tube” method as described previously (Murase et al., 1990; Dong and Xu, 2002; Li et al., 2003b). The tip of the drug tube was positioned between 50 and 100 μM away from the patched cells. This system allows a complete exchange of external solution surrounding a cell within 20 ms. Throughout the experiment, the bath was superfused continuously with the standard external solution.
Values were expressed as the mean ± S.E.M. One-way analysis of variance was used for statistical significance between multiple groups, and Student's t test was carried out for the comparison of two groups. P values < 0.05 were considered to be statistical significant. P and n represent the value of significance and the number of neurons or cells, respectively. The software Clampfit 9.0.1 (Molecular Devices) was used for data analysis. The continuous theoretical curves for concentration-response relationships of glycine in the presence or absence of curcumol were drawn according to a modified Michaelis-Menten equation by the method of least squares (the Newton-Raphson method) after normalizing the amplitude of the response: I = ImaxCnH/[CnH + (EC50)nH], where I is the normalized value of the current, Imax is the maximum whole cell current amplitude, C is the drug concentration, EC50 is the agonist concentration that induced the half-maximal response, and nH is the apparent Hill coefficient. The curve for the effect of curcumol on IGly was fitted by the equation: I = Imax(IC50)nH/[CnH + (IC50)nH], where IC50 represents the antagonist concentration producing a half-maximal inhibitory effect, and the others are the same as described above.
Curcumol Inhibits Glycine-Activated Current in Cultured Rat Hippocampal Neurons.
Glycine is one of the natural ligands for GlyRs in mammalian CNS. To investigate potential targets of R. curcumae, we examined the effects of main R. curcumae extracts on IGly in cultured rat hippocampal neurons. At a holding potential of −60 mV under the whole-cell voltage-clamp mode, bath application of glycine (50 μM) evoked an inward current in all tested neurons. This current was strychnine-sensitive and chloride-dependent (data not shown), indicating that it was mediated by the GlyR-chloride channels. Curcumol, curdione, and curcumin (for structures see Fig. 1,A–C) are representative compounds conferring the pharmacological effects of R. curcumae. We thus examined their effects on IGly. As shown in Fig. 1, prior administration of curcumol, curdione, or curcumin before glycine application resulted in significant reductions of IGly. The percentage of inhibition by 100 μM curcumol, curdione, and curcumin was 69.0 ± 2.1% (P < 0.001, n = 12), 31.1 ± 1.6% (P < 0.001, n = 10), and 22.0 ± 2.6% (P < 0.05, n = 10), respectively (Fig. 1D). Among the three compounds, curcumol had the greatest inhibitory effect on IGly. The onset and reversal of the inhibition of IGly by the three compounds were rapid, indicating the potential acute modulation of GlyRs by R. curcumae. These results prompted us to investigate the possible molecular mechanism of the inhibitory effect of curcumol on GlyRs.
Characterization of Curcumol-Mediated Inhibition of IGly.
As shown in Fig. 2, A and B, curcumol inhibited IGly in a concentration-dependent manner when coapplied with glycine after a short pretreatment (∼10 s). Because of the limited solubility, 300 μM curcumol was the maximal concentration used in this study. The average IC50 value for curcumol inhibition of IGly was 68.0 ± 5.4 μM, with a Hill coefficient of 2.0 ± 0.2. To make an obvious and significant effect of curcumol on GlyRs, we chose 100 μM as the effective concentration in most of the following study. Based on previous studies suggesting that the lipophilic curcumol is able to penetrate blood-brain barriers with the maximal concentration of curcumol after intravenous injection of R. curcuma oil up to 108.85 ± 65.91 and 92.38 ± 17.63 μg · g−1 in the liver and brain, respectively (Su et al., 1980; Zhang et al., 2009), we speculate that curcumol at the present concentrations can easily attain the effective concentrations to treat neurological disorders probably by targeting against GlyRs. Moreover, previous studies (Wang and Zhao, 2004; Sun et al., 2008) have suggested that a comparable level of R. curcuma oil (curcumol as the main ingredient accounting for 85%) always induced significant behavioral changes in the CNS.
