Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The N-methyl-D-aspartate (NMDA) subtype of excitatory amino acid receptor is a ligand-gated ion channel that gates Na⁺, Ca²⁺, and K⁺ and is blocked at resting membrane potentials by physiological concentrations of Mg²⁺. The receptor contains a number of distinct recognition sites for endogenous and exogenous ligands. Some of these modulatory substances include polyamines, Mg²⁺, Zn²⁺, glycine, redox agents, and neurosteroids (Nowak et al., 1984; Peters et al., 1987; Johnson and Ascher, 1987; Ransom and Stec, 1988; Aizenman et al., 1989; Wu et al., 1991).

Two families of NMDA receptor subunits have been cloned: the NMDAR1 and NMDAR2 subunits. The NMDA R1 subunit is transcribed as eight alternatively spliced mRNAs (Moriyoshi et al., 1991; Sugihara et al., 1992; Hollman et al., 1993). The NMDAR2 family consists of four different subunits, NR2A, NR2B, NR2C, and NR2D (Nakanishi, 1992; Seeburg, 1993). It has been shown that both the NR1 and NR2 subunits are required to form functional receptors in mammalian cells (Sucher et al., 1993; Boeckman and Aizenman, 1994). Although the precise subunit composition of native receptors is not known, the receptors are likely to be made up of NR1 and various NR2 subunits (Sheng et al., 1994).

Polyamines are complex modulators of the NMDA receptor. The endogenous polyamines, spermine and spermidine, have multiple effects, both positive and negative, on NMDA receptor activation. The effects of polyamines on the NMDA receptor were first observed in binding assays using radioligand channel blockers such as [³H]dizocilpine (MK-801). Spermine and spermidine (3-100 µM) enhanced the binding of open channel blockers such as [³H]MK-801 (Ransom and Stec, 1988; Reynolds and Miller, 1989; Williams et al., 1989; Sacaan and Johnson, 1990). At high concentrations of spermine and spermidine (>100 µM), [³H]MK-801 binding was inhibited, suggesting that polyamines modulate the NMDA receptor by at least two distinct actions.

Electrophysiological studies on the effects of polyamines on native NMDA receptors have produced various results. Spermine has been shown to potentiate, inhibit, or have no effect on native NMDA receptors (Rock and MacDonald, 1992a; Rock and MacDonald, 1992b; Lerma, 1992; Benveniste and Mayer, 1993; Williams et al., 1990; Araneda et al., 1993). Spermine, as studied electrophysiologically, has been shown to have four major effects on recombinant NMDA receptors in Xenopus oocytes: 1) glycine-independent stimulation, in the presence of saturating glycine concentrations; 2) glycine-dependent stimulation, in the presence of subsaturating concentrations of glycine; 3) voltage-dependent inhibition, seen more at hyperpolarized membrane potentials; and 4) a decrease in NMDA agonist affinity (Durand et al., 1993; Williams et al., 1994; Zhang et al., 1994; Williams, 1994; Williams, 1995). Each of these effects is dependent on subunit composition. However, it is not clear how the effects of polyamines in these assays relate to the phenomena observed using [³H]MK-801 binding.

In this study, we have examined the effects of several polyamines on [¹²⁵I]MK-801 binding to native and recombinant NMDA receptors. We used mouse fibroblast cells that were stably transfected with the NR1a splice variant (which indicates an absence of the N-terminal insert site but contains both C-terminal insert sites) and either the NR2A or NR2B subunit. The pharmacological characteristics of these two receptor compositions were studied to further understand each specific subunit and its possible ligand binding sites. Our data show that the polyamine pharmacology of native NMDA receptors cannot be modeled by simple NR1a/NR2A or NR1a/NR2B combinations.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals and Drugs. [¹²⁵I]MK-801 (2200 Ci/mmol) was obtained from DuPont/NEN. Spermine and spermidine, as the hydrochloride salts, were obtained from Sigma (St. Louis, MO). Arcaine and 1,5-(diethylamino)piperidine were obtained from Research Biochemicals Incorporated (Natick, MA). Other drugs and reagents were obtained from commercial sources.

