Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Chronic exposure of young children to lead produces irreversible damage to the central nervous system (Bellinger et al., 1987; Needleman et al., 1979). This is reflected in the impaired development of cognitive skills and increased antisocial and aggressive behavior (Edmonds et al., 1995). NMDA receptors are in vivo and in vitro targets of Pb²⁺ toxicity that may underlie these developmental and cognitive injuries (Alkondon et al., 1990; Cory-Slechta, 1995a). These receptors play an important role in synaptic development and plasticity and are believed to be components of learning and memory (Bashir et al., 1991; Kalb, 1994; Reymann et al., 1989; Sarvey et al., 1989). Pb²⁺ exposure reduces the response to MK-801 [(+)-5-methyl-10,11-dihydro-5H-dibenzo-[a,b]cyclohepten-5,10-imine maleate], a noncompetitive open channel blocker of NMDA-activated channels, in an operant drug discrimination test that is consistent with a direct action of Pb²⁺ on NMDA receptors in vivo (Cory-Slechta, 1995b). The performance of rats trained on a multiple schedule of repeated learning is less sensitive to MK-801 disruption after chronic Pb²⁺ exposure (Cohn and Cory-Slechta, 1993). Finally, the accuracy-impairing effects of NMDA in a repeated learning paradigm are increased by chronic Pb²⁺ exposure (Cohn and Cory-Slechta, 1994). All of these results are consistent with Pb²⁺ changing normal learning and behavioral patterns by a direct action on NMDA receptors in vivo.

Molecular cloning of NMDA receptors has identified two subunit families. The NR1 subunit (rat designation; 1 in mouse) exists in eight splice variants (Kutsuwada et al., 1992; Monyer et al., 1992; Moriyoshi et al., 1991). One splice variant, NR1b, has an in-frame insertion of 63 bp located immediately downstream of nucleotide 516. Three deletions are also observed between nucleotides 2536 and 3002, two of which can remove the stop codon and result in the use of a previously out-of-frame stop codon. Combinations of these splice sites produce the eight splice variants (Durand et al., 1993; Hollmann et al., 1993; Nakanishi et al., 1992; Sugihara et al., 1992). The NR2 subunit (rat designation; 1-4 in mouse) consists of a family of four members that do not form functional homomeric receptor channel complexes (Kutsuwada et al., 1992; Meguro et al., 1992; Monyer et al., 1992). Highly active NMDA receptors are present only when NR1 and NR2 subunits are coexpressed (Kutsuwada et al., 1992; Monyer et al., 1992; Tsuzuki et al., 1994). The characteristics of these heteromeric receptors, such as the strength of the Mg²⁺ block (Burnashev et al., 1992; Monyer et al., 1992), sensitivity to glycine (Ishii et al., 1993; Kutsuwada et al., 1992), the time course of deactivation (Monyer et al., 1992, 1994), Ca²⁺ inactivation (Krupp et al., 1996) and the sensitivity to reducing agents (Köhr et al., 1994), are dependent on the type of NR2 subunit present.

The expression of individual NMDA receptor subunits is developmentally regulated in a regionally specific manner (García et al., 1994; Sheng et al., 1994; Zhong et al., 1995). In vivo and in vitro, the efficacy of the block by Pb²⁺ by NMDA receptors changes with neuronal age (Alkondon et al., 1990; Guilarte et al., 1994, 1995; Guilarte and Miceli, 1992; Jett and Guilarte, 1995; Ujihara and Albuquerque, 1992). These data suggest that the sensitivity of the NMDA receptor to Pb²⁺ may depend on the specific subunits composing the receptor. We recently demonstrated that the NMDA receptors formed when 1 (NR1) subunits are expressed in combination with 1 (NR2A) or 2 (NR2B) subunits are significantly more sensitive to inhibition by Pb²⁺ than 112 (NR1-2AB) receptors (Omelchenko et al., 1996). NR1 combined simultaneously with multiple NR2 subunit types are expressed in the adult hippocampus, cortex and cerebellum (Chazot et al., 1994; Sheng et al., 1994; Wafford et al., 1993). These results, however, do not address whether such subunit-specific interactions between Pb²⁺ and NMDA receptors are unique for NR2A- or NR2B-containing NMDA receptors or are a common feature of NMDA receptors (Omelchenko et al., 1996). This issue is particularly important because the pharmacology of NMDA receptors shows regional variation and may contribute to selective Pb²⁺ actions within the brain (Porter and Greenamyre, 1995; Sakurai et al., 1993). Here, we report the effects of Pb²⁺ on other NMDA receptor compositions (NR1-2A, NR1-2C and NR1-2D) and compare the results with NR1-2AC receptors, a composition that is present in the mature cerebellum (Chazot et al., 1994).