To elucidate whether curcumol interferes with glycine binding to the GlyRs, we recorded IGly at different concentrations of glycine in the absence or presence of curcumol (100 μM). We found that curcumol effectively inhibited IGly evoked at various glycine concentrations used, including both subsaturating and saturating concentrations (Fig. 2C). In addition, the curcumol suppression of IGly was independent of glycine concentrations (close to 70% inhibition at all concentrations; Fig. 2, D and E). Therefore, curcumol probably inhibits IGly in a noncompetitive manner. However, as suggested by Fig. 2D, the EC50 and Hill coefficient of IGly changed from 50.1 ± 4.6 μM (1.8 ± 0.2) to 115.7 ± 8.6 μM (1.5 ± 0.1) in the absence or presence of 100 μM curcumol, respectively, implying that other action modes of curcumol cannot be absolutely excluded, although curcumol most likely exerts its antagonistic effect on GlyRs in a noncompetitive manner.
Curcumol-Mediated Inhibition of IGly Is Independent of GABAAR Activation.
In the hippocampus, GlyRs are thought to function extrasynaptically to produce tonic inhibition, whereas GABAA receptors (GABAARs) are considered to be responsible for both phasic and tonic inhibitions (Farrant and Nusser, 2005; Xu and Gong, 2010). It is noteworthy that asymmetric cross-inhibition between GlyRs and GABAARs has been identified in both rat hippocampal and SDH neurons (Li and Xu, 2002; Li et al., 2003a). Therefore, it might be possible that the inhibitory effect of curcumol on IGly was interfered with by GABAARs. To examine this possibility, we tested the effect of curcumol on IGly in the absence or presence of BMI (10 μM), a general antagonist of GABAARs. As shown in Fig. 3, A and B, curcumol induced a small current that was inhibited by BMI, indicating that the current induced by curcumol is through activation of GABAAR in cultured hippocampal neurons. However, the inhibitory effect of curcumol on IGly persisted, which was 69.1 ± 1.9 and 70.1 ± 1.2% in the absence and presence of BMI, respectively (P > 0.05; n = 5). Although curcumol may weakly activate GABAARs, whose effect may partially mask its inhibitory effect on GlyRs in cultured neurons, our data demonstrate that the inhibition of IGly by curcumol did not depend on the activation of GABAARs. To maintain a more natural interaction of curcumol with GlyRs, we performed electrophysiological recordings in the absence of a GABAAR antagonist in all subsequent experiments. Collectively, our data indicate that the curcumol-mediated inhibition of IGly most likely occurs through direct interaction between the compound and the GlyRs.
Curcumol-Mediated Inhibition of IGly Is State-Independent.
To determine whether curcumol-mediated inhibition of IGly is state-dependent, we compared its inhibitory effects on IGly by using different sequences of drug application. As shown in Fig. 3C, conditions a and e represent the initial and recovered responses to glycine application, respectively; conditions b, c, and d represent coapplication of curcumol (100 μM for all) with glycine without a pretreatment, curcumol pretreatment alone (∼10 s), and curcumol pretreatment plus glycine coapplication, respectively. Although significant inhibition was achieved under all conditions, curcumol pretreatment with glycine coapplication (condition d) gave the strongest inhibition (67.8 ± 3.4%; P < 0.001; n = 5), indicating that a short pretreatment helps curcumol to occupy the site for inhibition and the inhibition is easily reversible upon washout of the compound. Nonetheless, the finding that the inhibition still persisted, although partially, without the pretreatment (condition b; 25.5 ± 2.6%; P < 0.01; n = 9) and shortly after the washout (condition c; 41.0 ± 1.1%; P < 0.001; n = 7) suggests that although curcumol may bind slightly slower than glycine to the GlyRs and the dissociation is not instantaneous, it exerts an antagonistic effect on GlyRs irrespective of the state of the channel opening.
Curcumol-Mediated Inhibition of IGly Is Voltage-Independent.
