Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

N-Methyl-D-aspartate receptors (NMDARs) play an important role in neuronal development (Scheetz and Constantine-Paton, 1994) and synaptic plasticity (Collingridge and Singer, 1990). NMDAR activation by the coagonists glutamate and glycine leads to an open state of the Ca²⁺-permeable ion channel and subsequent activation of second messenger systems. NMDARs have multiple modulatory domains that regulate its function (McBain and Mayer, 1994). Within the channel pore there are recognition sites for noncompetitive channel blockers such as phencyclidine and (+)-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten 5,10-imine maleate (MK-801). Because these noncompetitive channel blockers require an open state of the channel to bind, tritiated channel blockers such as MK-801 have been used to monitor the functional state of the channel (Reynolds and Miller, 1988; Johnson et al., 1989).

Divalent cations play an important role in the modulation of NMDAR function. Mg²⁺ at physiological concentrations exerts a voltage-dependent block of the NMDAR at a site thought to be located deep within the ion pore (Mayer et al., 1984; Coan and Collingridge, 1985). An intracellular Mg²⁺ block of the NMDAR channel exhibiting voltage dependence opposite to the extracellular Mg²⁺ block has also been described (Kupper et al., 1998). Wang and MacDonald (1995) demonstrated that when the NMDAR glycine site is not saturated, low extracellular concentrations of Mg²⁺ potentiate NMDAR currents. Potentiating concentrations of Mg²⁺ are lower than concentrations producing voltage-dependent block of the channel. The Mg²⁺ potentiation of NMDA-evoked currents was the result of an increase in the affinity of the NMDAR for glycine. Paoletti et al. (1995) also reported a glycine-independent potentiation of the NMDAR by low extracellular concentrations of Mg²⁺.

Similar to Mg²⁺, low millimolar Ca²⁺ concentrations potentiate NMDA responses in trigeminal neurons (Gu and Huang, 1994). The modulatory effect of Ca²⁺ on NMDA-evoked currents also occurs at subsaturating glycine concentrations and is the result of a Ca²⁺-mediated increase in the NMDAR affinity for glycine. There is suggestive evidence that Ca²⁺ and Mg²⁺ bind at the same site to regulate the affinity of the NMDAR glycine site (Wang and MacDonald, 1995; Paoletti et al., 1995). Consistent with these electrophysiological findings (Gu and Huang, 1994; Paoletti et al., 1995; Premkumar and Auerbach, 1996), radioligand binding studies show that Ca²⁺ and Mg²⁺ potentiate [³H]MK-801 (Rajdev and Reynolds, 1992; Enomoto et al., 1992) and [³H]glycine (Marvizon and Skolnick, 1988) binding to the NMDAR channel and the glycine site, respectively. These earlier radioligand binding studies described the potentiating effects of Ca²⁺ and Mg²⁺ and other cations such as Ba²⁺, Mn²⁺, and Sr²⁺ on the NMDAR but did not define the mechanism of potentiation.

Divalent cations of toxicological importance, in particular the heavy metal Pb²⁺ and Zn²⁺, are potent inhibitors of the NMDAR (Guilarte, 1997). In vitro or in vivo exposure to Pb²⁺ impairs NMDAR-dependent long-term potentiation in the hippocampus, a cellular model of learning and memory (Altmann et al., 1991; Lasley et al., 1993; Zaiser and Miletic, 1997). Pb²⁺-induced impairment of the NMDAR is emerging as a principal mechanism for learning and memory deficits observed in children exposed to Pb²⁺ in their environment (Guilarte, 1998). The precise mechanism(s) by which Pb²⁺ produces NMDAR inhibition is not known. However, we have suggested that Pb²⁺ may inhibit the NMDAR by interacting at a Zn²⁺ regulatory site (Guilarte et al., 1995). Pb²⁺ may also have properties that mimic Ca²⁺ in some biological systems (Goldstein, 1993). Therefore, this study was undertaken to biochemically characterize NMDAR potentiation by Ca²⁺ and Mg²⁺ to determine whether divalent cations such as Pb²⁺ and Zn²⁺ can influence such effects. Our findings show that in the presence of subsaturating concentrations of NMDAR agonists, Ca²⁺ and Mg²⁺ potentiate NMDAR function by increasing the receptor affinity for glycine, and this effect is antagonized by Pb²⁺ and Zn²⁺.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

[³H]MK-801 with a specific activity of 23.9 Ci/mmol was purchased from DuPont-NEN (Boston, MA). Unlabeled (+) MK-801 hydrogen maleate was obtained from RBI (Natick, MA). All other supplies were purchased from commercial sources and were of the highest grade possible.

