Olfactory rule learning-induced enhancement in intrinsic neuronal excitability is maintained by shutdown of the cholinergic M-current

Abstract

Training rats in a particularly difficult olfactory discrimination task initiates a period of accelerated learning, manifested as a dramatic increase in the rats' capacity to discriminate between pairs of odors once they have learned the discrimination task, implying that rule learning has taken place. At the cellular biophysical level, rule learning is maintained by reduction in the conductance of the slow current (sI_AHP) simultaneously in most piriform cortex layer II pyramidal neurons. Such sI_AHP reduction is expressed in attenuation of the post-burst afterhyperpolarization (AHP) potential and thus in enhanced repetitive action potential firing. Previous studies have shown that a causal relationship exists between long-lasting post-burst AHP reduction and rule learning. A specific channel through which the sI_AHP flows has not been identified. The sI_AHP in pyramidal cells is critically dependent on membrane phosphatidylinositol 4,5-bisphosphate [PtdIns(4,5)P(2)]. PtdIns(4,5)P(2) regulates the calcium sensitivity of the sI_AHP by acting downstream from the rise in intracellular calcium. These findings led to the interesting hypothesis that PtdIns(4,5)P(2) activates a variety of potassium channels. Thus, the sI_AHP would not represent a unitary ionic current but the embodiment of a generalized potassium channel gating mechanism. We thus hypothesized that the learning-induced increase in intrinsic excitability is mediated by reduced conductance of one or more of the currents that contribute to the sI_AHP. Here we first show, using current-clamp recordings, that the post-burst AHP in piriform cortex pyramidal neurons is also mediated by the I_h, and the contribution of this current to the post-burst AHP is also affected by learning. We also show, using whole-cell patch-clamp recordings, that the sI_AHP in neurons from trained rats is not sensitive to blocking membrane phosphatidylinositol 4,5-bisphosphate [PtdIns(4,5)P(2)], and to the blocking of the current mediated by the cholinergic muscarinic acetylcholine receptor (M-current). Further current-clamp recordings also show that blocking PtdIns(4,5)P(2) synthesis and application of a specific IKCa blocker have no effect on the post-burst AHP in neurons from trained as well as control rats. Taken together with results from our previous studies, these data suggest that rule learning-induced long-lasting enhancement in intrinsic neuronal excitability results from reduced conductance of the M-current and thus the slow potassium currents, which control repetitive spike firing.

Keywords: cholinergic modulation; intrinsic neuronal excitability; olfactory rule learning; piriform cortex; pyramidal neurons; slow afterhyperpolarization.

Figure 1

Complex olfactory learning apparatus and complex OD rule learning. (A) Similar protocols were applied for “trained” and “control” rats. An electronic “start” command randomly opens two of eight valves (V), releasing a positive-cue odor (P) into one of the arms and a negative-cue odor (N) into another. After 8 s, the two corresponding guillotine doors (D) are lifted to allow the rat to enter the selected arms. Upon reaching the far end of an arm (90 cm long), the rat body interrupts an infrared beam (I, arrow) and a drop of drinking water is released from a water hose (W) into small drinking well (for a “trained” rat—only if the arm contains the positive-cue odor, for “pseudo-trained” rat—randomly). A trial ends when the rat interrupts a beam, or in 10 s, if no beam is interrupted. A fan is operated for 15 s between trials, to remove odors. (B) Trained rats (n = 20) demonstrate acquisition of rule learning. With 20 trials per day, 8 consecutive days of training was required for this group to reach the criterion for discriminating between the first pair of odors (80% correct choices). Other groups usually require a similar period. Values represent mean ± SE.

Figure 2

A widespread learning-induced decrease in the post-burst AHP amplitude mediates enhanced intrinsic neuronal excitability. (A) Examples for current-clamp recordings for AHPs and repetitive spike firing in neurons taken from naive (orange) and trained rat (navy) measurements in a piriform cortex neuron. For AHP recordings, the holding membrane potential is −60 mV, and the AHP is generated by a 100-ms depolarizing current step, with the intensity that generates a train of six action potentials. The presented trace is an average of five consecutive recordings made within intervals of 10 s. Repetitive spike firing is shown in the same neurons in response to intracellularly applied 1-s depolarizing, with a stimulus intensity of I_thx2. (B) The graph shows the difference between groups in the number of spikes. Dots denote the spikes generated in each neuron. Values represent mean ± SE. *p < 0.05. Between-group comparison was done using one-way ANOVA for the three learning groups, with post-hoc multiple t-tests for each pair of groups. Post-hoc tests show that values of the train group differed significantly from values of the naive (P < 0.05) and the pseudo-trained group (P < 0.05). The two control groups did not differ between them (p = 0.95). (C) The difference between groups in the AHP amplitude. Dots denote the AHP value in each neuron. Values represent mean ± SE. ***p < 0.001. Between-group comparison was carried out using one-way ANOVA for the three learning groups, with post-hoc multiple t-tests for each pair of groups. Post-hoc tests show that values of the train group differed significantly from values of the naive (P < 0.001) and the pseudo-trained group (P < 0.001). The two control groups did not differ between them (p = 0.21). (D) The relation between the AHP amplitude and the number of action potentials generated by each neuron. Results are shown for 48 neurons. A highly significant negative correlation between the AHP amplitude and the number of action potentials is apparent (r = −0.32, p < 0.001).

