Zinc modulates bidirectional hippocampal plasticity by effects on NMDA receptors

Abstract

Zinc has complex effects on NMDA receptors (NMDARs) and may be an endogenous modulator of synaptic plasticity. In the CA1 region of rat hippocampal slices, we observed that low micromolar concentrations of zinc depress NMDAR synaptic responses by 40-50% and inhibit long-term depression (LTD) but not long-term potentiation (LTP). A combination of zinc plus ifenprodil, an inhibitor of NR1/NR2B receptors, produced no greater inhibition of synaptic NMDARs than either agent alone, suggesting overlapping effects on NMDARs. Similar to low micromolar zinc, ifenprodil inhibited LTD but not LTP. In contrast, low concentrations of 2-amino-5-phosphonovalerate (APV) did not block either LTP or LTD despite producing >50% inhibition of synaptic NMDARs. NVP-AAM077 ([(R)-[(S)-1-(4-bromo-phenyl)-ethylamino]-(2,3-dioxo-1,2,3,4-tetrahydro-quinoxalin-5-yl)-methyl]phosphonic acid), an antagonist with relative NR1/NR2A selectivity at low concentrations, also inhibited synaptic NMDARs by approximately 50% at 0.05 mum but failed to completely block either LTP or LTD. These results suggest that LTD induction depends on specific NMDARs with sensitivity to low micromolar zinc and ifenprodil, but LTP is less dependent on specific NMDAR subtypes. Because high-affinity sites of NR2A are likely occupied by ambient zinc, we also examined effects of extracellular zinc chelators. Zinc chelation blocked LTP but had no effect on LTD. This LTP inhibition was overcome by APV and NVP-AAM077 but not ifenprodil, suggesting that zinc chelation unmasks tonic NR1/NR2A activation that negatively modulates LTP.

Figure 1.

Effects of antagonists on NMDAR EPSPs. A, NVP-AAM077 (white circles), d,l-APV (black circles), and zinc (black triangles) inhibited NMDAR EPSPs in a concentration-dependent manner. Traces to the right show representative responses from slices treated with these antagonists. B, Ifenprodil (10 μm; hatched bar) depressed NMDAR EPSPs by ∼50%. Application of 10 μm d,l-APV (white bar) in the presence of ifenprodil caused nearly complete depression (black circles). In the absence of ifenprodil, 10 μm d,l-APV only partially depressed NMDAR EPSPs (white circles). Traces to the right show a control response and the effects of APV (top) and ifenprodil alone and with APV (bottom). Numbers on the traces denote the time when traces were sampled. C, Zinc at 10 μm (gray bar) also partially inhibited NMDAR EPSPs. Addition of 10 μm APV (white bar) further suppressed these responses. Traces to the right show a control response and the effects of zinc alone and APV plus zinc. D, Addition of 10 μm zinc (gray bar) to 10 μm ifenprodil (hatched bar) did not further suppress NMDAR EPSPs. Traces below show inhibition by ifenprodil and the failure of zinc to cause additional inhibition in the presence of ifenprodil. E, After administration of 0.05 μm NVP-AAM077 (white bar), 10 μm zinc (gray bar) further depressed NMDAR EPSPs. Traces show representative waveforms. Calibration: 1 mV, 5 ms for all.

Figure 2.

Effects of NMDR antagonists and zinc on LTP. A, HFS (100 Hz, 1 s; arrow) failed to induce LTP in the presence of both 10 μm d,l-APV (white bar) and 10 μm ifenprodil (hatched bar) but induced LTP in the presence of ifenprodil alone. B, HFS (arrow) induced LTP in control slices (white triangles) and in the presence of 10 μm zinc (gray circles) but not in the presence of 100 μm zinc (black circles). Zinc was applied for 15 min before the tetanus (bars). C, HFS induced LTP in the presence of 10 μm APV (white circles). The combination of 10 μm zinc (gray bar) and 10 μm APV (white bar) inhibited LTP induction (gray circles). D, At 0.05 μm, NVP-AAM077 (white bar) did not completely inhibit LTP (white circles), whereas the combination of NVP-AAM077 and 10 μm zinc (gray bar) prevented LTP induction (gray circles).

