Sensitivity to theta-burst timing permits LTP in dorsal striatal adult brain slice

Abstract

Long-term potentiation (LTP) of excitatory afferents to the dorsal striatum likely occurs with learning to encode new skills and habits, yet corticostriatal LTP is challenging to evoke reliably in brain slice under physiological conditions. Here we test the hypothesis that stimulating striatal afferents with theta-burst timing, similar to recently reported in vivo temporal patterns corresponding to learning, evokes LTP. Recording from adult mouse brain slice extracellularly in 1 mM Mg(2+), we find LTP in dorsomedial and dorsolateral striatum is preferentially evoked by certain theta-burst patterns. In particular, we demonstrate that greater LTP is produced using moderate intraburst and high theta-range frequencies, and that pauses separating bursts of stimuli are critical for LTP induction. By altering temporal pattern alone, we illustrate the importance of burst-patterning for LTP induction and demonstrate that corticostriatal long-term depression is evoked in the same preparation. In accord with prior studies, LTP is greatest in dorsomedial striatum and relies on N-methyl-d-aspartate receptors. We also demonstrate a requirement for both Gq- and Gs/olf-coupled pathways, as well as several kinases associated with memory storage: PKC, PKA, and ERK. Our data build on previous reports of activity-directed plasticity by identifying effective values for distinct temporal parameters in variants of theta-burst LTP induction paradigms. We conclude that those variants which best match reports of striatal activity during learning behavior are most successful in evoking dorsal striatal LTP in adult brain slice without altering artificial cerebrospinal fluid. Future application of this approach will enable diverse investigations of plasticity serving striatal-based learning.

Keywords: LTP; learning; plasticity; striatum; theta.

Fig. 1.

Long-term potentiation (LTP) depends on intraburst and theta-burst timing. Example traces from end of experiment (red) overlay traces from baseline (gray) in the insets. Error bars represent ± SEM. A: schematic of induction variants. For each induction paradigm employed in this paper, a single train of stimuli is illustrated in brackets and annotated to show stimuli number is matched across conditions. Theta-burst (*intraburst period, **theta period) and nonbursty trains (50 Hz) are delivered with a 15-s intertrain interval. High-frequency stimulation (HFS) (100 Hz) and 20-Hz trains are delivered with a 10-s intertrain interval. B: intraburst frequency of 50 Hz is more effective than 100 Hz, both dorsomedial (DM) and dorsolateral (DL). Theta-burst frequency is 10.5 Hz for all groups. C: burst timing is critical to LTP. Theta-burst frequency of 10.5 Hz produces stronger, longer-lasting potentiation than 5 or 8 Hz in DM striatum. Bar graph indicates difference from nonstimulated controls at significance of ^#P < 0.0001 or ^&P < 0.05. In the nonbursty condition, the 40 stimuli within each train are delivered at a constant 50 Hz, and neither LTP nor long-term depression (LTD) results. Nonbursty experiments ended at 60 min, since long-term plasticity was not induced. D: picrotoxin decreases but does not eliminate induction of LTP using the optimal theta-burst timing of 50-Hz intraburst and 10.5-Hz interburst. TBS, theta-burst stimulation.

Fig. 2.

LTD confirms bidirectional plasticity in adult dorsal striatal slice. Four trains of moderate-frequency stimulation (20 Hz), but not HFS (100 Hz), evokes LTD, both DM and DL. Example traces from end of experiment (red) overlay baseline traces (gray). Error bars represent ± SEM.

Fig. 3.

TBS LTP requires N-methyl-D-aspartate (NMDA), group I metabotropic glutamate receptors (mGluR), m1 metabotropic acetylcholine receptors (AChR), and dopamine D1-type receptors. The drug-free TBS group (DM, 50 Hz intraburst, 10.5 Hz theta) was collected interleaved with pharmacology experiments and thus is different than the analogous TBS group in Fig 2. Example traces from end of experiment (red) overlay baseline traces (gray). Error bars represent ± SEM. A: NMDA receptor antagonist 2-amino-5-phosphonovaleric acid (APV) blocks LTP. B: group I mGluR antagonist (RS)-1-aminoindan-1,5-dicarboxylic acid (AIDA) blocks LTP. C: m1 AChR antagonist telenzepine blocks LTP. D: D1-type dopamine receptor antagonist SCH23390 blocks LTP.

Fig. 4.

TBS LTP requires protein kinase C (PKC), protein kinase A (PKA), and extracellular signal-regulated kinase (ERK). The drug-free TBS group (DM, 50 Hz intraburst, 10.5 Hz theta) was collected interleaved with pharmacology experiments and thus is different than the analogous TBS group in Fig 2. Example traces from end of experiment (red) overlay baseline traces (gray). Error bars represent ± SEM. A: PKC inhibitor chelerythrine (CHE) blocks LTP. Reduced drug concentration reduces amplitude without completely blocking LTP (green). B: PKA inhibitor PKI blocks LTP. Reduced drug concentration reduces amplitude without completely blocking LTP (blue). C: reduced concentrations of PKC and PKA inhibitors fully block LTP when combined. Mean effect from independent reduced concentration inhibitors are overlaid. D: preventing ERK activation with MAPK/ERK kinase inhibitor U-0126 blocks LTP.