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Oocyte isolation and cRNA injection. cDNAs for the rat NMDA receptor subunits NR1b, NR2A, NR2C and NR2D were kindly provided by the Molecular Neurobiology Laboratory of The Salk Institute. The NR1b subunit contains a 63-bp in-frame alternate splice insertion immediately downstream from nucleotide 516 (Hollmann et al., 1993). Plasmid DNA (1-5 µg) with coding inserts for the NMDA receptor subunits was linearized and cRNA was transcribed in vitro using the appropriate enzymes (GIBCO BRL, Gaithersburg, MD) under standard reaction conditions with the addition of 2.5 mM m⁷G(5)ppp(5)G (Pharmacia, Piscataway, NJ). The cRNA was evaluated by formaldehyde gel electrophoresis, quantified by optical density (260 nm) and stored at 80°C in 10 mM Tris-buffered solution, pH 8.0.

Voltage-clamp recording procedure. The steady-state currents generated by NMDA receptor activation were measured using standard two-electrode voltage-clamp techniques. Microelectrodes (1-3 m) were filled with a solution consisting of 3 M KCl or 3 M KCl, 10 mM HEPES and 20 mM 1,2-bis(2-aminophenoxy)ethane-N, N,N,N-tetraacetic acid, pH 7.5. The oocytes were placed in a 200-µl bath and perfused at 2 ml/min with a oocyte solution consisting of (in mM) NaCl 96, KCl 2, BaCl₂ 1.8 and HEPES 10, pH 7.5. Ba²⁺ was substituted for Ca²⁺ to prevent activation of the Ca²⁺-activated Cl channels present in the oocytes (Leonard and Kelso, 1990; Miledi and Parker, 1984). Glycine (10 µM) was present in all control and agonist-containing solutions. The resting membrane potential of the oocytes was determined and the oocytes were voltage-clamped using a Axoclamp 2A (Axon Instruments, Burlingame, CA) near the measured resting membrane potential. Before Glu application, the voltage was stepped to 60 mV until the holding current reached a new steady state. The test solutions (containing agonist with and without Pb²⁺) were applied from a theta tube and controlled by electronic valves (Omelchenko et al., 1996). The data were collected on-line using a Macintosh IIcx equipped with an ITC16 A/D interface (Instrutech, Elmont, NY) and AxoData software (Axon Instruments).

Data analysis. Concentration-response curves for Pb²⁺ inhibition of Glu-activated currents were performed using saturating concentrations of agonist. The IC₅₀ value was estimated by fitting the curves to the equation I_Pb/I_c = 1/[1 + ([Pb²⁺]/IC₅₀)^m]; where I_Pb/I_c is the ratio of the current amplitude in the presence of Pb²⁺ to that in the absence of Pb²⁺, [Pb²⁺] is the Pb²⁺ concentration (0.01-100 µM), IC₅₀ is the concentration that produced 50% inhibition and m is the Hc for the blocker. Concentration-response curves for Glu were estimated by fitting the data to the logistic equation: I_c = 1/[1 + (EC₅₀/[A])ⁿ]; where I_c is the current amplitude normalized to the amplitude of the threshold current, [A] is the Glu concentration (0.01 µM-1 mM), EC₅₀ is the concentration that produces a half-maximal response and n is the Hc for the agonist. The curve fitting was performed using Kaleidagraph 3.0 (Synergy Software, Reading, PA). All data were expressed as the mean ± standard error. Analysis of variance and Fisher's protected least significance difference post hoc analysis were performed using StatView (Abacus Concepts, Berkeley, CA).