To further explore the mechanism of curcumol-mediated inhibition of IGly, we investigated its voltage sensitivity. The current-voltage relationship of IGly was examined by using a voltage-step protocol with cells held at potentials ranging from −70 to +30 mV for 1 min before glycine was applied. As shown in Fig. 3E, curcumol did not significantly change the reversal potential of IGly, which in the absence and presence of curcumol was 3.2 ± 2.0 and 2.3 ± 1.1 mV, respectively (P > 0.05; n = 5) and close to the calculated chloride reversal potential (1.3 mV). In addition, curcumol reduced the amplitude of IGly to the same extent (close to 70%) at all potentials tested, including both positive and negative voltages (Fig. 3F). These results indicate that curcumol-mediated inhibition of IGly is voltage-independent, consistent with the view that the action of curcumol is independent of the channel states. Considering the noncompetitive nature of the curcumol action, we preferred that the drug most likely acts as an allosteric inhibitor on GlyRs.
Curcumol-Mediated Inhibition of IGly Depends on the Subunit Composition of GlyRs.
To examine whether there is subunit specificity of curcumol-mediated inhibition, we investigated the effect of curcumol on CHO cells that expressed different α subunits (α1, 2, and 3) of GlyRs without or with coexpression of the β subunit. As shown in Fig. 4, curcumol (100 μM) reversibly reduced IGly only in cells that expressed the homomeric α2 (P < 0.001; n = 8) but not that expressed the homomeric α1 or α3 subunits (P > 0.05; n = 5–8), indicating that the inhibitory action of curcumol is specific on GlyRs that contain the α2 subunit. It is noteworthy that with the introduction of the β subunit both heteromeric GlyRs composed of α2β (P < 0.001; n = 12) and α3β (P < 0.01; n = 10) showed significant inhibition by curcumol, suggesting that the β subunit also contributes to the inhibitory action of the compounds and alters the specificity to include GlyRs that contain the α3, but not α1, subunit, highlighting the importance of the GlyR β subunit in channel modulation. In addition, curcumol exerted inhibitory effects on those different recombinant GlyRs in a manner regardless of the concentrations of glycine (Fig. 4C), further supporting its subunit specificity on GlyRs, thus consequently dominating its effects on GlyRs by the noncompetitive mechanism. Taken together, these results indicate that curcumol selectively inhibits the current mediated by α2-containing GlyRs as well as those containing both α2/α3 and β subunits.
Threonine 59 Is Involved in the Selective Sensitivity of α2-GlyRs on Curcumol-Mediated Inhibition.
A previous study has noted that a threonine at position 59 in the α2 subunit differs from the corresponding residue (alanine) in the α1 and α3 subunits (Fig. 5A), and this residue plays a critical role in the susceptibility of α2-GlyRs to inhibition by cyclothiazide (Zhang et al., 2008). To explore whether this single amino acid difference may be responsible for the curcumol-mediated inhibition, we examined the effect of exchanging residues between the GlyR α1, α3, and α2 subunits at the equivalent positions of Ala52 (α1), Ala52 (α3), and Thr59 (α2) to generate α1-A52T, α3-A52T, and α2-T59A. Expressing these mutants in CHO cells resulted in functional GlyRs, with robust development of Cl− conductance in response to glycine. As shown in Fig. 5, B and D, curcumol (100 μM) significantly inhibited 50 μM IGly in CHO cells that expressed the α1-A52T and α3-A52T mutants. The percentage of inhibition of α1-A52T by 100 μM curcumol was 24.0 ± 1.9%, a significant increase from 8.1 ± 0.2% inhibition exhibited by the wild-type (WT) α1 (P < 0.05; n = 5). The change was more dramatic for α3-A52T; the inhibition increased from 5.2 ± 3.1% for the wild-type α3 to 79.4 ± 3.9% for the mutant (P < 0.001; n = 8). Conversely, the percentage of inhibition was reduced from 62.2 ± 2.0% exhibited by wild-type α2 to 33.5 ± 4.3% by the α2-T59A mutant (P < 0.01; n = 8). Although the consistent results on GlyR subunit specificity are encouraging, it may also be somewhat fortuitous, because possible differences in glycine dependence of curcumol inhibition on those mutant receptors may mediate the subunit-specific pharmacological differences. To further confirm the specificity of curcumol, we used 10 mM glycine on the three mutants and found that curcumol inhibited IGly to a similar extent with that of 50 μM glycine (Fig. 5, C and E). Therefore, these results demonstrate that Thr-59 at the N terminus of α2 subunit plays a major role in conferring the curcumol sensitivity of GlyRs. However, because α2-T59A did not display a complete loss of the curcumol inhibition and α1-A52T was only moderately inhibited by the drug, other regions of the GlyR subunits must also be involved.