Cell culture and Induction. Stably transfected mouse fibroblasts, containing NR1a/NR2A or NR1a/NR2B subunits, were grown and induced as previously described (Grimwood et al., 1996). Briefly, ML(tk) cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 1 mg/ml geneticin, 100 µM/ml penicillin, and 100 U/ml streptomycin at 37°C in 5% CO₂/95% air. For ligand-binding studies, cells were plated in 100 × 20-mm tissue culture dish at 3 × 10⁶ cells per dish. After 24 h, the culture medium was removed and replaced with medium as described above but without Geneticin and supplemented with 1 µM dexamethasone and 500 µM ketamine for protection. Cells were harvested 48 h after dexamethasone treatment by scraping in 10 mM HEPES-NaOH containing 1 mM EDTA. Cells were homogenized with a polytron and pelleted at 20,000g for 20 min and resuspended by homogenizing in 10 mM HEPES-NaOH for use in the ligand-binding assay. Control experiments using rat brain membranes indicated that the wash conditions were sufficient to remove the ketamine from the membrane preparation.

Receptor-Binding Assays. [¹²⁵I]MK-801 binding assays were performed in well washed rat brain membranes (Reynolds and Palmer, 1991). Binding assays were performed in a final volume of 1 ml of 10 mM HEPES-NaOH (pH 7.4) that contained 0.2 mg of protein, 0.05 nM [¹²⁵I]MK-801, 100 µM glutamate, and 30 µM glycine along with the appropriate polyamine. Nonspecific binding was determined using 30 µM MK-801. Binding assays were incubated for 2 h with BSA (5 mg/ml) at room temperature and terminated by filtration (10 rinses) over Schliecher and Schuell 32 glass-fiber filters, using a 24-well cell harvester (Brandel Inc., Gaithersburg, MD). Filters were soaked in 0.3% polyethyleneimine for 20 min to reduce nonspecific binding. Radioactivity was measured by a gamma counter at an efficiency of 70%.

Data Analysis. Most of the data in this study were analyzed using Prism version 2.0 (Graphpad Software, San Diego, CA). Data were fit to a sigmoid function of the following form:
where Bottom is the Y value at the bottom of the curve, Top is the Y value at the top of the curve, and LogEC₅₀ is the logarithm of the EC₅₀, the concentration that gives a response halfway between Bottom and Top.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

To characterize [¹²⁵I]MK-801 binding to the membrane preparations used in this study, a saturation curve and Scatchard analysis was performed (n = 3, data not shown). The equilibrium dissociation constant (K_d) was 1.06 ± 0.17 nM and the maximal binding (B_max) was 1.99 ± 0.14 pmol/mg protein for brain membranes. [¹²⁵I]MK-801 bound to membranes from cells transfected with NR1a/NR2A with a K_d of 3.84 ± 0.21 nM and a B_max of 0.72 ± 0.01 pmol/mg protein. [¹²⁵I]MK-801 bound to membranes from cells transfected with NR1a/NR2B with a K_d of 2.88 ± 0.25 nM and a B_max of 0.55 ± 0.01 pmol/mg protein. It is possible that our incubation conditions did not result in the saturation assays reaching equilibrium. However, the predominant effect of this would be a small underestimate of the affinity of [¹²⁵I]MK-801 (Reynolds, 1990).

To determine the effects of polyamines on [¹²⁵I]MK-801-binding properties on recombinant NMDA receptors, polyamine concentration response curves were generated with rat brain membranes and membranes of cells transfected with NR1a/NR2A or NR1a/NR2B. The polyamine agonists spermine and spermidine produced biphasic concentration-response curves with native NMDA receptors (Fig. 1, A and B) as described previously (Reynolds and Miller, 1989; Sacaan and Johnson, 1990). Both spermine and spermidine produced a biphasic curve with NR1a/NR2B receptors. The peaks of the biphasic curves with native NMDA receptors were >600% of control whereas the peak with NR1a/NR2B receptors was approximately 400% of control. However, neither spermine nor spermidine affected [¹²⁵I]MK-801 binding in NR1a/NR2A receptors (Fig. 1, A and B).

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Polyamine agonist effects on [125I]MK-801 binding in brain membranes and recombinant NMDA receptors. Spermine (A), spermidine (B), and DEAP (C) concentration-response curves were performed in brain membranes (), NR1a:NR2A (), and NR1a:NR2B () receptors in the presence of 100 µM glutamate and 30 µM glycine. The data represent the mean ± S.E.M. of three experiments performed in duplicate.

We also characterized the effect of the polyamine agonist 1,5-(diethylamino)piperidine (DEAP) on brain membranes, NR1a/NR2A and NR1a/NR2B receptors. DEAP is thought to be selective for the agonist polyamine site and only enhanced [¹²⁵I]MK-801 binding in brain membranes (EC₅₀ = 9.6 µM) (Fig. 1C). However, DEAP inhibited binding in NR1a/NR2A receptors with an IC₅₀ of 8.0 µM. DEAP produced a small increase in binding to NR1a/NR2B receptors with a peak at 150% of control (Fig. 1C) that reversed at the highest concentration of DEAP used.