Rat Brain Membrane Preparation. Adult rat brain membranes were prepared from whole brain without the cerebellum and brain stem due to the low density of NMDARs in these brain structures. The preparation of rat brain membranes and the [³H]MK-801 binding assay have been described (Guilarte and Miceli, 1992). Briefly, rat brain tissue was homogenized in 10 volumes of 0.32 M sucrose at 4°C and centrifuged at 1000g for 10 min. The supernatant was centrifuged at 18,000g for 20 min, the pellet resuspended in 10 volumes of 5 mM Tris-HCl (pH 7.7) with a polytron (6 setting) and centrifuged at 8000g for 20 min. The supernatant and upper buffy coat were centrifuged at 40,910g for 20 min. The resulting pellet was homogenized in 10 volumes of 5 mM Tris-HCl buffer with a polytron and centrifuged at 40,910g for 20 min. This washing procedure was done four times and the final pellet stored at 80°C overnight. The following day the pellet was thawed and homogenized in 10 volumes of Tris-HCl buffer with a polytron and centrifuged at 40,910g. The washing procedure was repeated four times and the pellet stored at 80°C until used.

[³H]MK-801 Binding Assay. [³H]MK-801 binding assays were performed with adult rat brain membranes unless otherwise indicated. On the day of assay, the rat brain membrane pellet was thawed and homogenized with a polytron in 5 mM Tris-HCl assay buffer, pH 7.5 (sodium-free) and distributed into assay tubes to provide approximately 200 to 300 µg protein/tube. Protein concentration of tissue homogenates was determined by the method of Lowry et al. (1951) using BSA as a standard. The final assay volume was 1 ml and all tubes were kept on ice during preparation. Membrane suspensions were incubated in triplicate and binding was determined with [³H]MK-801 concentrations ranging from 0.5 to 20 nM for saturation isotherms or at a 2- to 3-nM concentration for single point assay. Nonspecific binding was measured in the presence of 100 µM unlabeled (+)-MK-801 hydrogen maleate. Assay tubes were incubated for 1 h at 24°C (nonequilibrium conditions). The reaction was terminated by filtration through Whatman GF/B filter paper with a BRANDEL filtering system. Filters were washed three times with 4 ml of ice-cold assay buffer. Radioactivity retained in the filters was measured by liquid scintillation spectrometry using 10 ml of complete counting cocktail (Budget-Solve; RPI Corp., Mt. Prospect, IL).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Biphasic Effect of Ca²⁺ and Mg²⁺ on [³H]MK-801 Binding to NMDARs in Extensively Washed Adult Rat Brain Membranes. The addition of Ca²⁺ and Mg²⁺ (1-10,000 µM) to extensively washed adult rat brain membranes containing nominal subsaturating concentrations of glutamate and glycine (low nM range) resulted in a biphasic effect on [³H]MK-801 binding to NMDARs (Fig. 1). A potentiation of [³H]MK-801 binding was measured from approximately 1 to 600 µM concentrations. The mean ± S.E.M. Ca²⁺ EC₅₀ value (concentration that enhances binding to 50% of maximal) was 41.9 ± 6.3 µM (n = 14) with a maximal enhancement of 177.0 ± 7.8% of basal binding (no Ca²⁺ added). Concentrations of Ca²⁺ higher than 800 µM inhibited [³H]MK-801 binding with a mean ± S.E.M. IC₅₀ (concentration that inhibits binding to 50% of maximal) value of 6.44 ± 1.33 mM. The mean ± S.E.M. Mg²⁺ EC₅₀ value was 50.0 ± 7.0 µM (n = 6) with a maximal enhancement of 144.3 ± 6.5% of basal binding (no Mg²⁺ added). Similar to Ca²⁺, concentrations of Mg²⁺ higher than 800 µM inhibited [³H]MK-801 binding with a mean ± S.E.M. IC₅₀ value of 4.36 ± 1.78 mM. On a molar basis both Ca²⁺ and Mg²⁺ were equally potent in stimulating [³H]MK-801 binding, but Ca²⁺ appeared to be more efficacious.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Effect of increasing concentrations of Ca2+ or Mg2+ on the binding of [3H]MK-801 to the NMDA receptor channel from extensively washed adult rat brain membrane preparations. Binding was performed in the absence of added glutamate and glycine as described in Materials and Methods. Each value is the mean ± S.E.M. of 14 (calcium) or 6 (magnesium) different determinations.