Association of Curcumol-Mediated Inhibition of IGly with Changes of GlyR α2 Subunit Expression in Neurons.
Previous studies have established that the α2 subunit is the primary component of GlyRs in hippocampal neurons, whereas the expression of GlyR subunits in rat spinal neurons is developmentally regulated (Becker et al., 1988; Malosio et al., 1991; Lynch, 2004). The differential expression patterns prompted us to evaluate whether the curcumol-mediated inhibition of IGly changes with time in cultured hippocampal and spinal neurons. Rat hippocampal and SDH neurons were cultured for up to 21 days in vitro (DIV), and IGly was recorded in the absence or presence of curcumol (100 μM) on different days. We grouped the results for short-term (5–7 DIV), medium-term (12–14 DIV), and long-term (19–21 DIV) cultures and compared the curcumol-mediated inhibition of IGly. For hippocampal neurons, the degree of inhibition did not change among the three groups (Fig. 6; P > 0.05, comparing DIV 5–7, 12–14, and 19–21; n = 8–10). However, for SDH neurons, the inhibitory effect of curcumol declined significantly in the long-term culture (P < 0.001, comparing DIV 19–21 with DIV 5–7 or 12–14; n = 10–12). Consistent with a developmental switch of the GlyR subunit expression during spinal cord development, this specific decline of the curcumol-mediated inhibition of IGly in cultured spinal neurons further supports the notion that the action of curcumol is subunit-specific in native neurons.
In the present study, we showed that curcumol, a natural compound and active ingredient from the medicinal plant R. curcumae, inhibits IGly mediated by endogenous and recombinant GlyRs in neurons and CHO cells, respectively. In addition, curcumol exerts a subunit-specific inhibition of GlyRs. A single amino acid substitution, T59A, in the α2 subunit of GlyRs significantly reduced the inhibitory effect of curcumol, whereas the reciprocal mutations, A52T, in the α1 and α3 subunits resulted in a gain of curcumol-mediated inhibition. Furthermore, the inhibition by curcumol can be conferred in α3-containing GlyRs with the inclusion of the GlyR β subunit.
Generally, there are three mechanisms for inhibitors of ion channels to exert an inhibitory action based on where they bind to the channel. A competitive inhibitor binds to the site similar or close to the agonist binding domain to compete with the agonist for binding (Webb and Lynch, 2007), an open channel blocker binds to the channel pore to obstruct ion permeation, whereas an allosteric inhibitor binds to a distinct site from the above two sites to affect coupling of conformational changes to channel gating (Li et al., 2003a,b). Based on the downward shift of the concentration-response curve for glycine with a depression of the maximal responses (Fig. 2D), as well as the agonist concentration independence of the curcumol-mediated inhibition (Figs. 2E and 4C), the effect exerted by curcumol on GlyRs is considered to be noncompetitive. In addition, because curcumol is not charged under physiological pH as well as the condition tested here and its inhibition on IGly seems to be state- and voltage-independent (Fig. 3, C–F), it is unlikely to be an open-channel blocker. Therefore, it is most likely that curcumol is an allosteric inhibitor of the GlyRs, meaning that it might directly bind to the channel irrespective of the binding of glycine or channel opening and the binding by curcumol prevents channel activation.
