Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Feb 5;11(2):e0148800.
doi: 10.1371/journal.pone.0148800. eCollection 2016.

Presynaptic GABAB Receptors Regulate Hippocampal Synapses During Associative Learning in Behaving Mice

Free PMC article

Presynaptic GABAB Receptors Regulate Hippocampal Synapses During Associative Learning in Behaving Mice

M Teresa Jurado-Parras et al. PLoS One. .
Free PMC article


GABAB receptors are the G-protein-coupled receptors for GABA, the main inhibitory neurotransmitter in the central nervous system. Pharmacological activation of GABAB receptors regulates neurotransmission and neuronal excitability at pre- and postsynaptic sites. Electrophysiological activation of GABAB receptors in brain slices generally requires strong stimulus intensities. This raises the question as to whether behavioral stimuli are strong enough to activate GABAB receptors. Here we show that GABAB1a-/- mice, which constitutively lack presynaptic GABAB receptors at glutamatergic synapses, are impaired in their ability to acquire an operant learning task. In vivo recordings during the operant conditioning reveal a deficit in learning-dependent increases in synaptic strength at CA3-CA1 synapses. Moreover, GABAB1a-/- mice fail to synchronize neuronal activity in the CA1 area during the acquisition process. Our results support that activation of presynaptic hippocampal GABAB receptors is important for acquisition of a learning task and for learning-associated synaptic changes and network dynamics.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.


Fig 1
Fig 1. GABAB1a-/- mice exhibit deficits in executing an operant conditioning task.
(A) Mice were chronically implanted with stimulating (St.) and recording (Rec.) electrodes in the Schaffer collaterals/commissural pathway (Sch.) and the CA1 area, respectively. Representative photomicrographs illustrating the location (arrows) of recording (left) and stimulating (right) electrodes in post-experimental tissue are shown at the top. Calibration bars: 500 μm. (B) Mice were trained in a Skinner box to press a lever to obtain a food pellet. For operant conditioning we used two paradigms of increasing difficulty. In the first paradigm (Fixed-ratio of 1:1) mice had to press the lever ≥ 20 times per 20 min session for two successive sessions to successfully complete the task (criterion). In the second paradigm (Light/Dark), lever presses were rewarded only when a light bulb was switched on. Lever presses while the bulb was off were punished with a time penalty of 10 s during which the bulb would not turn on. In this case the animal had to press the lever at least the same number of times during the light and the dark periods for two successive sessions to successfully complete the task (criterion). (C) Performance of mice during 6 days of training with the 1:1 ratio schedule. WT mice (open circles) pressed the lever significantly more (P ≤ 0.02) and reached the criterion (arrow) in fewer sessions than GABAB1a-/- mice (closed circles). All WT mice but only 40% of the GABAB1a-/- mice successfully completed the task within 6 sessions (right panel). (D) Analysis of evoked fEPSP at the CA3-CA1 synapses during operant conditioning. The fEPSP slopes of WT mice were significantly (P < 0.02) larger than those of GABAB1a-/- mice during the 2nd session. Representative fEPSPs recorded during the indicated sessions (1 and 2) are shown at the top. E, Performance of mice during 6 days of training in the light/dark test. WT mice performed significantly better than GABAB1a-/- mice (*, P = 0.03; **, P ≤ 0.006) and reached the criterion by the 4th session (arrow). GABAB1a-/- mice failed to reach the criterion. The light/dark coefficient was calculated as follows: (number of lever presses during the light period—number of lever presses during the dark period) / total number of lever presses.
Fig 2
Fig 2. LFP changes in the CA1 area during learning sessions using 1:1 ratio schedule.
(A-F) Mean peak values (circles) and best fits (lines) of the spectral power corresponding to LPFs recorded during each learning session in WT (open circles and continuous lines) and GABAB1a-/- (closed circles and dotted lines) mice for all spectral bands in A (1–100 Hz), and for selected spectral bands in B-F (1–3 Hz in B, 3–8 Hz in C, 8–12 Hz in D, 12–30 Hz in E and 30–100 Hz in F). Note that peak spectral power values across the 3–8 Hz to the 30–100 Hz bands (C-F) tend to increase with each training session in WT mice, while spectral power values decreased or remained unchanged in GABAB1a-/- mice. Changes in peak spectral power across the learning sessions reflect physiological changes during the learning sessions. The linear or non-linear equations that best fitted (P ≤ 0.05) spectral power values as a function of session number are shown above the corresponding graphs.
Fig 3
Fig 3. GABAB1a-/- mice exhibit increased LTP of evoked fEPSPs in the CA1 area while input/output curves and paired-pulse facilitation are normal.
(A) Input/output curves for the CA3-CA1 synapse. A single (100 μs, biphasic) pulse was presented to Schaffer collaterals at increasing intensities (from 0.02 mA to 0.4 mA, in steps of 0.02 mA) while recording evoked fEPSPs in the CA1 area for WT (open circles) and GABAB1a-/- (closed circles) mice. Representative fEPSPs recorded from the stratum radiatum evoked with the intensities indicated in the graph (1, 2, 3) are shown at the top for each genotype. Equations corresponding to the best (r ≥ 0.996; P < 0.0001) sigmoid fits of the data [mean ± SEM; n ≥ 8 mice and ≥ 40 measurements for each of the 20 different stimulus intensities applied] are indicated. (B) Paired-pulse facilitation at the CA3-CA1 synapse. The graph shows the slopes of the second fEPSPs expressed as a percentage of the first (mean ± SEM) for six inter-stimulus intervals (10, 20, 40, 100, 200, and 500 ms). WT and GABAB1a-/- mice exhibited paired-pulse facilitation at intervals of 10–200 ms (P < 0.01) that was not significantly different between the genotypes (P = 0.342). Representative recordings at 20 ms (top) and 200 ms (bottom) of inter-stimulus interval are shown on top (open circles for WT and closed circles for GABAB1a-/-). (C) Time course of LTP in the CA1 area (fEPSP mean ± SEM) following HFS. The HFS was presented after 15 min of baseline recording, at the time marked by the dashed line. The fEPSP is given as a percentage of the baseline (100%) slope. WT and GABAB1a-/- mice showed a significant increase (ANOVA, two-tailed) in fEPSP slope following HFS when compared to baseline at day 1 (P < 0.001). fEPSP slope values were significantly (*, P ≤ 0.03) larger for GABAB1a-/- than WT mice during the 5 days of recording. fEPSPs collected from WT (open circles) and GABAB1a-/- (closed circles) mice before (baseline, B) and after (1, 2) HFS of Schaffer collaterals are shown on top.
Fig 4
Fig 4. GABAB1a-/- mice exhibit a drastic increase in the spectral power of LFP activity in the presence of kainate.
(A) Scheme of the experimental protocol used. LFP activity in the CA1 area was recorded for 30 min to establish a baseline. Then, a HFS protocol (consisting of five 200 Hz, 100 ms trains of pulses at a rate of 1/s) was presented and the evoked LFP activity was recorded for another 30 min. After that, mice were injected with kainate (8 mg/kg), and 60 min later LFP activity was recorded again. Calibration for the selected LFP traces is indicated on the right. (B-G) Histograms representing the maximum values of the spectral power of LFP activity recorded during baseline (BL) recordings, and before and after kainate injection into WT (green histograms) and GABAB1a-/- (red histograms) mice. Values of all spectral bands (1–100 Hz) are depicted in B, selected spectral bands in C-G (1–3 Hz in C, 3–8 Hz in D, 8–12 Hz in E, 12–30 Hz in F, and 30–100 Hz in G). Note that for all spectral bands the maximum values of the spectral power were seen in GABAB1a-/- mice in the presence of kainate and the second HFS.*, P < 0.05; **, P < 0.01; ***, P < 0.001.
Fig 5
Fig 5. Differences in the PSD of LFPs recorded in the hippocampal CA1 area during different states of neuronal activation.
(A) Scheme of the experimental protocol used. (B-G) PSD plots for the LFP activity recorded during baseline (WT, in B; GABAB1a-/-, in E), and before (WT group, in C; GABAB1a-/- group, in F) and after (WT group, in D; GABAB1a-/- group, in G) kainate injection (8 mg/kg). Colored arrows in B-G indicate the dominant frequency of the spectra [PSD values, in log (mV/√Hz) are indicated on the y-axis; corresponding frequencies, in log (Hz), are indicated on the x-axis]. The mean values of the represented traces are indicated by the colored traces. (H) Histograms representing PSDs corresponding to each group and the three different recording situations [baseline (BL, green and red histograms), and before (light green and light red histograms) and after (dark green and dark red histograms) kainate administration]. (I) Histogram representing the dominant frequency of the LFP activity recorded during baseline (BL, green and red histograms), and before (light green and light red histograms) and after (dark green and dark red histograms) kainate injection for both genotypes. Note in H and I that the maximum PSD value and the minimum value of the fundamental frequency was recorded in the GABAB1a-/- group following HFS after injection of kainate. For both H and I, collected data was quantified for each experimental animal (n = 5 per group) and then averaged. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Similar articles