Figure 3

I_h is active during the post-Burst AHP in neurons from control rats. (A) Averaged traces from pseudo-trained and trained neurons, before and 20 min after the ZD7288 application. The I_h blocker reduced the post-burst AHP in the pseudo-trained neuron only. (B) The I_h blocker, ZD7288, reduces the post-burst AHP amplitude in neurons from the controls, but not from the trained rats. Bars represent mean ± SE. **p < 0.051. *p < 0.05. A paired t-test was used to compare the effect of ZD7288 of each group of neurons. An unpaired t-test was used to examine the difference between neurons from the trained and control rats. *p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant.

Figure 4

Rundown of the sI_AHP upon prolonged recordings shows that the M current is not active in piriform cortex pyramidal neurons after learning. (A) Whole-cell voltage-clamp recordings from a pyramidal neuron. The cell was voltage-clamped at −60 mV. A 200-ms depolarizing pulse to +60 mV generated an unclamped calcium current (truncated). After its termination, an outward current appears with a reversal potential of −80 mV. Traces illustrating the rundown of the sI_AHP. The blue trace shows the sI_AHP when the stable recording was established. Red and orange traces were taken after 15 and 30 min. (B) Summary plot illustrating the rundown of the sI_AHP in neurons from trained and control rats. Recordings were carried out in the presence of the SK channel blocker apamin, to ensure that only the sI_AHP is activated. (C) Rundown of the sI_AHP in the presence of apamin and wortmannin, to inhibit PtdIns(4,5)P(2) biosynthesis. (D) Rundown of the sI_AHP in the presence of apamin and XE991, to block the cholinergic M current. (B,D) values represent mean ± SE, and n denotes the numbers of neurons. (E) Bars show the average normalized decline for each treatment. In apamin only, there was no difference in the rundown kinetics between the controls and trained neurons. The presence of wortmannin enhanced the sI_AHP rundown in neurons from controls only. Application of the M current blocker greatly enhanced the sI_AHP decay in neurons from controls only. Notably, both treatments had no effect on the sI_AHP decay in neurons from trained rats. *p < 0.05. ***p < 0.001. The paired t-test was used to compare the effect of drug application on each group of neurons. The unpaired t-test was used to examine the difference between neurons from trained and control rats.

Figure 5

Blocking PtdIns(4,5)P(2) biosynthesis does not affect the post-burst AHP amplitude. (A) Averaged traces from pseudo-trained and trained neurons, before and 20 min after wortmannin application. Wortmannin had no effect on the post-burst AHP in both neurons. (B) Timeline of continuous AHP recordings before (−10 to 0 min) and after wortmannin application in neurons from control and trained rats. Values represent mean ± SE. (C) Direct comparison of the AHP values in each cell before and 20 min after wortmannin application in neurons from control and trained rats. Wortmannin did not change the post-burst AHP amplitude in controls or trained neurons. Values represent mean ± SE. The paired t-test was used to compare the effect of wortmannin on each group of neurons. ns, not significant.

Figure 6

Blocking the IKCa channels does not affect the post-burst AHP amplitude. (A) Averaged traces from a naive neuron before and 30 min after TRAM-34 application. The IKCa blocker had no effect on the post-burst AHP. (B) Timeline of continuous AHP recordings before (−10 to 0 min) and after TRAM-34 application in neurons from control and trained rats. Values represent mean ± SE. Inset: Direct comparison of the AHP values in each cell before and 30 min after TRAM-34 application. Values represent mean ± SE. (C) The relation between the initial post-burst AHP amplitude and the effect of TRAM-34 in each neuron. TRAM-34 did not affect the AHP amplitude, regardless of its initial value (r = −0.13, p = 0.87). The paired t-test was used to compare the effect of TRAM.

Figure 7

Difference in the post-burst AHP amplitude between neurons from control and trained rats is abolished in the presence of a cholinergic agonist. Application of the cholinergic agonist, carbachol, reduced significantly the post-burst AHP amplitude in neurons from control rats (P < 0.001) but did not affect the post-burst AHP in neurons from trained rats. Moreover, in the presence of the agonist, the difference in the post-burst AHP amplitude between the two groups (p < 0.01) was abolished. Notably, intracellular application of the calcium chelator BAPTA via the recording electrode (Saar et al., 2001) further reduced the post-burst AHP in neurons from control and trained rats. Bars represent mean ± SE. Each point reorients the post-burst AHP value in a cell. All data are taken from our previous publication (Saar et al., 2001). The unpaired t-test was used to examine the difference between neurons from trained and control rats and between neurons from the same groups at different conditions. *p < 0.05; **p < 0.01; ***p < 0.001; n.s., not significant.