Inhibitory GlyRs are pentamers that can be either homomeric or heteromeric. It has been demonstrated that the subunit composition and distribution of GlyRs exhibit differential patterns and functions in different brain areas and at different developmental stages (Malosio et al., 1991; Sato et al., 1992; Xu and Gong, 2010), implicating a regulated diversification of GlyR subtypes for distinct physiological roles. In situ hybridization studies showed that the α2-containing GlyRs are the predominant forms in the hippocampus, whereas, to a lesser extent, α3 subunit is expressed in pyramidal and granular cell layers (Malosio et al., 1991; Lynch, 2004). In addition, it has been shown that the subunit composition of GlyRs changes with a developmental switch from α2 to α1 at approximately postnatal day 20 in rat spinal cord neurons either in vivo or in vitro (Malosio et al., 1991; Watanabe and Akagi, 1995; Bechade et al., 1996). Consistent with the above observation, we present here the subunit-specific and developmentally regulated inhibitory effect of GlyRs by curcumol (Figs. 4 and 6), indicating the unique selectivity of curcumol on the modulation of inhibitory signaling mediated by different GlyR subtypes. Also important is that the GlyR β subunit, which does not form a functional channel by itself but coassembles with the α subunits to form functional heteromeric channels (Bormann et al., 1993; Kuhse et al., 1993; Grudzinska et al., 2005), confers the curcumol sensitivity of the GlyR α3 subunit. It is known that the β subunit plays an important role in binding to the postsynaptic scaffolding protein gephyrin, as well as in the transduction process between ligand binding and channel opening (Meyer et al., 1995; Grudzinska et al., 2005). Mechanistically, incorporation of the β subunit also affects the sensitivity of GlyRs to both positive and negative modulators, such as ginkgolide B, 1H-indole-3-carboxylic acid (ICS 205-930), certain β-carbolines, and picrotoxinin (Pribilla et al., 1992; Supplisson and Chesnoy-Marchais, 2000; Kondratskaya et al., 2005). Similar to those modulators, curcumol-mediated inhibition of IGly is more pronounced in the presence of the β subunit than in its absence. Not only is the β subunit required for the curcumol sensitivity of the α3-containing GlyRs, but also it showed a tendency to enhance the curcumol-mediated inhibition of the α2-containing GlyRs (Fig. 4), extending the pharmacological usefulness of curcumol and calling for more functional studies of β subunit in the future.
As one type of the inhibitory receptors in the CNS, GlyRs have become an attractive molecular target for drug development because of their important roles in regulating neuronal excitability, synaptic plasticity, and potential involvement in some neurological disorders (Xu and Gong, 2010). In the spinal cord, the roles of GlyRs have been well established. GlyRs in spinal motoneurons are reported to be abnormal in a transgenic mouse model of amyotrophic lateral sclerosis (Lorenzo et al., 2006; Chang and Martin, 2011), raising a possible strategy of enhancing glycinergic inhibition for treating certain neurodegenerative diseases. In addition, GlyRs in spinal dorsal horn are one of the molecular targets for analgesia and are potentiated by cannabinoid for its pain relief effects (Xiong et al., 2011). In that study, Xiong et al. suggested that α1 and α3, but not α2, subunit-forming GlyRs, contribute to pain sensation in the spinal level by using cannabinoids as probes in virtue of their receptor subunit-specific enhancement effects. Unfortunately, because of the less sensitivity of α2 subunit-forming GlyRs to cannabinoids, the exact role of this subunit-specific receptor in pain modulation has not been investigated. Here, identified α2 subunit-specific modulation of GlyRs by curcumol raises a possibility of studying the contribution of α2 GlyR subunit for pain sensation in the future. On the basis of the present allosteric mechanisms of curcumol modulation on GlyRs, obtaining positive modulators of GlyRs by generation of the curcumol analogs will shed more light on the drug development of targeting neurodegenerative disorders and pain by enhancing the glycinergic inhibition.
