Adenosine A2A receptors control generalization of contextual fear in rats

Abstract

Fear learning is essential to survival, but traumatic events may lead to abnormal fear consolidation and overgeneralization, triggering fear responses in safe environments, as occurs in post-traumatic stress disorder (PTSD). Adenosine A_2A receptors (A_2AR) control emotional memory and fear conditioning, but it is not known if they affect the consolidation and generalization of fear, which was now investigated. We now report that A_2AR blockade through systemic administration of the A_2AR antagonist SCH58261 immediately after contextual fear conditioning (within the consolidation window), accelerated fear generalization. Conversely, A_2AR activation with CGS21680 decreased fear generalization. Ex vivo electrophysiological recordings of field excitatory post-synaptic potentials (fEPSPs) in CA3-CA1 synapses and of population spikes in the lateral amygdala (LA), showed that the effect of SCH58261 is associated with a reversion of fear conditioning-induced decrease of long-term potentiation (LTP) in the dorsal hippocampus (DH) and with increased amplitude of LA LTP in conditioned animals. These data suggest that A_2AR are engaged during contextual fear consolidation, controlling long-term potentiation mechanisms in both DH and LA during fear consolidation, impacting on fear generalization; this supports targeting A_2AR during fear consolidation to control aberrant fear processing in PTSD and other fear-related disorders.

Fig. 1. Blockade of A_2AR immediately after contextual fear conditioning increases fear consolidation and accelerates fear generalization.

A Scheme of the experimental design. Vehicle or SCH58261 (0.1 mg/kg) were administered intraperitoneally (i.p.) immediately after contextual fear conditioning (CFC - 3 shocks of 0.7 mA). B Individual values and mean ± SEM (n = 9–11) of the percentage of time freezing in context A (paired with foot-shocks) or in the unpaired context B, at 1 and 2 days after conditioning, respectively. C Discrimination index of CFC animals at 1 and 2 days after conditioning; when probed for recent fear memory, SCH58261-injected rats had a lower discrimination index compared with the saline (control) group, showing a worst ability to distinguish between contexts A and B. D Scheme of the experimental design, where vehicle or SCH58261 (0.1 mg/kg) were administered i.p. immediately after a weak CFC protocol (1 shock of 0.5 mA). E Individual values and mean ± SEM (n = 10) of the percentage of time freezing in the paired context A or in the unpaired context B, at 1 and 2 days after CFC, respectively. F Discrimination index of rats subjected to a weak CFC, probed at 1 and 2 days after conditioning. Only the animals that were injected with SCH58261 after CFC discriminated between contexts A and B when probed for recent fear memory. B, E *p < 0.05 and ***p < 0.01 compared to the control group, treated with vehicle (two-way ANOVA followed by Fisher’s LSD multiple comparison test); C, F **p < 0.01, compared to the group treated with vehicle (Student’s t test) and ^#p < 0.05 or ^##p < 0.01, one sample t test comparing with the hypothetical value of 0.5 (i.e., no discrimination between contexts A and B).

Fig. 2. The blockade of A_2AR accelerates fear generalization only when done during the early stages of memory consolidation.

A Scheme of the experimental design. Vehicle or SCH58261 (0.1 mg/kg) were administered i.p. 3 or 6 h after contextual fear conditioning (CFC - 3 shocks of 0.7 mA). B Individual values and mean ± SEM (n = 9) of the percentage of time freezing in context A (paired with foot-shocks), 1 day after conditioning. C Individual values and mean ± SEM (n = 9) of the percentage of time freezing in the unpaired context B at 2 days after conditioning. D Discrimination index probed at 1 and 2 days after conditioning; animals that were injected with SCH58261 3 h after CFC did not discriminate between contexts when probing for recent fear memory, unlike saline-treated animals and the rats injected with SCH58621 6 h after CFC. B, C *p < 0.05 compared to the control group treated with vehicle (one-way ANOVA followed by a Dunnett’s post hoc test); D ** p < 0.01 in relation to the control group treated with vehicle (one-way ANOVA followed by a Dunnett’s post hoc test) and #p < 0.05, one sample t test comparing with the hypothetical value of 0.5 (no discrimination between contexts A and B).

Fig. 3. Activation of A_2AR during fear consolidation delays fear generalization.

A Scheme of the experimental design. Vehicle or CGS21680 (0.2 mg/kg) were administered i.p. immediately after a strong intensity fear conditioning session (3 shocks of 1.2 mA). B Individual values and mean ± SEM (n = 7–10) of the percentage of time freezing in the paired context A and in the unpaired context B at 1 and 2 days after conditioning, respectively. C Discrimination index at 1 and 2 days after fear conditioning; both groups of animals discriminate between contexts A and B, when probed for recent fear memory. D Percentage of time freezing in the paired context A and in the unpaired context B at 14 and 15 days after conditioning, respectively, in the same group of animals. E Discrimination index at 14 and 15 days after fear conditioning; animals that were injected with CGS21680 still discriminate between contexts, in contrast to the saline group, when probed for remote fear memory. B, D *p < 0.05 in relation to the group treated with vehicle (two-way ANOVA followed by Fisher’s LSD multiple comparison test); C, E *p < 0.05 compared to the control group treated with vehicle (Student’s t test) and ^#p < 0.05, one sample t test when compared with the hypothetical value of 0.5 (no discrimination between contexts A and B).

Fig. 4. Blockade of A_2AR immediately after contextual fear conditioning attenuates conditioning-induced decrease in long-term potentiation in the dorsal hippocampus.