Chemicals. All chemicals were purchased from Sigma Chemical (St. Louis, MO) except lead(III) acetate trihydrate, which was purchased from Aldrich Chemical (Milwaukee, WI). The Pb²⁺ solutions were made fresh just before application to the oocytes.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Agonist potency of Glu. The application of Glu with a saturating concentration of glycine (10 µM) to oocytes injected with cRNAs for different combinations of NMDA receptor subunits (NR1b-2A, NR1b-2C, NR1b-2D and NR1b-2AC) evoked inward currents, indicating the functional expression of heteromeric NMDA receptors. Tris-injected oocytes had no responses to Glu application. These receptors had a high affinity for Glu with EC₅₀ values ranging from 1.6 to 8.9 µM (fig. 1, table 1), similar to the affinity of native NMDA receptors (Kutsuwada et al., 1992; Williams, 1994). Of these receptors, those consisting of NR1b-2A subunits had the lowest affinity for Glu (table 1). These data support the idea that the type of NR2 subunit present determines the agonist sensitivity of NMDA receptors and supports a regulatory role of NR2 subunits for receptor properties (Ishii et al., 1993; Köhr et al., 1994; Kutsuwada et al., 1992; Tsuzuki et al., 1994; Monyer et al., 1992, 1994). Interestingly, the EC₅₀ value for NR1b-2C-containing receptors was very similar to that for NR1b-2AC receptors (table 1). In contrast, the EC₅₀ value for NR1b-2A receptors was significantly larger than that for NR1b-2C or NR1b-2AC receptors (table 1). This comparison suggests a dominant role for the NR2C subunit in determining NR1b-2AC receptor properties. This idea, however, was not pursued with other combinations of NR2 subunits expressed with NR2C because it was beyond the goals of this study.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Inhibition of recombinant NMDA receptors by Pb2+. The saturating Glu concentration was chosen as the Glu concentration for each experiment. Glu-activated currents were obtained in the absence and presence of 10 µM Pb2+.

Pb²⁺ inhibition of Glu-activated currents. Pb²⁺ (0.1-30 µM) inhibited the Glu-activated currents in all NMDA receptor combinations tested (fig. 1). The IC₅₀ values for Pb²⁺ inhibition ranged from 1.52 to 4.86 µM, with the rank of Pb²⁺ potency NR1b-2A > NR1b-2C > NR1b-2D > NR1b-2AC (table 2).

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Pb2+ inhibition of currents evoked by Glu from recombinant NMDA receptors. A-D, Concentration-response curves for Glu were recorded in the absence () and presence of Pb2+ (). The amplitudes of Glu-activated currents obtained in the presence of an IC50 concentration of Pb2+ (see table 2) were plotted as ratios of the current amplitude produced by threshold Glu concentration, and each point represents the mean ± S.E.M. of 4 to 9 oocytes. E and F, The effect of agonist concentration on the magnitude of the Pb2+ block of Glu-activated NMDA receptors. These data points were plotted using the formula: [(control current/current in the presence of Pb2+)  1]. E, Pb2+ produced a greater inhibition of NR1b-2A, NR1b-2C or NR1b-2D NMDA receptors at higher agonist concentrations. F, The potency of Pb2+ (8 µM) inhibition was dependent on the Glu concentration at NR1b-2AC NMDA receptors.

Complex effects of Pb²⁺ on NR1b-2AC receptors. In contrast to the NR1b-2A, NR1b-2C and NR1b-2D NMDA receptors, Pb²⁺ had complex actions at NR1b-2AC receptors. At low Glu concentrations (0.01-3 µM), the ratio of the amplitudes of Glu-activated currents in the absence and presence of Pb²⁺ increased with increasing Glu concentrations, consistent with a noncompetitive type of block. However, at Glu concentrations of >3 µM, a decreased current amplitude ratio was observed with increasing agonist concentrations (fig. 2F). In addition, the Glu EC₅₀ value increased in the presence of Pb²⁺ (table 1). Pb²⁺ had a higher potency as an antagonist at 3 µM Glu (IC₅₀ = 3.93 µM) compared with saturating Glu (300 µM) concentrations (IC₅₀ = 8.19 µM) (table 2, fig. 3B). These data suggest that Pb²⁺ may alter the affinity of the NMDA receptor for Glu. At Glu concentrations of <1 µM, Pb²⁺ produced a small (~10%) potentiation of Glu-activated currents from NR1b-2AC-containing receptors (fig. 3B). This effect was also previously observed for 112 (NR1a-2AB) NMDA receptors expressed in oocytes, in which it was more prominent (Omelchenko et al., 1996). Such a potentiating action may contribute to the lower efficiency of the antagonist actions of Pb²⁺ at NR1b-2AC and 112 NMDA receptors.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Concentration-response curves for the inhibition of Glu-activated currents by Pb2+. The current amplitudes recorded in the presence of Pb2+ were plotted as a percentage of those activated in the absence of Pb2+. A, Comparison of the inhibition of NMDA receptors by Pb2+ at saturating Glu concentrations. B, Pb2+ effects at NR1b-2AC NMDA receptors were inhibitory at 3 µM Glu () but become more complex at 300 µM Glu ().