Moreover, GlyR signaling in brain areas such as in hippocampal neurons is also being actively investigated. Generally, hippocampal GlyRs are thought to be extrasynaptically located and exert a tonic inhibitory role, which can be highly regulated under many pathophysiological conditions (Xu and Gong, 2010). It is reasonable to speculate that the inhibitory effect of curcumol on GlyRs might largely result in the reduction of hippocampal tonic inhibition in vivo, an effect comparable with the down-regulation of this form of inhibition mediated by GABAA α5 subunit-containing receptors through its inverse agonists; therefore it can be a cognition enhancer (Chambers et al., 2004; Ballard et al., 2009). In that case, the down-regulation of GlyR-mediated tonic inhibition might be a novel strategy for cognition enhancement. In support of this assumption, the curcumol-containing R. curcumae oil has been reported to enhance learning and memory behavior (Sun et al., 2008), presumably through the mechanism with down-regulation of the GlyR-mediated tonic inhibition identified here.
In addition, it was reported that the expression of RNA-edited high-affinity GlyR subunit α2192L and α3185L is increased after experimentally induced brain lesion in rats and in patients with temporal lobe epilepsy with severe disease (Meier et al., 2005; Eichler et al., 2008), arguing for potential roles of GlyRs, edited or not edited, in the development of epilepsy and other neurological diseases. In the present study, we found that curcumol was effective in reducing IGly mediated by wild-type GlyRs as well as the high-affinity GlyRs (data not shown). Moreover, the R. curcumae oil (curcumol as the main ingredient accounting for 85%) has been reported to exert antiepilepsy effects in animal models (Wang and Zhao, 2004). Thus, it will be interesting in the future to investigate the exact roles of curcumol, presumably through GlyRs, in the termination of seizure.
R. curcumae oil has attracted a great interest because of its traditional use and newly confirmed therapeutic activities, especially the potential pharmacological effects in the CNS such as neuroprotection and cognition enhancement (Dohare et al., 2008; Sun et al., 2008). Because curcumol is a major bioactive compound in the R. curcumae oil, investigation into the mechanism of curcumol-mediated inhibition on GlyRs will probably promote further characterizations of new pharmacological activities and therapeutic use of this traditional drug as well as the development of novel drugs based on its structure. Moreover, the lipophilic curcumol is able to penetrate blood-brain barriers (Su et al., 1980; Zhang et al., 2009), strengthening the possibility of drug discovery led by natural products for therapies against CNS diseases.
In conclusion, curcumol, in a subunit-specific manner, reversibly and concentration-dependently inhibits IGly in cultured rat hippocampal and SDH neurons, as well as in CHO cells expressing certain isoforms of GlyRs. In addition, a threonine at position 59 of the GlyR α2 subunit seems to be critical for the subtype-specific inhibition of GlyRs by curcumol. Although the exact mechanism by which curcumol inhibits IGly needs further investigation, our results suggest that curcumol may act as an allosteric inhibitor of GlyRs. Our findings implicate GlyRs as a novel molecular target of curcumol, which may underlie its pharmaceutical effects in the central nervous system.
Participated in research design: Wang, W.-G. Li, Zhu, Xu, Wu, and Y. Li.
Conducted experiments: Wang, W.-G. Li, and Huang.
Performed data analysis: Wang, W.-G. Li, and Y. Li.
Wrote or contributed to the writing of the manuscript: Wang, W.-G. Li, Zhu, Xu, Wu, and Y. Li.
We thank Dr. J. J. Celentano for critical reading of the manuscript and Drs. D. S. Liu, X. Xiao, and Q. Fang for helpful technical assistance.
This study was supported by the National Natural Science Foundation of China [Grants 91132303, 30970937]; the Shanghai Municipal Education Commission (Leading Academic Discipline Project J50201); the Shanghai Science and Technology Committee [Grant 09JC1408700]; and the China Postdoctoral Science Foundation [Grant 2012M511105].
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- glycine receptor
- glycine-activated current
- central nervous system
- spinal dorsal horn
- Chinese hamster ovary
- bicuculline methiodide
- wild type
- days in vitro
- γ-aminobutyric acid type A
- A-type GABA receptor
- ICS 205-930
- 1H-indole-3-carboxylic acid
- no significant difference.
- Received April 17, 2012.
- Accepted August 13, 2012.
- Copyright © 2012 by The American Society for Pharmacology and Experimental Therapeutics