D2 dopamine receptors recruit a GABA component for their attenuation of excitatory synaptic transmission in the adult rat prefrontal cortex

Abstract

The dopamine modulation of neuronal excitability in the prefrontal cortex (PFC) changes during critical late periods of postnatal development. In particular, D2 receptors activate fast-spiking interneurons after, and not before, adolescence. To test the functional impact of this change, we investigated the effects of dopamine agonists on PFC excitatory synaptic transmission with whole-cell recordings from deep-layer pyramidal neurons in brain slices obtained from prepubertal [postnatal day (PD) 28-35] and postpubertal (PD>51) rats. Electrical stimulation of superficial layers elicited a fast AMPA/kainate excitatory postsynaptic potential (EPSP). In the adult PFC, the D2 agonist quinpirole decreased EPSP amplitude, an effect that lasted for at least 25 min after drug washout and was blocked by the D2 antagonist eticlopride. The late component of this effect was blocked by the GABA-A antagonist picrotoxin without affecting the early inhibition. Quinpirole also decreased EPSP amplitude in deep-layer pyramidal neurons from prepubertal rats, but this response was not affected by picrotoxin. A D1 agonist, on the other hand, did not affect the pyramidal neuron EPSP. These results indicate that D2, not D1, receptors attenuate local excitatory synaptic transmission in the adult PFC, and this effect of D2 involves a recruitment of local GABAergic activity.

Fig. 1

Whole-cell recordings of deep-layer PFC pyramidal neurons obtained from slices from a young adult (PD 56) rat. (A) Characteristic voltage responses (top) to depolarizing and hyperpolarizing somatic current pulses (bottom; 300 ms duration, −300 to +100 pA in amplitude) in a representative neuron. (B) Current-voltage (IV) plot obtained from the traces shown in A. Typically, currents larger than −100 pA yielded inward rectification in the hyperpolarizing direction (arrowhead). The oblique line highlights the regression slope for the linear part of the plot (−100 to +100 pA). (C) IR-DIC image of a deep-layer pyramidal neuron recorded from a PFC slice. Arrowheads indicate the shadow of the patch electrode. (D) Neurobiotin labeling of a representative pyramidal neuron recorded from the medial PFC (same cell from which traces shown in A were obtained). Small arrowheads point to the apical dendrite and the large arrowhead indicates the cell body.

Fig. 2

Electrical stimulation of superficial layers typically elicits a glutamatergic EPSP in deep-layer pyramidal neurons of the medial PFC. (A) Diagram illustrating the spatial arrangement of stimulating electrodes (layers I–II) and recording sites (layers V–VI). (B) Bath application of the GABA-A antagonist picrotoxin (10 μM) failed to change the amplitude of the evoked responses in all cells tested (n = 6). Left: bar graph summarizing EPSP amplitudes; right: example of the evoked response recorded before (black line) and after 5 min. of picrotoxin (gray line). Traces in this and subsequent figures are representative examples and not averages. (C) Bar graph (left) summarizing the effect of bath application of the AMPA/kainate antagonist CNQX (10 μM). The amplitude of the evoked response was gradually reduced and completely eliminated after 5 min. of CNQX in all cells tested (n = 8, ***P < 0.0001, paired t-test). Representative traces (right) showing the evoked EPSP before (baseline) and after 5 min. of CNQX. (D) Bath application of the NMDA antagonist APV (50 μM) failed to change EPSP amplitude in all cells tested (n = 7; left). Traces (right) recorded before (baseline) and after 5 min. of drug application illustrate the slight reduction of EPSP decay observed with APV.

Fig. 3

Quinpirole depresses deep-layer pyramidal neuron EPSP amplitude in the PFC of postpubertal animals. (A) Graph summarizing the effect of the D₂ agonist quinpirole on pyramidal neuron EPSP amplitude. Bath application of quinpirole (1 μM for 5 min.) significantly reduced EPSP amplitude in all cells tested (n = 10, ***P < 0.0001, paired t-test). (B) Graph illustrating the effect of quinpirole in presence of the D₂ antagonist eticlopride (20 μM) in all cells tested. (C) Bar graph summarizing the effect of quinpirole and quinpirole + eticlopride as percentage changes relative to baseline. After 5 min. of quinpirole (1 μM), the average EPSP amplitude decreased by around 23%, an effect that was not evident in presence of eticlopride (***P < 0.0001, unpaired t-test). (D) Representative traces of evoked EPSP recorded in pyramidal neurons before and after bath application of 1 μM quinpirole alone (top) or in presence of 20 μM eticlopride (bottom).

Fig. 4

Time course of the effect of quinpirole on PFC pyramidal neuron EPSP amplitude, recorded in slices from postpubertal animals. Quinpirole (1 μM) significantly attenuated EPSP amplitude in pyramidal neurons after 5–7 min. of drug application (indicated with a gray shading), and this was blocked by 20 μM eticlopride (open triangles). In the absence of eticlopride, however, the EPSP attenuation remained even after quinpirole was removed from the bath (solid squares, n = 10). A period of at least 25 min. was required to partially washout the effect of quinpirole. The GABA-A antagonist picrotoxin (10 μM; open squares, n = 6) shortened the duration of this inhibition (*P < 0.01, **P < 0.001, ***P < 0.0001, Tukey posthoc test after significant 2-way ANOVA, interaction between drug and time P < 0.001). The initial D₂-dependent EPSP attenuation (white and black arrowheads) was not affected by picrotoxin (open circle, n = 6).

Fig. 5

In slices from prepubertal animals the GABA component was not observed. (A) Plot summarizing the effect of quinpirole on EPSP amplitudes in pyramidal neurons recorded in the PFC of prepubertal animals. Bath application of quinpirole (1 μM) significantly reduced EPSP amplitude in all cells tested (n = 5, P < 0.001, paired t-test). (B) Plot summarizing the effect of quinpirole on EPSP amplitude in presence of the GABA-A antagonist picrotoxin (10 μM). In these conditions, quinpirole still reduced pyramidal neuron EPSP amplitude to a similar degree to that observed with quinpirole alone (n = 4, P < 0.005, paired t-test). (C) Line graph showing the time course of the quinpirole effect on EPSP amplitude in pyramidal neurons from prepubertal and postpubertal PFC slices. Bath application of quinpirole decreased EPSP amplitude by near 14% (solid squares/solid line) in prepubertal pyramidal neurons; a more pronounced effect was obtained in the PFC of postpubertal animals (23%, grey squares/dashed line;*P < 0.01, **P < 0.001, ***P < 0.0001, Tukey posthoc test after significant 2-way ANOVA, interaction between drug and time P < 0.001). In addition, the early postquinpirole inhibition observed in slices from prepubertal animals (open circles/solid line) was not affected by picrotoxin as it was in the adult PFC (arrowheads).

Fig. 6

Bath application of the D₁ agonist SKF38393 failed to elicit significant changes on medial PFC pyramidal neuron EPSP amplitude. (A) Plot illustrating the effect of 8 μM SKF38393 on EPSP amplitude recorded in deep-layer pyramidal neurons. No consistent changes were observed after 5–7 min. of SKF38393 alone (n = 5; open circles) or in presence of picrotoxin (10 μM, n = 4; open triangles). (B) Time course analysis of normalized EPSP amplitude revealing that SKF38393 does not affect significantly evoked synaptic responses.