Calcium-stimulated adenylyl cyclase subtype 1 (AC1) contributes to LTP in the insular cortex of adult mice

Abstract

Long-term potentiation (LTP) of synaptic transmission in the central nervous system is a key form of cortical plasticity. The insular cortex (IC) is known to play important roles in pain perception, aversive memory and mood disorders. LTP has been recently reported in the IC, however, the signaling pathway for IC LTP remains unknown. Here, we investigated the synaptic mechanism of IC LTP. We found that IC LTP induced by the pairing protocol was N-methyl-D-aspartate receptors (NMDARs) dependent, and expressed postsynaptically, since paired-pulse ratio (PPR) was not affected. Postsynaptic calcium is important for the induction of post-LTP, since the postsynaptic application of BAPTA completely blocked the induction of LTP. Calcium-activated adenylyl cyclase subtype 1 (AC1) is required for potentiation. By contrast, AC8 is not required. Inhibition of Ca²⁺ permeable α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (CP-AMPARs) or protein kinase M zeta (PKMζ) reduced the expression of LTP. Our results suggest that calcium-stimulated AC1, but not AC8, can be a trigger of the induction and maintenance of LTP in the IC.

Keywords: Neuroscience.

Fig. 1

Postsynaptic calcium triggers IC LTP. (a) Brain diagram of adult mouse IC. (b) Top: sample traces show input-output relationship of AMPA receptor-mediated EPSCs in the IC. Bottom: Plots of input-output curve in WT mice IC (n = 8 neurons/5 mice). (c) The evoked EPSC was completely blocked by CNQX (20 μM). (d) A scheme illustrating the pairing protocol consisting of pairing 80 presynaptic pulses at 2 Hz with postsynaptic depolarization (holding at + 30 mV). (e) Sample traces at the indicated time points are shown above the plot. Top: sample traces of EPSCs with single-pulse stimulation during baseline (1) and 60 min after pairing protocol (2) at a holding membrane potential of −60 mV. (f) Pooled data to illustrate the time course of LTP (black, n = 14/12) and control (white, n = 11/11). There is significant difference between LTP and control (two-way ANOVA, F_1,46 = 70.52, ***p < 0.001). (g) Bath application of AP-5 (50 μM) completely blocked the induction of LTP in one neuron. (h) Pooled data of AP-5 (n = 5 neurons/4 mice). (i) Postsynaptic application of BAPTA (20 mM in the recording pipette) completely blocked the induction of LTP in one neuron. (j) Pooled data of BAPTA (n = 5/4). (k) Summary of AP-5 and BAPTA on the induction of LTP. The amplitudes of EPSCs in AP-5 or BAPTA were significantly decreased compared with LTP (one-way ANOVA, F_2,21 = 14.17, ***p < 0.001). Calibration, 50 pA, 50 ms. The mean amplitudes of EPSCs were determined at 50–60 min after pairing protocol. The arrow donates the time of pairing protocol. Error bars represent SEM; ^***p < 0.001.

Fig. 2

Pairing protocol purely induce post-LTP. (a) Top: sample traces of EPSCs with paired-pulse stimulation during baseline (1) and 60 min after the pairing protocol (2) at a holding membrane potential of −60 mV. Bottom: a time course plot of a representative single example. (b) Pooled data to illustrate the time course of PPR (n = 7/5). (c) Summary of PPR data before and after the pairing protocol. There is no significant difference between before and after (paired t-test, t₆ = -0.64, p > 0.05). Calibration, 100 pA, 50 ms. The arrow denotes the time of pairing protocol.