NMDAR Agonists Alter the Ca²⁺ and Mg²⁺ Potentiation and Inhibition of [³H]MK-801 Binding to the NMDAR Channel. To understand the effect of NMDAR agonists on the Ca²⁺ and Mg²⁺ potentiation of [³H]MK-801 to the NMDAR channel, divalent cation/[³H]MK-801 dose-effect curves were performed in the absence or presence of saturating concentrations of glutamate with or without increasing concentrations of glycine. Saturation of the glutamate site by 50 µM glutamate (glycine site not saturated), increased potency (decreased EC₅₀ value) and decreased efficacy (maximal enhancement) of Ca²⁺ and Mg²⁺ on [³H]MK-801 binding (Table 2). The decrease in efficacy was related to increased basal levels of binding in the presence of glutamate so that the additional effect of Ca²⁺ on binding was markedly reduced (data not shown). Saturation of the glutamate site (50 µM) with the addition of a near saturating concentration of glycine (2 µM), increased the potency and reduced efficacy of Ca²⁺ and Mg²⁺ potentiation of the NMDAR beyond that obtained with glutamate alone. Again, this was the result of an agonist-induced increase in basal binding. When both the glutamate (50 µM) and glycine (20 µM) sites were maximally saturated, Ca²⁺ and Mg²⁺ were not effective in potentiating [³H]MK-801 binding (Table 2).

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Effect of glutamate, glycine, and the glycine site-competitive antagonist DCKA on the Ca2+ potentiation of [3H]MK-801 to the NMDA receptor channel. Each value is the mean ± S.E.M. of four different determinations. Similar results were obtained with Mg2+.

Ca²⁺ and Mg²⁺ Increase the Affinity of the NMDAR for Glycine. The previous experiments suggested that µM concentrations of Ca²⁺ and Mg²⁺ potentiate NMDAR function by modulating the glycine site. To study this possibility, glycine/[³H]MK-801 dose-effect curves were performed in the presence of saturating concentrations of glutamate (50 µM) and increasing cation concentrations. In the presence of saturating concentrations of glutamate, Ca²⁺ and Mg²⁺ increased glycine affinity (decreased glycine EC₅₀ value) in a dose-dependent manner (Table 3 and Fig. 3). No significant cation effect was measured for the glycine-induced maximal enhancement (efficacy) of [³H]MK-801 binding (Table 3).

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Effect of Ca2+ on the affinity of the NMDA receptor for glycine. The addition of 10 µM Ca2+ produced a left shift in the glycine concentration effect curve, indicative of an increase in the potency of glycine to stimulate [3H]MK-801 binding in the presence of Ca2+. Each value is the mean ± S.E.M. of four different determinations. Similar results were obtained with Mg2+.

Developmental Profile of NMDAR Potentiation by Ca²⁺ and Mg²⁺. We determined the Ca²⁺ and Mg²⁺ potentiation of [³H]MK-801 binding to extensively washed neuronal membranes from rats at different postnatal ages (Fig. 4). In these experiments the Ca²⁺ and Mg²⁺ potentiation of [³H]MK-801 binding was also performed in the absence of exogenously added glutamate and glycine. An age-dependent effect was measured for the amount of [³H]MK-801 binding associated with Ca²⁺ potentiating effects (F_6,30 = 15.0; p = .0001). [³H]MK-801 binding to neuronal membranes from 2-day-old rats were not potentiated by Ca²⁺ or Mg²⁺. The amount of [³H]MK-801 binding associated with the Ca²⁺ enhancement was present at low levels during early development and increased as a function of age peaking at 17 days before decreasing to adult levels (Fig. 4). These data indicate that developmental changes occur at the divalent cation-binding site associated with NMDAR potentiation. A similar developmental pattern to that of Ca²⁺ was present for Mg²⁺ (Fig. 4). The effects of age on the potency (EC₅₀) of Ca²⁺ or Mg²⁺ to enhance [³H]MK-801 binding are also presented in Fig. 4. Age did not alter the potency of these divalent cations to enhance [³H]MK-801 binding (F_6,30 = 2.2; p > .05).

