Signalling pathway of nitric oxide in synaptic GABA release in the rat paraventricular nucleus

Abstract

In the paraventricular nucleus (PVN) of the hypothalamus, nitric oxide (NO) inhibits sympathetic outflow through increased GABA release. However, the signal transduction pathways involved in its action remain unclear. In the present study, we determined the role of cGMP, soluble guanylyl cyclase, and protein kinase G in the potentiating effect of NO on synaptic GABA release to spinally projecting PVN neurones. The PVN neurones were retrogradely labelled by a fluorescent tracer injected into the thoracic spinal cord of rats. Whole-cell voltage-clamp recordings were performed on labelled PVN neurones in the hypothalamic slice. Bath application of the NO donor, S-nitroso-N-acetyl-penicillamine (SNAP), reproducibly increased the frequency of miniature GABAergic inhibitory postsynaptic currents (mIPSCs) without changing the amplitude and the decay time constant. Neither replacement of Ca2+ with Co2+ nor application of Cd2+ to block the Ca2+ channel altered the effect of SNAP on mIPSCs. Also, the effect of SNAP on mIPSCs was not significantly affected by thapsigargin, a Ca2+-ATPase inhibitor that depletes intracellular Ca2+ stores. Application of a membrane-permeant cGMP analogue, pCPT-cGMP, mimicked the effect of SNAP on mIPSCs in the presence of a phosphodiesterase inhibitor, IBMX. Furthermore, both the soluble guanylyl cyclase inhibitor, ODQ, and the specific protein kinase G inhibitor, Rp pCPT cGMP, abolished the effect of SNAP on mIPSCs. Thus, these data provide substantial new information that NO potentiates GABAergic synaptic inputs to spinally projecting PVN neurones through a cGMP-protein kinase G pathway.

Figure 1. identification of retrogradely labeled spinally projecting PVN neurons

A, a PVN neurone in the brain slice labelled with FluoSphere identified with fluorescence illumination. B, the same neurone recorded with a glass electrode (^∧) viewed with Nomarski optics. C, photomicrograph showing the FluoSphere injection site (red) at the spinal intermediolateral cell column (IML) in one rat. Note that the bright field and fluorescence images were taken from the same tissue section and superimposed to show the location and diffusion of the FluoSphere injection. Scale bar, 50 μm in A and B; 500 μm in C. DH, dorsal horn.

Figure 2. Effect of SNAP on GABAergic mIPSCs of labeled PVN neurons

A, representative tracings showing mIPSCs recorded from a labelled PVN neurone during control, application of 100 μm SNAP, and application of 20 μm bicuculline. Note that bicuculline completely abolished mIPSCs. B, superimposed averages of 100 consecutive mIPSCs obtained during control and SNAP application. The decay time constants during control (τ_fast= 4.83 ms and τ_slow= 25.75 ms) and SNAP administration (τ_fast= 4.86 ms and τ_slow= 25.89 ms) were similar. C and D, summary data showing the effect of 100 μm SNAP on the frequency (C) and the amplitude (D) of mIPSCs of nine labelled PVN neurones. E, summary data showing the decay time constants of mIPSCs of nine labelled PVN neurones during control and administration of 100 μm SNAP. Data presented as means ±s.e.m.*P < 0.05 compared to the control (Kruskal-Wallis test). Bic, bicuculline.

Figure 3. Effect of SNAP on mIPSCs of labeled PVN neurons in Ca²⁺-free solution

A, representative tracings showing the effect of 100 μm SNAP on mIPSCs in normal aCSF and Ca²⁺-free aCSF. B and C, cumulative plot analysis of mIPSCs of the same neurone as in A showing the distribution of the interevent interval (B) and peak amplitude (C) during control, application of 100 μm SNAP, and SNAP application in Ca²⁺-free aCSF. SNAP decreased the interevent interval of mIPSCs (P < 0.05, Kolmogorov-Smirnov test) without changing the distribution of the amplitude in both normal and Ca²⁺-free aCSF. D and E, summary data showing the effect of SNAP on the frequency (D) and the amplitude (E) of mIPSCs of eight labelled PVN neurones in both normal and Ca²⁺-free aCSF. Data presented as means ±s.e.m.*P < 0.05 compared to the control (Kruskal-Wallis test).

Figure 4. Effect of SNAP on mIPSCs after blockade of Ca²⁺ channels with Cd²⁺

A, raw tracings showing the effect of 100 μm SNAP and SNAP + 50 μm Cd²⁺ on mIPSCs of a labelled PVN neurone. Note that treatment with Cd²⁺ failed to alter the SNAP-induced increase in the frequency of mIPSCs. B and C, cumulative probability analysis of mIPSCs of the same neurone as in A showing the distribution of the interevent interval (B) and peak amplitude (C) during control, SNAP, and application of SNAP + Cd²⁺. SNAP decreased the interevent interval of mIPSCs (P < 0.05, Kolmogorov-Smirnov test) without changing the distribution of the amplitude in the presence of Cd²⁺. D and E, summary data showing the effect of SNAP and SNAP + Cd²⁺ on the frequency (D) and amplitude (E) of mIPSCs of nine labelled PVN neurones. *P < 0.05 compared to the control (Kruskal-Wallis test).