Figure 3.

Effects of NMDAR antagonists and zinc on LTD. A, LFS (1 Hz, 900 s, vertical striped bar) failed to induce LTD in the presence of 10 μm ifenprodil (hatched bar). B, LFS induced LTD in the presence of 10 μm (white circles) but not 100 μm d,l-APV (black circles). C, LFS induced LTD in the presence of 0.05 μm NVP-AAM077. D, LFS failed to induce LTD in the presence of 10 μm zinc (gray bar) but after zinc washout LTD was induced. Traces on the right are field EPSPs obtained before (dashed traces) and 60 min after (solid traces) LFS. Calibration: 1 mV, 5 ms.

Figure 4.

NVP-AAM077 prevents NMDA-mediated LTP inhibition. As reported previously (Izumi et al., 1992), a 100 Hz, 1 s HFS (arrow) delivered in the presence of 1 μm NMDA (white bar) inhibits LTP induction (white circles; n = 7). Administration of 0.05 μm NVP-AAM077 (black bar) with NMDA allowed LTP (black circles; n = 4), whereas 10 μm ifenprodil (gray triangles; n = 3) was ineffective. Drugs were administered for the duration noted by the bars, and the tetanus (HFS) was delivered at the time denoted by the arrow. These results indicate that LTP inhibition mediated by untimely activation of NMDA receptors results from activation of NR1/NR2A receptors. We reported previously that 10 μm d,l-APV overcomes the LTP inhibition by 1 μm NMDA (Izumi et al., 1992a,b). Traces depict EPSPs before and 60 min after HFS. Calibration: 1 mV, 5 ms.

Figure 5.

Effects of NMDAR antagonists on EDTA-mediated LTP inhibition. A, Administration of 2 mm Ca-EDTA for 15 min does not alter NMDAR EPSPs (104.0 ± 4.6% of control; n = 4). In contrast, administration of 2 mm Zn-EDTA for 15 min partially inhibited NMDAR EPSPs (34.3 ± 14.9% of control; n = 5). B, The graph shows the effects of 15 min perfusion of 2 mm Ca-EDTA (white circles) or 2 mm Zn-EDTA (black circles) on LTP induction. HFS (arrow) was delivered in the presence of EDTA (bars). The results indicate that a Ca-EDTA chelatable, but not Zn-EDTA chelatable, agent inhibits LTP induction. C, The graph shows the effects of 10 μm d,l-APV (black circles), 0.05 μm NVP-AAM 077 (gray circles), and 10 μm ifenprodil (white triangles) on Ca-EDTA-mediated LTP inhibition. Whereas APV and NVP-AAM077 allowed LTP induction, ifenprodil failed to promote LTP in the presence of Ca-EDTA, suggesting that LTP inhibition by zinc chelation results from activation of NMDARs expressing NR1/NR2A but not NR1/NR2B. Traces to the right show representative EPSPs before (dashed traces) and 60 min after (solid traces) the tetanus. Calibration: 1 mV, 5 ms.

Figure 6.

Other zinc chelators inhibit LTP. A, PHN, a zinc chelator, mimics the effects of 2 mm Ca-EDTA. HFS failed to induce LTP in the presence of 200 μm PHN (white bar, white triangles). Coadministration of 10 μm APV (black bar) with PHN allowed LTP induction (n = 5; black triangles). B, Similarly, in the presence of 50 μm TPEN, a third zinc chelator, LFS failed to induce LTP. TPEN administered with 10 μm zinc (black bar) allowed LTP induction. Traces depict EPSPs before and 60 min after HFS. Calibration: 1 mV, 5 ms.