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Postnatal ontogeny of Ca2+ and Mg2+ potentiation of [3H]MK-801 binding to extensively washed rat brain membranes. A, [3H]MK-801 binding in fmol/mg protein associated with divalent cation potentiation. B, potency (EC50) of divalent cation stimulation of [3H]MK-801 binding. Each value is the mean of three to six different preparations.

Pb²⁺ and Zn²⁺ Antagonize the Ca²⁺ and Mg²⁺ Potentiation of the NMDAR. Pb²⁺ and Zn²⁺ decreased (increased the EC₅₀ value) the Ca²⁺ and Mg²⁺ potentiation of [³H]MK-801 binding to extensively washed adult rat brain membranes in a dose-dependent fashion (Table 5). Both cations also reduced the maximal magnitude of [³H]MK-801 potentiation (Table 5). However, Zn²⁺ appeared to be more potent than Pb²⁺ in altering the Ca²⁺ and Mg²⁺ EC₅₀ value and reducing the maximal potentiation of [³H]MK-801 binding to adult rat brain membranes.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

NMDAR activity is exquisitely sensitive to regulation by physiologically relevant divalent cations such as Ca²⁺, Mg²⁺, and Zn²⁺ as well as divalent cations of toxicological importance such as Pb²⁺. We demonstrate that the glycine site of the NMDAR is susceptible to modulation by divalent cations and the modulation is most pronounced when the NMDAR glycine site is not saturated. Concentrations of Ca²⁺ and Mg²⁺ in the 1 to 600 µM range potentiate [³H]MK-801 binding to the NMDAR channel in extensively washed adult rat brain membranes (Fig. 1). The potentiating effects of Ca²⁺ and Mg²⁺ on [³H]MK-801 binding are strikingly similar to those described by Gu and Huang (1994) and Wang and MacDonald (1995) on NMDA-evoked currents. We also show that divalent cation potentiation was the result of an increase in the apparent number of [³H]MK-801 binding sites (Table 1), and a direct consequence of an increased affinity of the NMDAR for glycine (Table 3 and Fig. 3). The latter finding is consistent with the observations of Gu and Huang (1994) and Wang and MacDonald (1995).

The site responsible for the potentiating effects of Mg²⁺ and Ca²⁺ is distinctly different from the voltage-dependent Mg²⁺ site located deep within the NMDAR ion channel (Mayer et al., 1984; Coan and Collingridge, 1985). Concentrations of Mg²⁺ that inhibit NMDA-evoked currents also inhibit [³H]MK-801 binding (Fig. 1). Calcium concentrations in the mM range are also known to block channel conductance possibly by binding to a site located near the entrance of the channel (Jahr and Stevens, 1993; Premkumar and Auerbach, 1996; Sharma and Stevens, 1996). These findings suggest that fluctuations in synaptic concentrations of Ca²⁺ and Mg²⁺ that result from changes in synaptic activity may produce different functional effects on native NMDARs.

