Mechanisms of decreased cholinergic arousal in focal seizures: In vivo whole-cell recordings from the pedunculopontine tegmental nucleus

Abstract

Focal limbic seizures often impair consciousness/awareness with major negative impact on quality of life. Recent work has shown that limbic seizures depress brainstem arousal systems, including reduced action potential firing in a key node: cholinergic neurons of the pedunculopontine tegmental nucleus (PPT). In vivo whole-cell recordings have not previously been achieved in PPT, but are used here with the goal of elucidating the mechanisms of reduced PPT cholinergic neuronal activity. An established model of focal limbic seizures was used in rats following brief hippocampal stimulation under light anesthesia. Whole-cell in vivo recordings were obtained from PPT neurons using custom-fabricated 9-10 mm tapered patch pipettes, and cholinergic neurons were identified histologically. Average membrane potential, input resistance, membrane potential fluctuations and variance were analyzed during seizures. A subset of PPT neurons exhibited reduced firing and hyperpolarization during seizures and stained positive for choline acetyltransferase. These PPT neurons showed a mean membrane potential hyperpolarization of -3.82 mV (±0.81 SEM, P < .05) during seizures, and also showed significantly increased input resistance, fewer excitatory post-synaptic potential (EPSP)-like events (P < .05), and reduced membrane potential variance (P < .01). The combination of increased input resistance, decreased EPSP-like events and decreased variance weigh against active ictal inhibition and support withdrawal of excitatory input as the dominant mechanism of decreased activity of cholinergic neurons in the PPT. Further identifying synaptic mechanisms of depressed arousal during seizures may lead to new treatments to improve ictal and postictal cognition.

Figure 1.. Reduced-firing hyperpolarizing (RfHp) neuron in the pedunculopontine tegmental nucleus (PPT) during a focal limbic seizure.

(A) Whole-cell current clamp recording of membrane potential (Vm) in PPT neuron shows reduced firing and hyperpolarization (see inset) during a focal seizure induced in the hippocampus by a 2s, 60 Hz stimulus (gray bar). Concomitant recording of local field potential (LFP) shows polyspike seizure activity in the hippocampus (Hc). Orbital frontal cortex (Ctx) LFP exhibit slow-waves and multiunit activity (MUA) recordings show Up/Down state firing at seizure onset lasting into the postictal period. (B) Histology demonstrates co-localization of biocytin (from intracellular electrode solution) staining and ChAT immunohistochemistry for neuron recorded in A. Scale bar is 20 microns.

Figure 2.. Reduced action potential firing in RfHp neurons during seizures.

(A) Raster plots (five cells from five animals) show decreased firing during the Ictal period compared with Baseline. Boxes in ictal panel indicate duration of seizures up to the first 20 s. (B) Histograms of firing rate in 1 s bins across neurons. Mean reduction in firing rate of Ictal compared to Baseline epochs was −51.6% (SEM 12.8%, * p < 0.05, paired t-test). Baseline is defined as the 20 s preceding seizure onset. Ictal is defined as up to the first 20 s following seizure onset. Postictal is the first 20 s following seizure end. Recovery is the 20 s following the Postictal period.

Figure 3.. Increased input resistance and reduced membrane potential fluctuations in RfHp neurons during seizures.

(A) Example of response to continuous 10Hz 30pA square current pulses in an RfHp neuron (top trace, Voltage; bottom trace, Current). Action potentials are truncated. The trace shows reduced firing and hyperpolarization in the ictal period (blue), as well as a larger magnitude of V_m response to constant current steps, indicating an increase in ictal input resistance. Inset on right shows expanded view of baseline (black) and ictal (blue) V_m responses to current pulses, with resting potentials aligned vertically to enable differences in response amplitude to be compared more easily. (B) Mean time course of absolute change in mean input resistance (R_in) during seizures reveal that the increase in mean R_in for RfHp neurons (blue) was significant during the ictal period (P<0.05, two-tailed t-test). Seizure onset was at time 0, and ends of seizure epochs are indicated by vertical dashed lines for RfHp (blue, n=3) and non-RfHp (red, n=4) neurons. Traces show mean change in R_in (R_in – baseline R_in) from baseline with standard error of mean (SEM). (C) Mean time course of change in mean Vm from baseline (Vm – baseline Vm) for the same neurons in (B) and over the same time course, showing significant hyperpolarization during the ictal period (P<0.05, two-tailed t-test). (D, E) Changes in input resistance (R_in) for individual RfHp and non-RfHp neurons. RfHp neurons (D) show consistent increases in R_in during the ictal period compared to baseline. (* P < 0.05, paired, two-tailed t-test, Holm-Bonferroni corrected). None of the non-RfHp neurons (E) show significant increase in R_in during seizures. For (D) and (E) mean and SEM are shown across each neuron that received current pulses, with SEMs obtained across current pulses for each period. (F) Examples of EPSP-like events defined as positive fluctuations in membrane potential of >0.4mV with a rise-time of <2ms (Koch, 2004; Mason et al., 1991). Top (black) trace shows the baseline V_m of an RfHp neuron, while the lower (blue) trace shows ictal V_m of the same RfHp neuron. Orange markings overly EPSP-like events, with each event also marked as a black (baseline) or blue (ictal) tick-mark for easier comparison of relative frequency. In group analysis, frequency of EPSP-like events was significantly decreased during the ictal periods of RfHp neurons compared to baseline (see text). (G,H) Average membrane potential histograms for RfHp neurons (n=4) (G) and non-RfHp neurons (n=10) (H) in the Baseline, Ictal and Recovery periods, presented side-by-side with bar graphs quantifying variance. Mean histogram for RfHp neurons was narrower for the ictal period (blue), and the corresponding membrane potential variance of RfHp neurons was significantly decreased during seizures compared to baseline (* P < 0.05, paired two-tailed t-test, Holm- Bonferroni corrected)