Next, we examined the effects of polyamine antagonists on [¹²⁵I]MK-801 binding in recombinant receptors. We performed concentration-response curves with arcaine and a novel polyamine antagonist, N,N'-bis(propyl)guanidinium (BPG) (Sharma et al., 1999). Both arcaine and BPG inhibited [¹²⁵I]MK-801 binding in brain membranes (Fig. 2). Similar effects were demonstrated with arcaine in brain membranes, NR1a/NR2A receptors, and NR1a/NR2B recombinant receptors, producing an IC₅₀ = 4.6 µM, IC₅₀ = 8.4 µM, and IC₅₀ = 14.1 µM respectively (Fig. 2A). BPG inhibited [¹²⁵I]MK-801 binding in brain membranes (IC₅₀ = 0.8 µM), NR1a/NR2A (IC₅₀ = 2.2 µM), and NR1a/NR2B receptors (IC₅₀ = 2.1 µM) (Fig. 2B).

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Polyamine antagonists effects on [125I]MK-801 binding in brain membranes and recombinant NMDA receptors. Arcaine (A) and BPG (B) concentration-response curves were performed in brain membranes (), NR1a:NR2A (), and NR1a:NR2B () receptors in the presence of 100 µM glutamate and 30 µM glycine. The data represent the mean ± S.E.M. of three experiments performed in duplicate.

To examine the competitive antagonism between arcaine and BPG with spermidine, arcaine and BPG concentration-response curves were generated in the absence (control) or presence of 100 µM or 1 mM spermidine (Figs. 3 and 4). This paradigm was performed in brain membranes and the recombinant receptors NR1a/NR2A and NR1a/NR2B. As the spermidine concentration was increased, both arcaine and BPG concentration-response curves in brain membranes, NR1a/NR2A, and NR1a/NR2B receptors shifted to the right (Figs. 3 and 4). The IC₅₀ values for each concentration-response curve are summarized in Table 1. The Hill slopes for the arcaine concentration-response curves vary between conditions. In all receptor membrane preparations, the Hill slope of the arcaine concentration-response curves became steeper as the concentration of spermidine increased. Interestingly, the Hill slopes of NR1a/NR2A and NR1a/NR2B membranes under the control condition was shallower (~ 0.55) than the control condition in native brain membranes. The basis for the variation in Hill slopes between the various experiments with arcaine is not clear. However, all Hill slopes for the BPG concentration-response curves were in proximity to 1.0, ranging from 0.99 to 1.35. This might indicate that BPG is more selective than arcaine for the polyamine site on the NMDA receptor.

View larger version (12K): [in this window] [in a new window]   Fig. 3.   Effects of spermidine on the inhibition of [125I]MK-801 binding by arcaine in brain membranes and recombinant NMDA receptors. Arcaine concentration-response curves were performed in the absence () or presence of 100 µM () or 1 mM () spermidine in brain membranes (A), NR1a/NR2A receptors (B), and NR1a/NR2B receptors (C). Binding assays contained 100 µM glutamate and 30 µM glycine. The data represent the mean ± S.E.M. of three experiments performed in duplicate.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Effects of spermidine on the inhibition of [125I]MK-801 binding by BPG in brain membranes and recombinant NMDA receptors. BPG concentration-response curves were performed in the absence () or presence of 100 µM () or 1 mM () spermidine in brain membranes (A), NR1a/NR2A receptors (B), and NR1a/NR2B receptors (C). Binding assays contained 100 µM glutamate and 30 µM glycine. The data represent the mean ± S.E.M. of three experiments performed in duplicate.

To further examine the DEAP-binding site in NR1a/NR2A receptors that caused [¹²⁵I]MK-801-binding inhibition, we performed DEAP concentration-response curves in the absence (control) or presence of 100 µM or 1 mM spermidine (Fig. 5). As we increased the spermidine concentration, the DEAP curves shifted to the right, increasing the IC₅₀ values from 25.7 µM in control to 91.1 µM in the presence of 100 µM spermidine and to 352.9 µM in the presence of 1 mM spermidine.