Ca²⁺ and Mg²⁺ Potentiate NMDAR Function by Modulation of the Glycine Site. The potentiating effects of Ca²⁺ and Mg²⁺ on the NMDAR are masked if saturating concentrations of glutamate and glycine are present (Table 2 and Fig. 2). This suggests that the potentiating divalent cation site is not independent from the glutamate and glycine sites as has been shown with other modulatory sites such as the polyamine site (Williams et al., 1989). The polyamine site is independent from the glutamate and glycine sites because it potentiates [³H]MK-801 binding beyond the enhancement obtained in the presence of saturating concentrations of glutamate and glycine. Because NMDAR potentiation by Ca²⁺ and Mg²⁺ is not measurable at saturating glutamate and glycine concentrations, the Ca²⁺ and Mg²⁺ site must be associated to one of the agonist sites. Evidence linking the divalent cation site with the NMDAR glycine site is shown by cation potentiation of NMDAR with the competitive glycine site antagonist DCKA in the presence of saturating agonists concentrations (Table 2 and Fig. 2). Our data are consistent with those of Gu and Huang (1994) and Wang and MacDonald (1995) in which the addition of potentiating concentrations of Ca²⁺ or Mg²⁺ decreased the inhibitory potency of the NMDAR glycine site antagonist 7-chlorokynurenic acid on NMDA-evoked currents.

Ontogeny of Divalent Cation-Induced Potentiation of the NMDAR. The effect of neuronal development on the NMDAR potentiation by divalent cations is undefined and was investigated in the present study. The potency of Ca²⁺ or Mg²⁺ to stimulate [³H]MK-801 binding in developing animals did not change during development (Fig. 4). Brain tissue from 2-day-old rats did not exhibit Ca²⁺ or Mg²⁺ potentiation of [³H]MK-801 binding. The earliest measurable potentiation of the NMDAR by Ca²⁺ and Mg²⁺ was observed in 5- to 6-day-old rat brain. The amount of [³H]MK-801 binding associated with the Ca²⁺ and Mg²⁺ potentiation increased from low levels shortly after birth, to reach peak levels at 17 days of age before declining to adult levels (Fig. 4). This developmental pattern of [³H]MK-801 binding associated with NMDAR potentiation by Ca²⁺ and Mg²⁺ is similar to the developmental profile of NR1 and NR2A subunit mRNA and protein (Monyer et al., 1994; Riva et al., 1994; Luo et al., 1996; Wenzel et al., 1997). In rat forebrain, the NR2A subunit mRNA or protein is almost undetectable shortly after birth and increases after the first week of life (Monyer et al., 1994; Riva et al., 1994). Thus, the NR1 and/or NR2A subunits may contribute to the potentiating effect of Ca²⁺ and Mg²⁺. It is unlikely that the NR2B subunit would be responsible for the potentiation of the NMDAR by Ca²⁺ and Mg²⁺ as this subunit is abundantly expressed at 2 days of age, a time when divalent cation potentiation of [³H]MK-801 binding is not measurable.

Divalent Cation Regulation at the NMDAR Glycine Site. Our studies suggest that divalent cations interact at a site that modulates the NMDAR glycine-binding site. Low concentrations of Ca²⁺ and Mg²⁺ potentiate [³H]MK-801 binding by increasing NMDAR affinity for glycine (Table 3, Fig. 3). Increasing concentrations of Ca²⁺ diminish the inhibition of NMDAR currents by Zn²⁺ (Mayer et al., 1989) or Pb²⁺ (Marchioro et al., 1996). These studies imply that Ca²⁺, Zn²⁺, and Pb²⁺ interact at the same site on the NMDAR. NMDAR currents and [³H]MK-801 binding are inhibited by Pb²⁺ and Zn²⁺ at high- and low-affinity sites (Christine and Choi, 1990; Guilarte, 1997). Concentrations of Pb²⁺ and Zn²⁺ associated with the low-affinity site (>10 µM) inhibit the NMDAR potentiation by Ca²⁺ and Mg²⁺ (Table 5). Other data show that Pb²⁺ competitively interacts with Ca²⁺ (Marchioro et al., 1996), and that Mg²⁺ potentiates [³H]glycine binding (Marvizon and Skolnick, 1988). Together, these studies suggest that Ca²⁺ and Mg²⁺ have opposing effects on glycine binding from Pb²⁺ and Zn²⁺, but that all the cations may interact at the same site. The cation site could exist independently from the glycine site where cation binding causes stoichiometric changes in the conformation of the glycine binding pocket. Alternatively, the cation site may be located within the glycine site. Cations could then directly interact with glycine to influence the affinity of glycine inside the pocket. The latter is less likely because Zn²⁺ has been shown to be a noncompetitive antagonist of [³H]glycine binding (Yeh et al., 1990), suggesting an allosteric rather than a direct interaction.