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The present study provides further support for the hypothesis that the vulnerability of NMDA receptors to Pb²⁺ depends specifically on the subunit composition (Omelchenko et al., 1996). Pb²⁺ may produce either potentiation or inhibition depending on the subunit composition, agonist concentration and Pb²⁺ concentration. When the NMDA receptors are saturated by agonist, Pb²⁺ at low concentrations (<1 µM), is a positive modulator of agonist action at NR1b-2AC and 112 (NR1a-2AB) receptors (table 2 and Omelchenko et al., 1996)). The average concentration of Glu in the synaptic cleft has been calculated to be 1 to 5 mM, which is high enough to saturate postsynaptic NMDA receptors (Clements, 1996). We therefore expect that the adult postsynaptic NMDA receptors would be potentiated by low Pb²⁺ concentrations after presynaptic release of Glu, similar to what was observed in experiments with 300 µM Glu (fig. 3). Potentiation of the NMDA-mediated responses should promote an additional Ca²⁺ influx through NMDA receptors and thus affect Ca²⁺-dependent processes, including modulation of NMDA receptors themselves by Ca²⁺-dependent enzymes (Lieberman and Mody, 1994; Tong et al., 1995). At higher concentrations, Pb²⁺ was a potent inhibitor of all recombinant NMDA receptors tested and is least potent at 112 and NR1b-2AC NMDA receptors (table 2 and Omelchenko et al., 1996)). This demonstrates that NMDA receptors, comprising NR1 and a single NR2 subunit, independent of the subunit type, are more vulnerable to Pb²⁺ inhibition than NMDA receptors containing two distinct NR2 subunits.

The biphasic effects of Pb²⁺ on NR1b-2AC and 112 receptors could involve Pb²⁺ interacting with a Zn²⁺ binding site or Pb²⁺ activating protein kinases directly or indirectly and phosphorylating the NMDA receptor (Guilarte et al., 1995; Kapoor et al., 1984; Markovac and Goldstein, 1988; Omelchenko et al., 1996; Zheng et al., 1994). However, data presented here suggest an additional hypothesis. The NMDA receptor complex contains two binding sites for Glu and two binding sites for the coagonist glycine (Benveniste and Mayer, 1991; Clements and Westbrook, 1991). The Glu binding sites have similar affinities, yet the binding of agonist to one site reduces the affinity of the second site for agonist (Clements and Westbrook, 1994). Pb²⁺ binding to the NMDA receptor may change the cooperative interaction between these two sites. Further, the binding sites for Glu and glycine are allosterically linked so binding of Glu reduces the affinity for glycine (Benveniste et al., 1990). Marchioro et al. (1996) demonstrated that Pb²⁺ potentiates NMDA-activated currents in the presence of nonsaturating glycine concentrations (0.01-0.05 µM). They postulated that Pb²⁺ may increase the affinity of the NMDA receptor for glycine (Marchioro et al., 1996). In the presence of a saturating Glu concentration and low Pb²⁺ concentrations, binding of agonist could change the receptor conformation and increase the affinity for either Glu or glycine leading to potentiation. However, if Glu affinity is affected, it happens at a low affinity site because no potentiation by Pb²⁺ was seen with low concentrations of agonist (at 3 µM Glu; fig. 3B).