Figure 5. Role of introcellular Ca²⁺ store in the effect of SNAP on mIPSCs

A, Original tracings showing mIPSCs recorded from a labelled PVN neurone during control, application of 100 μm SNAP, and application of SNAP in the presence of 10 μm thapsigargin (TSG). B and C, summary data showing the effect of 100 μm SNAP on the frequency (B) and the amplitude (C) of mIPSCs before and after treatment with 10 μm thapsigargin in six labelled PVN neurones. Data presented as means ±s.e.m.*P < 0.05 compared to the control (Kruskal-Wallis test).

Figure 6. Lack of effect of pCPT-cGMP on mEPSCs in labeled PVN neurons

A, raw tracings showing the spontaneous mIPSCs during control and application of 30 μm pCPT–cGMP in the presence of 100 μm IBMX in a labelled PVN neurone. B and C, cumulative plot analysis of mIPSCs of the same neurone as in A showing the distribution of the interevent interval (B) and peak amplitude (C) during control and pCPT–cGMP application in the presence of IBMX. pCPT–cGMP decreased the interevent interval of mIPSCs (Kolmgorov-Smirnov test, P < 0.05) without changing the distribution of the amplitude. D, superimposed averages of 100 consecutive mIPSCs obtained during control and pCPT–cGMP application. The decay time constants during control (τ_fast= 4.79 ms and τ_slow= 23.12 ms) and pCPT–cGMP administration (τ_fast= 4.74 ms and τ_slow= 22.25 ms) were similar. E,F, summary data showing the effect of pCPT–cGMP on the frequency (E) and amplitude (F) of mIPSCs of nine labelled PVN neurones. Data presented as means ±s.e.m.*P < 0.05 compared to the control (Kruskal-Wallis test).

Figure 7. pCPT-cGMP potentiates mIPSCs in labeled PVN neurons

A, representative tracings from a PVN-labelled neurone showing mEPSCs during control, application of 50 μm pCPT–cGMP, and application of 20 μm CNQX in the presence of 100 μm IBMX. Note that CNQX completely eliminated mEPSCs. B and C, cumulative plot analysis of mEPSCs of the same neurone as in A showing the distribution of the interevent interval (B) and peak amplitude (C) during control and pCPT–cGMP application. Neither the frequency nor the amplitude of mEPSCs was altered by the pCPT–cGMP. D, superimposed averages of 100 consecutive mEPSCs obtained during control and pCPT–cGMP application. The decay phase of mEPSCs was best fitted with a single exponential function. The decay time constant was similar during control (τ= 1.77 ms) and pCPT–cGMP application (τ= 1.74 ms). E, F, summary data showing the effect of 50 μm pCPT–cGMP on the frequency (E) and amplitude (F) of mEPSCs in eight labelled PVN neurones. Data presented as means ±s.e.m. (Wilcoxon Signed rank test).

Figure 8. ODQ blocks the effect of SNAP on mIPSCs

A, representative tracings showing the spontaneous mIPSCs during control, perfusion of 100 μm SNAP and SNAP plus 30 μm ODQ in labelled PVN neurones. Note that treatment with ODQ abolished SNAP-induced increase in the frequency of mIPSCs. B,C, cumulative plot analysis of mIPSCs of the same neurone showing the distribution of the interevent interval (B) and peak amplitude (C) during control, application of SNAP, and application of SNAP + ODQ. D,E, summary data showing the effect of SNAP and SNAP + ODQ on the frequency (D) and amplitude (E) of mIPSCs in 11 labelled PVN neurones. Data presented as means ±s.e.m.*P < 0.05 compared to control (Kruskal-Wallis test).

Figure 9. Rp-pCPT-cGMP inhibits the effect of SNAP on mIPSCs

A, histograms showing the effect of 100 μm SNAP and SNAP + 1 μm Rp-pCPT-cGMP on the frequency (upper panel) and amplitude (lower panel) of spontaneous mIPSCs in a labelled PVN neurone. B, representative tracings showing the spontaneous mIPSCs during control, application of SNAP, and SNAP + Rp-pCPT-cGMP in the same neurone as in A. Note that Rp-pCPT-cGMP completely abolished the SNAP-induced increase in the frequency of mIPSCs. C and D, summary data showing the effect of SNAP and SNAP + Rp-pCPT-cGMP on the frequency (C) and amplitude (D) of mIPSCs of nine labelled PVN neurones. Data presented as means ±s.e.m.*P < 0.05 compared to control (Kruskal-Wallis test).