Oscillatory activity within rat substantia gelatinosa in vitro: a role for chemical and electrical neurotransmission

Abstract

Although rhythmic behaviour of mammalian spinal ventral horn networks has been extensively studied little is known about oscillogenesis in the spinal dorsal horn. The aims of this in vitro study were to record and determine the underlying mechanisms of potassium-evoked network field oscillations in the substantia gelatinosa of the neonatal rat dorsal horn, a lamina involved in nociceptive processing. Transient pressure ejection of a potassium solution evoked reproducible rhythmic activity in discrete areas of the substantia gelatinosa which lasted for 5-15 s with a single prominent peak in the 4-12 Hz frequency band (7.7 +/- 0.1 Hz, n = 60). Oscillations of similar frequency and amplitude were also observed in isolated dorsal horn quadrants. Application of CNQX (10 microm) reduced peak power amplitude and integrated power area (from 4 to 12 Hz) of the power spectrum, whereas D-AP5 (50 microm) had no effect on the potassium-evoked rhythm. Bicuculline (30 microm) or strychnine (10 microm) reduced the power amplitude and area. On combination of bicuculline (30 microm) and strychnine (10 microm) the reductions in power amplitude and area were not significantly different (P > 0.05) when compared with application of either drug alone. The gap junction blockers carbenoxolone (100 microm) or octanol (1 mM) significantly reduced power amplitude and area. Although TTX (1 microm) or a calcium-free perfusate both caused reductions in the power amplitude and area, potassium-evoked rhythmic activity persisted. However, this persistent rhythm was further reduced on combination of calcium-free perfusate with octanol (1 mM) and was abolished using a cocktail of drugs. Blockade of the potassium delayed rectifier current by tetraethylammonium (5 mM) or the hyperpolarization-activated current (I(h)) by ZD7288 (10 microm) disrupted the synchronization of the potassium-induced oscillation. The frequency of potassium-induced rhythms was unaffected by any of the drugs tested. These novel findings demonstrate that transient pressure ejection of potassium evokes oscillatory activity in the substantia gelatinosa in vitro. This rhythm is partly dependent upon various receptors (AMPA/kainate, GABA(A) and glycine), ion channels (potassium delayed rectifier and I(h)) and gap junctions. Oscillatory behaviour in the substantia gelatinosa could potentially play a role in the processing of nociceptive signals.

Figure 1. Potassium-induced rhythmic activity in the substantia gelatinosa of the rat spinal cord slice

A, extracellular field recording of rhythmic activity after brief pressure ejection of 1.5 m KCH₃SO₄ (arrow). Evoked oscillatory activity was maintained for approximately 10 s. Bi, expanded records of records presented in A (black bars). Bii, power spectral analysis of records in Bi reveals a time-dependent decay of the amplitude of the power. Biii, autocorrelation analysis of 1 s epochs (black bars in A) was used to determine the frequency of the oscillation. Note that the frequency was unchanged despite a substantive decrease in power (Bi and Bii). C, pooled data (n = 6) showing that the peak frequency was unchanged across the duration of the evoked activity (left panel). In contrast, there was an initial increase in power that gradually decayed with time (right panel).

Figure 2. Potassium-evoked oscillations in the rat substantia gelatinosa in vitro are partly dependent upon AMPA/kainate receptors but not on NMDA receptors

A, example of a single experiment where bath application of 50 μmd-AP5 for 30 min (middle panel) was followed by 10 μm CNQX for 30 min (right panel). B, CNQX reduced the peak amplitude and area of the power spectrum (B, right panel), without affecting frequency (C, right panel). In contrast, d-AP5 had no effect on either parameter (middle panels of B and C). D, quantified data (n = 6) reveal that CNQX but not d-AP5 significantly reduces the peak power and power area. Neither antagonist had an effect on the frequency of the oscillation. *P < 0.05 paired t test versus control.