The multiple actions of Pb²⁺ on NMDA receptors may play a role in the unique vulnerability of some developing brain regions to Pb²⁺ toxicity. The expression of NMDA receptor subunits is developmentally regulated in a regionally specific pattern (García et al., 1994; Sheng et al., 1994; Zhong et al., 1995). In the rat cortex, NR1 and NR2B subunits are present before birth and remain expressed into adulthood. On the other hand, expression of NR2A subunit mRNA begins after birth and increases for the first 3 weeks, at which time it reaches adult levels (Sheng et al., 1994). The adult form of the NMDA receptor in the cortex consists of heteromers of NR1, NR2A and NR2B subunits, whereas the early developmental forms consist of NR1 and NR2B subunits (Sheng et al., 1994). In the cerebellum, NR1 and NR2B subunits are expressed at early stages of development, with a subsequent reduction of NR2B expression and an increase in NR2C and NR2A subunit expression (Akazawa et al., 1994; Farrant et al., 1994; Laurie and Seeburg, 1994; Zukin and Bennett, 1995). The expression of NR1, NR2A and NR2C subunits in vitro is required to recreate the physiological characteristics of cerebellar NMDA receptors (Chazot et al., 1994). These observations suggest that in the fetal cortex and cerebellum, NMDA receptors are composed of NR1-2B subunits that are inhibitable in vitro by Pb²⁺ with an IC₅₀ value of 1 µM (Omelchenko et al., 1996). NMDA receptors in the adult cortex (NR1-2AB subunits) and the adult cerebellum (NR1-2AC subunits) would be potentiated by low concentrations (<1 µM) of Pb²⁺ and inhibited by higher Pb²⁺ concentrations.

The physiological and pharmacological characteristics of NMDA receptors are dependent on the NR2 subunit expressed (Köhr et al., 1994; Kutsuwada et al., 1992; Monyer et al., 1992, 1994). This observation is consistent with previous binding studies that have demonstrated a critical role of the NR2 subunit in formation of binding sites for competitive antagonists on NMDA receptors (Lynch et al., 1994). Because affinity for Glu varies significantly among the NMDA receptors studied here (table 1), it also suggests that each NR2 subunit can form a site for Glu with distinct characteristics. Our data indicate that the 2C subunit plays a dominant role in NR1-2AC receptor. In particular, the affinity of Glu for this receptor is the same as for the NR1-2C receptor (table 1). It is not surprising that another agonist binding site formed by the 2A subunit is undetectable in the dose-response of the NR1b-2AC receptor because NR1-2A receptor has a much lower affinity for Glu than NR1-2C.

Data obtained using animal behavioral techniques demonstrate that NMDA-mediated synaptic transmission is affected by Pb²⁺ exposure (Cory-Slechta, 1995b; Cohn and Cory-Slechta, 1993, 1994). This is important because synaptic plasticity mediated by NMDA receptors is believed to play a role in learning and memory (Bashir et al., 1991; Kalb, 1994; Reymann et al., 1989; Sarvey et al., 1989). An important toxicological question is whether the Pb²⁺ concentrations that produce a direct inhibition of NMDA receptors are similar to those that are observed in Pb²⁺-exposed individuals. The Pb²⁺ concentrations that are believed to produce clinical effects in children are in the range of 10 to 20 µg/dl (0.48-0.97 µM) (Bellinger et al., 1987; McMichael et al., 1988; Pocock et al., 1994). However, the majority of Pb²⁺ in whole blood is bound to erythrocytes, giving a ratio of free to bound Pb²⁺ of ~40:1 and resulting in blood Pb²⁺ concentrations of ~30 nM (Cavalleri et al., 1984). Therefore, the concentrations of Pb²⁺ used in this study appear to be above those observed clinically. However, several points must be considered before such a conclusion can be reached. First, the neurotoxic actions of Pb²⁺ at NMDA receptors may involve not only inhibition but the potentiation of NMDA-activated currents observed with 112 (NR1a-2AB)- and NR1b-2AC-containing receptors. Second, Pb²⁺ is stored in locations similar to Zn²⁺ and may be released from these sites by a like mechanism. Under these conditions, although the concentration of Zn²⁺ may be low in the whole brain, the localized concentration of Zn²⁺ may reach several hundred micromolar (Swanson et al., in press). A similar mechanism may produce similar increases in the concentrations of Pb²⁺ during synaptic release of Pb²⁺ from presynaptic sites (Swanson et al., in press). Finally, extended Pb²⁺ exposure (10-15 min) irreversibly inhibits NMDA-activated currents, suggesting that Pb²⁺ dissociates very slowly from the NMDA receptor (Alkondon et al., 1990; Büsselberg et al., 1994).

The present data together with previous work indicate that independent of brain area, Pb²⁺ has a higher efficiency at NMDA receptors containing only one type of NR2 subunit regardless of their specific composition (Omelchenko et al., 1996). Both the subunit composition and Glu concentration are crucial variables for understanding the actions of Pb²⁺ at NMDA receptors.