Loss of long-term depression in the insular cortex after tail amputation in adult mice

Abstract

The insular cortex (IC) is an important forebrain structure involved in pain perception and taste memory formation. Using a 64-channel multi-electrode array system, we recently identified and characterized two major forms of synaptic plasticity in the adult mouse IC: long-term potentiation (LTP) and long-term depression (LTD). In this study, we investigate injury-related metaplastic changes in insular synaptic plasticity after distal tail amputation. We found that tail amputation in adult mice produced a selective loss of low frequency stimulation-induced LTD in the IC, without affecting (RS)-3,5-dihydroxyphenylglycine (DHPG)-evoked LTD. The impaired insular LTD could be pharmacologically rescued by priming the IC slices with a lower dose of DHPG application, a form of metaplasticity which involves activation of protein kinase C but not protein kinase A or calcium/calmodulin-dependent protein kinase II. These findings provide important insights into the synaptic mechanisms of cortical changes after peripheral amputation and suggest that restoration of insular LTD may represent a novel therapeutic strategy against the synaptic dysfunctions underlying the pathophysiology of phantom pain.

Figure 1

Loss of LFS-evoked LTD in the superficial layer of the IC after tail amputation. (A) A schematic diagram showing the tail amputation model (up) and the experimental procedure (lower). All insular slices are obtained at 2 weeks after amputation in the present study. (B) Light microscopy photograph showing the relative location of the IC slice with the MED64 probe, the stimulation site (red dot) and the layer designation. (C and D) An overview of the 64-channel multi-electrode array recordings in the tail-amputated IC slice (C: before LFS; D: 60 min after LFS). No synaptic depression was revealed. Red dots mark the stimulation sites in the deep layer. Calibration: 100 μV, 10 ms. (E) Pooled data of LFS-elicited LTD in the superficial layer of the IC for sham control (n = 9 slices/7 mice) and tail amputation (n = 9 slices/5 mice) groups. The sham group showed typical LTD lasting for 1 h, while tail amputation abolished the LTD induction. Sample fEPSP recordings taken at the times indicated by the corresponding numbers are shown above the plot. Calibration: 100 μV, 10 ms. Horizontal bars denote the period of LFS delivery. Error bars represent SEM.

Figure 2

Loss of LFS-evoked LTD in the deep layer of the IC after tail amputation. (A) Pooled data of LFS-elicited LTD in the deep layer of the IC for sham control (n = 9 slices/7 mice) and tail amputation (n = 9 slices/5 mice) groups. Tail amputation also abolished the LTD induction in the deep layer. Sample fEPSP recordings taken at the times indicated by the corresponding numbers are shown above the plot. Calibration: 100 μV, 10 ms. Horizontal bars denote the period of LFS delivery. (B) Bar histogram showing the induction ratio of LTD (the percentage of LTD-showing channels among all activated channels) in the superficial layer and deep layer of the IC for sham control (n = 9 slices/7 mice) and tail-amputated (n = 9 slices/5 mice) groups. ***P < 0.001. Error bars represent SEM.

Figure 3

Tail amputation has no effect on DHPG-LTD in the IC. (A and B) Pooled data showing the comparable magnitude and temporal progression of DHPG-LTD between sham control (superficial layer: n = 7 slices/7 mice, A; deep layer: n = 8 slices/8 mice, B) and tail amputation (superficial layer: n = 9 slices/9 mice, A; deep layer: n = 9 slices/9 mice, B) groups. Sample fEPSP recordings taken at the times indicated by the corresponding numbers are shown above the plot. Calibration: 100 μV, 10 ms. Horizontal bars denote the period of DHPG application. (C) The induction ratio of DHPG-LTD in either superficial layer or deep layer does not differ between the two groups (sham control: n = 8 slices/8 mice; tail amputation: n = 9 slices/9 mice). NS, no significance. Error bars represent SEM.

