Two distinct oscillatory states determined by the NMDA receptor in rat inferior olive

Abstract

1. The effects of N-methyl-D-aspartate (NMDA) receptor activation and blockade on subthreshold membrane potential oscillations of inferior olivary neurones were studied in brainstem slices from 12- to 21-day-old rats. 2. Dizocilpine (MK-801), a non-competitive NMDA antagonist, at 1-45 microM abolished spontaneous subthreshold oscillations, without affecting membrane potential, input resistance, or the low-threshold calcium current, I(T). Ketamine (100 microM), a non-competitive NMDA antagonist, and L-689,560 (20 microM), an antagonist at the glycine site of the NMDA receptor, also abolished the oscillations, while the competitive non-NMDA antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; 20-50 microM) had no effect. 3. NMDA (100 microM) induced 4.1 Hz subthreshold oscillations and reversibly depolarized olivary neurones by 13.7 mV. In contrast, 10 microM alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA) and 20 microM kainic acid depolarized the membrane equivalently but did not induce oscillations. 4. Both NMDA-induced and spontaneous subthreshold oscillations were unaffected by 1 microM tetrodotoxin and were prevented by substituting extracellular calcium with cobalt. 5. Removing magnesium from the perfusate did not affect spontaneous subthreshold oscillations but did prevent NMDA-induced oscillations. 6. NMDA-induced oscillations were resistant to 50 microM mibefradil, an I(T) blocker, in contrast to spontaneous oscillations. Both oscillations were inhibited by 20 microM nifedipine, an L-type calcium channel antagonist, and 200 nM omega-agatoxin IVA, a P-type calcium channel blocker. Bay K 8644 (10 microM), an L-type Ca(2+) agonist, significantly enhanced the amplitude of both spontaneous and NMDA-induced oscillations. 7. The data indicate that NMDA receptor activation induces olivary neurones to manifest high amplitude membrane potential oscillations in part mediated by L- and P- but not T-type calcium currents. Moreover, the data demonstrate that NMDA receptor currents are necessary for generation of spontaneous subthreshold oscillations in the inferior olive.

Figure 1. MK-801 suppresses spontaneous oscillations without affecting I_T

A, recordings from a representative olivary neurone showing that 1 μm MK-801 suppresses spontaneous subthreshold oscillations without significantly changing membrane potential. B, cumulative statistics on the effects of MK-801 concentration (1, 5, 15 and 45 μm) on the time required for abolition of oscillations (□) and percentage of cells whose oscillations were abolished by MK-801 (♦). MK-801, at all doses, was successful in 100 % of the experiments at annihilating the oscillations. C, 15 μm MK-801 did not affect the low-threshold Ca²⁺ spike (open arrow) triggered by I_T. D, the amplitude and slope of the low-threshold calcium spike were not altered by 15 μm MK-801.

Figure 2. NMDA but not AMPA or kainate receptors are required for spontaneous oscillations

A, treatment of olivary neurones with 100 μm ketamine, a non-competitive antagonist at NMDA receptors, annihilated spontaneous oscillations. B, application of 20 μm L-689,560, an antagonist at the glycine site of the NMDA receptor, also abolished spontaneous oscillations. C, in contrast, 50-100 μm APV, a competitive NMDA antagonist, failed to suppress oscillations. D, spontaneous oscillations were not affected by 20-50 μm CNQX, a competitive non-NMDA receptor antagonist.

Figure 3. NMDA depolarizes the membrane and induces subthreshold oscillations

A, recording from an olivary neurone showing the depolarization and induction of oscillations by 100 μm NMDA. The oscillations persisted when membrane potential was returned to the control value with DC bias and were reversed by removing NMDA from the bath. B, cumulative statistics on the effects of NMDA concentration (10, 25, 50 and 100 μm) on membrane depolarization (□) and percentage of cells with oscillations in the presence of NMDA (♦).

Figure 4. NMDA-induced oscillations differ from spontaneous oscillations in their frequency and amplitude

A, NMDA at 100 μm depolarized the membrane and appeared to slow the oscillation frequency reversibly. B, superposition of spontaneous (continuous line) and NMDA-induced (dotted line) oscillations from the same neurone indicated that NMDA-induced oscillations had lower frequency and larger amplitude (left). Power spectrum analysis confirmed that the two oscillatory states were of different frequency (right). The power spectral data were normalized to the peak coefficient of the NMDA spectrum. C, plot of the oscillation frequency of 15 neurones before and during NMDA application. D, cumulative probability distribution of peak spectral power for the 15 neurones during spontaneous (continuous line) and NMDA-induced oscillations (dotted line). E, the amplitude of NMDA-induced oscillations was larger than the amplitude of spontaneous oscillations. *P < 0.05.