Figure 3. Involvement of GABA_A receptor-mediated inhibition in rhythmic activity within rat substantia gelatinosa in vitro

A, potassium-induced rhythmic activity before (left panel) and after 30 min 30 μm bicuculline (middle panel). B, bicuculline reduced the power amplitude and area of the spectrum (middle panel) and on drug washout there was a partial reversal of this effect (right panel). C, bicuculline had no effect on the frequency of the oscillation. D, quantified data (n = 6) reveal that bicuculline reduces the peak power amplitude and the power area with no significant effect on the frequency of activity. *P < 0.05 paired t test versus control.

Figure 4. Contribution of glycine receptor-mediated inhibition to potassium-evoked rhythmic activity in rat substantia gelatinosa in vitro

A, potassium-induced rhythm before (left panel) and after 30 min 10 μm strychnine (middle panel). B, strychnine reversibly reduced the power of the spectrum. C, there was no effect of strychnine on the frequency of the oscillation. D, quantified data (n = 6) reveal that strychnine significantly reduces the parameters of peak power amplitude and power area with no significant effect on the frequency of activity. *P < 0.05 paired t test versus control.

Figure 5. Potassium-evoked oscillations without action potential dependent chemical neurotransmission in the rat substantia gelatinosa in vitro and the involvement of gap junction coupling

A, example traces from the same experiment showing potassium-induced oscillations before (left panel) and after bath application of calcium-free perfusate either alone (middle panel) or in subsequent combination for 45 min with the gap junction blocker 1 mm octanol (right panel). B, calcium-free perfusate caused a reduction in the peak amplitude of the power spectrum (middle panel). There was a further reduction in peak amplitude on subsequent addition of octanol to the calcium-free perfusate (right panel). C, autocorrelograms showing that frequency of the oscillation was unchanged. D, quantified data (n = 6) showing that the peak power and area power are significantly reduced (P < 0.05) by the calcium-free perfusate (left panel). A combination of calcium-free perfusate and octanol caused a further significant reduction in the peak power and area (left panel). Neither the calcium-free perfusate alone or in combination with octanol had any significant effect (P > 0.05) on the frequency of the oscillation (right panel). *P < 0.05 paired t test versus calcium-free perfusate.

Figure 6. Partial attenuation of rhythmic activity in the rat substantia gelatinosa in vitro by the gap junction uncoupling agents carbenoxolone and octanol

A, field recording of potassium-induced rhythmic activity before (left panel) and after 45 min 100 μm carbenoxolone (middle panel). B, carbenoxolone reversibly reduced the power of the spectrum (middle panel), without affecting frequency (C). D, quantified data (n = 6) reveal that carbenoxolone reduces the peak power and area with no significant effect on the frequency of activity. A similar profile was obtained with octanol (45 min, 1 mm), another gap junction uncoupler (n = 6). *P < 0.05 paired t test versus control.

Figure 7. Block of the delayed rectifier potassium current by TEA disrupts synchronization of the potassium-evoked rhythmic activity in rat substantia gelatinosa in vitro

A, potassium-induced rhythmicity before (left panel) and after 45 min 5 mm TEA (middle panel). TEA also caused spontaneous activity in the baseline prior to potassium ejection (not shown). Such spontaneous activity was also observed during the potassium-evoked rhythm (middle panel) and consequently the power amplitude for the potassium-induced rhythm could not be determined from the power spectrum. B, TEA causes a disruption in the synchronization of the rhythm.

Figure 8. Block of the I_h current by ZD7288 reduces potassium-evoked rhythmic activity and synchronization in rat substantia gelatinosa in vitro

A, potassium-induced rhythmicity before (left panel) and after 45 min 10 μm ZD7288 (middle panel). B, ZD7288 reversibly reduced the power of the spectrum. C, there was no effect of ZD7288 on the frequency of the oscillation although there was a disruption in the synchrony of the rhythm. D, quantified data (n = 6) reveal that ZD7288 significantly reduces the parameters of peak power amplitude and power area with no significant effect on the frequency of activity. *P < 0.05 paired t test versus control.