A Scheme of the experimental design. Animals were injected i.p. with vehicle or SCH58261 (0.1 mg/kg) immediately after CFC (3 shocks of 0.7 mA), corresponding to the groups labeled as CFC-veh (black circles) and CFC-SCH (black squares), or exposure to the conditioning chamber (context A) without application of electrical shocks, corresponding to the groups labeled as veh (gray circles) and SCH (gray squares). Electrophysiological recordings (field excitatory post-synaptic potential—fEPSP) were performed on slices of the dorsal hippocampus (DH) collected 2 h after the i.p. injections. B Input–output (I/O) curves at CA3-CA1 synapses of the DH showed no significant differences between the different experimental groups. C Paired pulse ratio (PPR) showed a paired-pulse facilitation with an interpulse interval of 50 ms with no differences between groups. D Time-course of the variation of the slope of fEPSPs, expressed as percentage of baseline values, in the CA1 stratum radiatum upon stimulation of afferent Schaffer collateral fibers, before and after induction of a long-term potentiation (LTP) with a train of high-frequency stimulation (HFS, 1 train of 100 Hz for 1 s, arrow). The inserts show representative recordings of the fEPSPs obtained for the indicated experimental groups, prior to LTP induction (filled traces) and 50–60 min after LTP induction (dashed lines). E Bar graph of LTP magnitude at 50–60 min after HFS. The blockade of A_2AR after CFC prevented the decrease of LTP magnitude induced by CFC in the DH. Individual values and mean ± SEM (n = 5–10) are presented in the bar graphs. *p < 0.05 in relation to the value of 1 (C) or 0 (E) (one sample t test, when comparing with a hypothetical value); ^#p < 0.05 between the indicated groups (two-way ANOVA and Fisher’s LSD multiple comparison test).

Fig. 5. The blockade of A_2AR immediately after contextual fear conditioning increases long-term potentiation in the lateral amygdala.

A Scheme of slices containing the amygdala showing the position of the electrodes for extracellular electrophysiological recordings of population spikes (PS) at the lateral amygdala (LA). The animals received vehicle or SCH58261 (0.1 mg/kg), i.p., immediately after CFC (3 shocks of 0.7 mA), corresponding to the groups labeled as CFC-veh (black circles) and CFC-SCH (black squares), or exposure to the conditioning chamber (context A) without application of electrical shocks, corresponding to the groups labeled as veh (gray circles) and SCH (gray squares). Electrophysiological recordings were performed on horizontal slices containing the lateral nuclei of the amygdala (LA), collected 2 h after the i.p. injections. B Input-output (I/O) curves at LA excitatory synapses showed no significant differences between the different experimental groups. C Paired pulse ratio (PPR) showed that a paired-pulse facilitation (with an interpulse interval of 30 ms) in vehicle-injected groups, which was abrogated after exposure to SCH58261, regardless of CFC. D Time-course of the variation of the amplitude of PS (expressed as percentage of baseline values) in the LA before and after induction of LTP with high-frequency stimulation (HFS, 3 trains of 100 Hz for 1 s, with 5 s inter-train interval). The inserts show representative recordings of the PS obtained for the indicated experimental groups, prior to LTP induction (filled traces) and 50–60 min after LTP induction (dashed lines). E Bar graph of LTP magnitude at 50–60 min after HFS. The blockade of A_2AR immediately after CFC increased LTP magnitude when compared with animals subjected to CFC and treated with vehicle. Values are expressed as mean ± SEM of n = 5–9 rats; * indicates a significant difference (p < 0.05) in relation to the value of 1 for C or 0 for E (one sample t test, comparing with a hypothetical value). # indicates a significant difference (p < 0.05) between the indicated groups, observed with two-way ANOVA followed by Fisher’s LSD multiple comparison test.

Fig. 6. The ability of blocking A_2AR immediately after contextual fear conditioning to normalize the increased long-term potentiation are independent of A₁R in the dorsal hippocampus but not in the lateral amygdala.

A, E Time-course of the variation, expressed as percentage of baseline values, of the slope of field excitatory post-synaptic potentials (fEPSPs) in the CA1 stratum radiatum upon stimulation of the afferent Schaffer collateral fibers (A) or of the amplitude of population spikes in synapses of the lateral amygdala (E), upon exposure to increasing concentrations of 2-chloroadenosine (CADO) in slices of rats receiving vehicle or SCH58261 (0.1 mg/kg), i.p., immediately after contextual fear conditioning (CFC - 3 shocks of 0.7 mA), corresponding to the groups labeled as CFC-veh (black circles) and CFC-SCH (black squares), or exposure to the conditioning chamber (context A) without application of electrical shocks, corresponding to the groups labeled as veh (gray circles) and SCH (gray squares). B, F The CADO concentration-dependent inhibition of fEPSPs or population spikes (and the fitted sigmoids constrained at 0 and 100%) indicate a lower efficiency of A₁R after CFC, irrespective of the exposure to SCH58261 in the DH (B), but no modification of A₁R efficiency in the LA (E). C, G The disinhibition by the A₁R antagonist, DPCPX (100 nM), of synaptic transmission is not modified by CFC or SCH58261 in the DH (C), whereas it is abolished by SCH58261 exposure in the LA, irrespective of CFC (G). D, H LTP magnitude in the DH was reduced by CFC and restored by exposure to SCH58261, irrespective of the absence of presence of DPCPX (D), whereas in the LA the magnitude of LTP was increased by exposure to SCH58261 in the absence, but not in the presence of DPCPX (H). Values are expressed as mean ± SEM of n = 5–7 rats; * indicates a significant difference (p < 0.05) in relation to the value of 1 for C (one sample t test, comparing with a hypothetical value). # indicates a significant difference (p < 0.05) between the indicated groups, observed with two-way ANOVA followed by Fisher’s LSD multiple comparison test.