View larger version (12K): [in this window] [in a new window]   Fig. 5.   Effects of spermidine on the inhibition of [125I]MK-801 binding by DEAP in NR1a/NR2A receptors. DEAP concentration-response curves were performed in the absence () or presence of 100 µM () or 1 mM () spermidine in NR1a/NR2A receptors. Binding assays contained 100 µM glutamate and 30 µM glycine. The data represent the mean ± S.E.M. of three experiments performed in duplicate.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In this study, we report the effects of polyamines on [¹²⁵I]MK-801 binding to NR1a/NR2A and NR1a/NR2B receptors. The recombinant NR1a/NR2A and NR1a/NR2B receptors exhibited slightly lower affinity and decreased B_max values compared with native NMDA receptors in forebrain by [¹²⁵I]MK-801 saturation analysis. As reported previously, [¹²⁵I]MK-801 binding to NR1a/NR2B receptor combinations was enhanced by spermidine at low micromolar concentrations and then inhibited at concentrations greater than 100 µM (Lynch et al., 1995), whereas spermidine had no effect in NR1a/NR2A receptors. Another endogenous polyamine agonist, spermine, had similar effects on [¹²⁵I]MK-801 binding NR1a/NR2B and no effect on NR1a/NR2A. Surprisingly, the polyamine antagonists arcaine and BPG inhibited [¹²⁵I]MK-801 binding in both NR1a/NR2A and NR1a/NR2B receptors. These observations suggest that both NR1a/NR2A and NR1a/NR2B receptors contain a binding site for the inhibitory component of polyamine action. However, the stimulatory polyamine site is only found in the NR1a/NR2B recombinant receptors.

To further investigate the pharmacology of polyamine-binding sites on recombinant receptors, we used the specific polyamine site agonist DEAP. DEAP has been shown to increase [¹²⁵I]MK-801 binding in rat brain membranes similar to spermine and spermidine without the low-affinity inhibition of [¹²⁵I]MK-801 binding (Reynolds, 1992). Very different results were seen in the recombinant receptors compared to native receptors. DEAP produced a slight increase of [¹²⁵I]MK-801 binding in NR1a/NR2B receptors (150% of control) and then inhibition to return the binding to 100% control. However, NR1a/NR2A receptors showed a pronounced decrease in [¹²⁵I]MK-801 binding. This stands in contrast to the conclusion that DEAP specifically activates the stimulatory NMDA receptor-associated polyamine site as previously described by this laboratory (Reynolds, 1992). Our observations showed that the effects of DEAP is dependent on subunit combination and that DEAP does not solely bind to the site involved in the stimulatory component of the effects of polyamines. It is interesting that spermine and spermidine do not inhibit [¹²⁵I]MK-801 binding in NR1a/NR2A like DEAP. This suggests that the structure of DEAP is unique enough when compared to spermine and spermidine to permit it to bind to an inhibitory polyamine site. DEAP contains an aromatic spacer that is not found in spermine or spermidine that could account for the ability of DEAP to bind to the inhibitory polyamine site found on NR1a/NR2A receptors.

It has been shown that arcaine antagonizes the enhancement of [¹²⁵I]MK-801 binding produced by spermine, spermidine and DEAP (Reynolds, 1990, 1992). To determine whether the polyamine antagonists arcaine and BPG are competing with spermidine at a polyamine-binding site, we performed arcaine and BPG concentration-response curves on brain membranes, NR1a/NR2A and NR1a/NR2B receptors in the absence or presence of increasing spermidine concentrations (Table 1). The increase in spermidine concentration shifted the response curves to the right and increased the IC₅₀ values. This is surprising given that spermidine alone did not alter the binding of [¹²⁵I]MK-801 to the NR1a/NR2A receptors, and suggests that this modulatory effect is mediated by a distinct binding site. However, the IC₅₀ values were not increased by 10-fold, suggesting that arcaine and BPG are not competing with spermidine for the same binding site but that spermdine is binding to a site that allosterically modulates the inhibitory polyamine site. This also suggests that arcaine and BPG were binding to a site on the recombinant receptors that mirrored the binding site on native NMDA receptors.

To examine the inhibitory effect on [¹²⁵I]MK-801 binding by DEAP in NR1a/NR2A receptors, we performed DEAP concentration-response curves in the absence or presence of increasing spermidine concentrations (Fig. 5). This experiment produced results similar to those of the arcaine and BPG-response curves in the presence of increasing spermidine. Spermidine shifted the DEAP-response curves to the right and increased the IC₅₀ values. It appears that DEAP is acting like a polyamine antagonist in NR1a/NR2A recombinant receptors. This inhibitory effect of DEAP on [¹²⁵I]MK-801 binding must be masked in native NMDA receptors by the prominent enhancement of [¹²⁵I]MK-801 binding. However, it can be observed in NR1a/NR2B receptors, as evidenced by the inversion of the concentration-response curve at higher concentrations.