Fig. 3

Loss of LTP in the IC of AC1-/- mice. (a) The pairing stimulation failed to induce LTP in the IC in one neuron of AC1-/- mouse. (b) Pooled data of AC1-/- mouse (n = 7 neurons/5 mice). (c) NB001 (20 μM) blocked the induction of LTP in one neuron. (d) Pooled data of NB001 (n = 7/5). (e) KT5720 (1 μM) blocked the induction of LTP. (f) Pooled data of KT5720 (n = 7/5). (g) Summary of the effect of AC1-/-, NB001 and KT5720 on the induction of LTP. The amplitudes of EPSCs in AC1-/-, NB001 or KT5720 were significantly decreased compared with control LTP (one-way ANOVA, F_3,37 = 13.82, ***p < 0.001). Insets in a,b and e are example EPSC traces at time points indicated by the numbers in the graph. Calibration, 50 pA, 50 ms. The arrow denotes the time of pairing protocol. Error bars represent SEM; ^***p < 0.001.

Fig. 4

AC8^−/− mice show normal LTP in the IC. (a) LTP in the IC was induced in one neuron of AC8^−/− mouse. Sample traces at the indicated time points are shown above the plot. Calibration, 100 pA, 50 ms. (b) Pooled data to illustrate the induction of LTP in AC8-/- mice (n = 5 neurons/4 mice). The arrow denotes the time of pairing protocol. Error bars represent SEM.

Fig. 5

NMDAR-mediated EPSCs and effect of PPR in AC1^−/− mice. (a) Sample trace showing the input-output relationship of NMDAR-mediated EPSCs in a WT mice IC neuron. (b) There was no difference in the input-output curve of NMDAR-mediated EPSCs in the IC between WT mice (square, n = 7 neurons/5 mice), AC1-/- (triangle, n = 5/5) or AC8-/- (circle, n = 12/6) (two-way ANOVA, F_2,147 = 0.1, p > 0.05). (c) Sample trace showing NMDAR-mediated EPSCs evoked at holding potentials of −85 mV to approximately +55 mV in a WT IC neuron. (d) Current-voltage plots for NMDAR-mediated EPSCs between WT mice (n = 7/6), AC1-/- mice (n = 5/4) and AC8^−/− (n = 10/7) in the IC (two-way ANOVA, F_2,152 = 0.23, p > 0.05). (e) Pooled data of PPR in AC1-/- mice (n = 5/3). (f) Pooled data of NB001 (n = 5/3). (g) Summary of PPR data before and after the pairing protocol. There is no significant difference among baseline, AC1-/- and NB001 (one-way ANOVA, F_2,12 = 0.28, p > 0.05). Calibration, 100 pA, 50 ms. The arrow denotes the time of pairing protocol. Error bars represent SEM.

Fig. 6

GluA1 and PKMζ are involved in the maintenance of IC LTP. (a) Pooled data of NASPM (50 μM) on the maintenance of LTP (n = 5 neurons/3 mice), applied 20 min after the LTP induction. (b) Pooled data of ZIP (5 μM) on the maintenance of LTP in the IC, applied 20 min after the LTP induction (n = 5/5). (c) Summary of effect for the maintenance of LTP. There are significant differences comparing LTP with NASPM and ZIP (one-way ANOVA, F_2,21 = 18.94, p < 0.001). (d) Left: A simplified diagram shows that insular cortex (IC) receives projections from the spinal cord and amygdala. Right: A model for IC LTP. Activation of glutamate NMDARs triggers an increase in postsynaptic Ca²⁺ in dendritic spines. Ca²⁺ is an important intracellular signal for triggering a series of biochemical events that contribute to the induction and expression of LTP. After the activation of NMDARs, Ca²⁺ binds to calmodulin (CaM) and leads to activation of Ca²⁺-stimulated AC1 as well as Ca²⁺/CaM-dependent protein kinases. Subsequently, postsynaptic GluA1-containing AMPA receptors may be transferred into synaptic sites and contribute to enhanced synaptic responses. In addition, activation of AC1 leads to activation of PKA. Activation of cAMP-PKA drives the insertion of GluA1 homomers (CP‐AMPARs). PKMζ may maintain LTP by upregulating GluA1–GluA2 heteromers. Insets in a and b are example eEPSC traces at time points indicated by the numbers in the graph. The arrow denotes the time of pairing protocol. Calibration, 50 pA, 50 ms. Error bars represent SEM; ^***p < 0.001.