See all similar articles

Cited by 7 articles

See all "Cited by" articles


    1. Chalifoux JR, Carter AG (2011) GABAB receptor modulation of synaptic function. Curr Opin Neurobiol 21: 339–344. 10.1016/j.conb.2011.02.004 - DOI - PMC - PubMed
    1. Gassmann M, Bettler B (2012) Regulation of neuronal GABAB receptor functions by subunit composition. Nat Rev Neurosci 13: 380–394. 10.1038/nrn3249 - DOI - PubMed
    1. Kulik A, Vida I, Luján R, Haas CA, López-Bendito G, et al. (2003) Subcellular localization of metabotropic GABAB receptor subunits GABAB1a/b and GABAB2 in the rat hippocampus. J Neurosci 23: 11026–11035. - PMC - PubMed
    1. Davies CH, Collingridge GL (1996) Regulation of EPSPs by the synaptic activation of GABAB autoreceptors in rat hippocampus. J Physiol (Lond) 496: 451–470. - PMC - PubMed
    1. Scanziani M (2000) GABA spillover activates postsynaptic GABAB receptors to control rhythmic hippocampal activity. Neuron 25: 673–681. - PubMed

Publication types

Grant support

This study was supported by grants from the Spanish Ministry of Economy and Competitiveness (BFU2014-56692-R) to AG and JMD-G, and from the Swiss National Science Foundation (3100A0-117816) to BB, and a grant from the European Community’s Seventh Framework Program (FP7/2007-2013) under grant agreement N° 201714 (DEVANX) to AG, JMD-G, and BB.