Perhaps the most clear conclusion from this set of experiments is that the receptors expressed in this cell line do no recapitulate the pharmacological properties of NMDA receptors obtained from rat brain membrane preparations. There are several possible reasons for this disparity. One notable difference is in the species of origin of the receptors. The ML(tk) cells express human NMDA receptors (Grimwood et al., 1996) rather than rat receptors. However, it is unlikely that the species difference can account for the differences alone, because MK-801 binding to human brain membranes is stimulated by spermidine in a similar way to that observed in rat (Steele et al., 1990). It is also possible that heterotrimeric receptors (i.e., receptors that contain NR1, NR2A, and NR2B) represent the predominant receptor in the brain (Sheng et al., 1994; Luo et al., 1997), and that the combination of all three subunits is necessary to replicate the pharmacology of the polyamine sites. Finally, it is also possible that there are differences in the post-translational processing of NMDA receptors between fibroblasts and neurons, although rather little evidence exists to suggest that this is an important factor in governing the pharmacological properties of NMDA receptors. Given the ubiquitous presence of polyamines and divalent cations in biological systems, it is also possible that there is a differential contamination of our membrane preparations with polyamines, or Ca²⁺ or Mg²⁺, although the membrane preparation procedures are intended to avoid this problem.

One of the original goals of this study was to provide a more robust basis for the comparison of results from electrophysiological and radioligand-binding studies. On the basis of the properties identified here, it remains reasonable to suggest that the glycine-independent enhancement of electrophysiological responses corresponds to the stimulation of [¹²⁵I]MK-801 binding associated with the NR1a/NR2B combination. Because the binding assays were performed in saturating concentrations of glutamate and glycine we are not in a position to assess the subunit dependence of alterations in glycine and glutamate site affinity. However, it is possible that the voltage-dependent inhibition of NMDA receptors is associated with the inhibition of binding that occurs with high concentrations of spermidine and spermine under most conditions, presumably at a site linked to that recognized by arcaine and BPG. More extensive comparisons await electrophysiological studies using DEAP and BPG.

Based on the data obtained from these experiments, we have formulated a three-site model (Table 2) that explains the binding characteristics of [¹²⁵I]MK-801 to brain membranes and NR1a/NR2A and NR1a/NR2B receptors. Site 1 is the binding site that potentiates [¹²⁵I]MK-801 binding to the NMDA receptor by binding spermine, spermidine, or DEAP. Site 2 is a low-affinity inhibitory binding site for spermine and spermidine. At concentrations greater than 100 µM, spermine and spermidine now inhibit [¹²⁵I]MK-801 binding, producing a biphasic binding curve. Site 3 is the predominant inhibitory site at which DEAP, arcaine, and BPG bind to decrease [¹²⁵I]MK-801 binding. Site 3 is allosterically modulated by site 2. This is demonstrated by adding increasing concentrations of spermidine to arcaine, BPG, or DEAP concentration-response curves. As the spermidine concentrations were increased, the IC₅₀ values increased but not by 10-fold to assume competitive antagonism. Hence, spermidine does not appear to be competitive with arcaine, BPG, or DEAP but probably alters the concentration of polyamine antagonist bound by an allosteric interaction.

According to this model, native NMDA receptors from rat forebrains contain all three polyamine-binding sites found in Table 2. NR1a/NR2A receptors contain sites 2 and 3. NR1a/NR2B receptors contain all three binding sites, similar to forebrain NMDA receptors. A concentration-dependent increase and then decrease in [¹²⁵I]MK-801 binding by spermine and spermidine acting on sites 1 and 2, respectively, was seen with native NMDA receptors and NR1a/NR2B receptors. DEAP produced an increase in [¹²⁵I]MK-801 binding in brain membranes. This increase in [¹²⁵I]MK-801 binding by DEAP was not as robust as spermine and spermidine which caused a 600 to 800% increase over control [¹²⁵I]MK-801 binding. DEAP produced a 400% change in binding. This smaller effect on binding may be due to the fact that DEAP is also binding to the predominant inhibitory site (site 3), decreasing DEAP's maximal stimulatory effect.

In conclusion, our present data lends itself to a three-site model to explain the effects of polyamine on [¹²⁵I]MK-801 binding to the NMDA receptor complex. Our data also suggest that DEAP is not selective for the stimulatory polyamine site but binds to two polyamine sites, one that is the stimulatory site which spermine and spermidine exhibit their potentiating effects (site 1) and another that is an inhibitory site that also binds arcaine and BPG (site 3). Also, our findings suggest that arcaine and BPG do not competitively antagonize spermine or spermidine which differs from previous conclusions. They appear to act at a site distinct from the site responsible for increasing binding which can be allosterically modulated by the low-affinity inhibitory site when occupied by spermine or spermidine. The precise relationship between these sites and those described electrophysiologically remains to be determined.