Figure 5. AMPA and kainic acid do not induce oscillations

A, application of 100 μm NMDA depolarized the membrane and induced subthreshold oscillations. Same neurone as in Fig. 2A. B, application of 10 μm AMPA significantly depolarized the membrane but did not induce oscillations. C, treatment with 20 μm kainic acid also depolarized the membrane without inducing oscillations.

Figure 6. Basic properties of NMDA-induced oscillations

A, spontaneous subthreshold oscillations were not affected by the presence of 1 μm TTX. Subsequent application of 100 μm NMDA significantly depolarized the membrane and induced oscillations with frequency and amplitude indistinguishable from those obtained with NMDA alone. B, when Ca²⁺ was removed from the perfusate and replaced with an equimolar amount of Co²⁺, spontaneous oscillations were abolished and 100 μm NMDA subsequently failed to depolarize the membrane and induce oscillations.

Figure 7. Mg²⁺ is required for NMDA-induced but not spontaneous oscillations

A, spontaneous subthreshold oscillations were not affected by removing Mg²⁺ from the perfusate. B, removing Mg²⁺ from the perfusate did not impair the ability of 100 μm NMDA to depolarize the neurone but did prevent the induction of oscillations. The effects of NMDA in normal ACSF are shown on the left for comparison. C, application of 100 μm NMDA in ACSF containing 0 Mg²⁺ and 4 mm Ca²⁺ completely and irreversibly depolarized the neurone.

Figure 8. I_T is not required for NMDA-induced oscillations

A, application of 50 μm mibefradil did not affect the after-depolarization (filled arrow) produced by high-threshold Ca²⁺ currents. B, the low-threshold Ca²⁺ spike (open arrow) produced by I_T was abolished by 50 μm mibefradil. Same neurone as in A. C, mibefradil significantly reduced the amplitude and the slope of the low-threshold calcium spike. *P < 0.05. D, mibefradil (50 μm) abolished spontaneous subthreshold oscillations, but subsequent application of 100 μm NMDA induced oscillations with frequency and amplitude indistinguishable from those induced by NMDA alone.

Figure 9. Cd²⁺ abolishes spontaneous oscillations and prevents induction of oscillations by NMDA

A, the after-depolarization (filled arrow) generated by high-threshold calcium currents and the calcium-dependent after-hyperpolarization were blocked by 100 μm Cd²⁺. B, in contrast, Cd²⁺ had no effect on the low-threshold Ca²⁺ spike (open arrow). Same cell as in A. C, spontaneous subthreshold oscillations were abolished by 100 μm Cd²⁺ in the presence of 1 μm TTX. D, when 100 μm NMDA was given to olivary neurones in the presence of 100 μm Cd²⁺ and 1 μm TTX, the associated depolarization was reduced and the oscillations were completely prevented. The effects of NMDA in normal ACSF are shown on the left for comparison.

Figure 10. L-type Ca²⁺ currents amplify both spontaneous and NMDA-induced oscillations

A, application of 20 μm nifedipine in the presence of 1 μm TTX did not affect input resistance or the low-threshold Ca²⁺ spike (open arrow). B, the high-threshold Ca²⁺ spike, on the contrary, was reduced by nifedipine. C, comparison of high-threshold Ca²⁺ spike parameters before and in the presence of nifedipine revealed that the spike peak, amplitude and slope were significantly decreased by nifedipine. *P < 0.05. D, nifedipine abolished spontaneous subthreshold oscillations in the presence of 1 μm TTX. E, application of 100 μm NMDA in the presence of 1 μm TTX and 20 μm nifedipine resulted in a robust depolarization but diminished oscillations.

Figure 11. Bay K 8644 enhances the amplitude of spontaneous and NMDA-induced oscillations

A, application of 10 μm Bay K 8644 in the presence of 1 μm TTX enhanced the amplitude of spontaneous subthreshold oscillations. B, when 100 μm NMDA was given to olivary neurones in the presence of 1 μm TTX and 10 μm Bay K 8644, the amplitude of the oscillations increased significantly. The recordings are from the same neurone.

Figure 12. P-type Ca²⁺ currents contribute to spontaneous and NMDA-induced oscillations

A, high-threshold Ca²⁺ spikes elicited in the presence of 1 μm TTX were attenuated by 200 nmω-agatoxin IVA. The recordings are from 2 different neurones and were superimposed to facilitate comparison. B, statistical analysis showed that high-threshold Ca²⁺ spike amplitude and slope were reduced by ω-agatoxin IVA. *P < 0.05. C, in 2 neurones spontaneously manifesting subthreshold oscillations, 200 nmω-agatoxin IVA appeared to reduce the amplitude of the oscillations. D, application of 100 μm NMDA to a neurone in the presence of 1 μm TTX and 200 nmω-agatoxin IVA depolarized the membrane but failed to produce oscillations.