Figure 4

Enhancement of synaptic transmission in the IC after tail amputation. (A) The input–output relationship of the fEPSP slope in the superficial layer of the IC. Shown are the percentage changes of the fEPSP slope (normalized to the slope value at 8 μA) in response to series of ascending stimulation intensities. Tail amputation (n = 6 slices/4 mice) caused a leftward shift of the input–output curve compared to the sham control group (n = 6 slices/6 mice). (B) Pooled data of the input–output relationship of the fEPSP slope in the deep layer of the IC. Similarly, tail amputation (n = 6 slices/4 mice) resulted in a leftward shift of the curve compared to the sham control (n = 5 slices/5 mice). The insets in (A) and (B) show the representative fEPSP traces recorded at 18 μA for both sham (left) and tail-amputated (right) groups. Calibration: 100 μV, 10 ms. (C) The input-out curve of the number of activated channels obtained at graded stimulation intensities in the IC slice. Significant difference was detected between the sham control (n = 6 slices/6 mice) and tail-amputated (n = 8 slices/5 mice) group at higher stimulation intensities. Error bars represent SEM.

Figure 5

Pharmacological rescue of LFS-evoked insular LTD in the tail-amputated mice by group I mGluR activation. (A and B) An overview of 64-channel multi-electrode array recordings in the tail-amputated IC slice (A: before LFS; B: 60 min after LFS). DHPG (20 μM) was applied for 20 min followed by washout for 30 min. Then LFS was given. Bath application of low dose of DHPG only elicited acute depression. However, subsequent LFS induced clear LTD after priming with DHPG. Red dots denote the stimulation sites in the deep layer. Calibration: 100 μV, 10 ms. (C and D) One representative example slice showing the rescue of LFS-induced LTD by DHPG priming in both superficial layer (C) and deep layer (D) of the IC. Inset traces show representative fEPSPs at the time points indicated by the numbers in the graph. Calibration: 100 μV, 10 ms. (E and F) Summarized data for the superficial layer (n = 6 slices/5 mice) and deep layer (n = 5 slices/5 mice). Horizontal bars denote the period of DHPG application or LFS delivery as indicated. Error bars represent SEM.

Figure 6

PKC, but not CaMKII or PKA, is involved in the rescue of LFS-evoked insular LTD in the superficial layer. (A) Vehicle control group exhibited the normal rescue of insular LTD (n = 5 slices/3 mice). (B) Co-application of the PKC inhibitor chelerythrine (Che, 3 μM) together with DHPG (20 μM, 20 min) blocked the LTD rescue (n = 6 slices/6 mice). (C) CaMKII inhibitor KN62 (5 μM) could not affect the LTD recovery (n = 6 slices/5 mice). (D) PKA inhibitor KT5720 (1 μM) had no effect on DHPG-primed insular LTD (n = 7 slices/5 mice). Inset traces in (A-D) show representative fEPSPs at the time points indicated by the numbers in the graph. Calibration: 100 μV, 10 ms. Horizontal bars denote the period of DHPG application or LFS delivery as indicated. (E) Bar histogram summarizing the averaged data within the last 10 min of the LTD recording. **P < 0.01. NS, no significance. Error bars represent SEM.

Figure 7

PKC, but not CaMKII or PKA, is involved in the rescue of LFS-evoked insular LTD in the deep layer. (A) Vehicle control group exhibited the normal rescue of insular LTD (n = 5 slices/3 mice). (B) Co-application of the PKC inhibitor chelerythrine (Che, 3 μM) together with DHPG (20 μM, 20 min) blocked the LTD rescue (n = 6 slices/6 mice). (C) CaMKII inhibitor KN62 (5 μM) could not affect the LTD recovery (n = 6 slices/5 mice). (D) PKA inhibitor KT5720 (1 μM) had no effect on DHPG-primed insular LTD (n = 7 slices/5 mice). Inset traces in (A-D) show representative fEPSPs at the time points indicated by the numbers in the graph. Calibration: 100 μV, 10 ms. Horizontal bars denote the period of DHPG application or LFS delivery as indicated. (E) Bar histogram summarizing the averaged data within the last 10 min of the LTD recording. ***P < 0.001. NS, no significance. Error bars represent SEM.