Role of the vasopressin 1b receptor in rodent aggressive behavior and synaptic plasticity in hippocampal area CA2

Abstract

The vasopressin 1b receptor (Avpr1b) is critical for social memory and social aggression in rodents, yet little is known about its specific roles in these behaviors. Some clues to Avpr1b function can be gained from its profile of expression in the brain, which is largely limited to the pyramidal neurons of the CA2 region of the hippocampus, and from experiments showing that inactivation of the gene or antagonism of the receptor leads to a reduction in social aggression. Here we show that partial replacement of the Avpr1b through lentiviral delivery into the dorsal CA2 region restored the probability of socially motivated attack behavior in total Avpr1b knockout mice, without altering anxiety-like behaviors. To further explore the role of the Avpr1b in this hippocampal region, we examined the effects of Avpr1b agonists on pyramidal neurons in mouse and rat hippocampal slices. We found that selective Avpr1b agonists induced significant potentiation of excitatory synaptic responses in CA2, but not in CA1 or in slices from Avpr1b knockout mice. In a way that is mechanistically very similar to synaptic potentiation induced by oxytocin, Avpr1b agonist-induced potentiation of CA2 synapses relies on NMDA (N-methyl-D-aspartic acid) receptor activation, calcium and calcium/calmodulin-dependent protein kinase II activity, but not on cAMP-dependent protein kinase activity or presynaptic mechanisms. Our data indicate that the hippocampal CA2 is important for attacking in response to a male intruder and that the Avpr1b, likely through its role in regulating CA2 synaptic plasticity, is a necessary mediator.

Figure 1

Aggressive behavior is rescued in Avpr1b KO mice following bilateral injections of a lentivirus expressing the mouse Avpr1b into hippocampal area CA2. (a) The schematic diagrams on the left depict the mouse hippocampus at the three levels targeted for bilateral lentiviral replacement (at coordinates 1.2 to 2.2 mm posterior to the bregma). Image plates are adapted from a brain atlas. Images on the right highlight the expression of Avpr1b in the hippocampus detected by in situ hybridization histochemistry in a representative WT mouse (Wildtype), a Avpr1b KO mouse injected with a GFP control lentivirus (KO+GFP), and a KO mouse injected with the Avpr1b lentivirus (KO+Replace). Arrowheads indicate examples of high Avpr1b expression in the hippocampus (calibration bar = 2 mm). (b) An expanded view of the lentiviral expression from the boxed region in the image marked with a star in (a) is presented with the hippocampal cells layers superimposed. Note that the high Avpr1b expression from the lentivirus localized bilaterally at the injection sites in area CA2. (c) Aggressive behavior was restored in Avpr1b KO mice following bilateral injections of the Avpr1b lentivirus into area CA2 (KO+Replace), but not in mice injected with a virus to express GFP (KO+GFP). Behavior data are from 16 WT mice, 13 KO mice injected with a lentivirus containing a cytomegalovirus promoter-driven green fluorescent protein (CMV-GFP) construct (KO+GFP), and 19 KO mice injected with a lentivirus containing a construct for the Avpr1b (KO+Replace).

Figure 2

Synaptic potentiation is induced by Avpr1b and Oxtr agonists in slices of rat and mouse hippocampus. (a) Reflecting the enrichment of Avpr1b in CA2, (b) 50 nM d[Leu⁴,Lys⁸]-Avp, a selective Avpr1b agonist, induced potentiation of EPSCs recorded in rat CA2 (red circles; n=9) but not in CA1 (blue squares; n=6). The duration of drug application is indicated by the black bar in this and subsequent figures. Representative synaptic currents from time points before (1) and after (2) drug application are shown above the averaged results for areas CA2 and CA1. (c) Similar effects were observed in CA2 neurons in slices from WT (grey squares; n=6) and Oxtr KO (pink triangles; n=4) mice, but not in slices from Avpr1b KO mice (red circles; n=7). Representative currents are shown above the group data at the time points indicated by the numbers. Similar results were observed with 100 nM of the Oxtr agonist [Thr⁴,Gly⁷]-Oxt (d–f). The high levels of oxytocin receptor binding in areas CA2 and CA3 are depicted in the schematic diagram shown in (d), and (e) synaptic potentiation induced with the Oxtr agonist is observed in CA2 (pink circles; n=11) and CA3 (green triangles; n=7), but not CA1 (blue squares; n=8). Conventions in (e) and (f) are the same as in (b) and (c). (f) In addition, potentiation induced with [Thr⁴,Gly⁷]-Oxt was also observed in slices from WT mice (grey squares; n=6) and Avpr1b KO mice (red circles; n=11), but not in slices from Oxtr KO mice (pink triangles; n=7).

Figure 3

The potentiation of EPSCs in area CA2 induced by d[Leu⁴,Lys⁸]-Avp is mediated by a postsynaptic change in AMPA receptor function. AMPA- and NMDA-mediated currents were recorded before and after bath application of 50nM d[Leu⁴,Lys⁸]-Avp and synaptic stimulation for 15 min. Sample currents in (a) show that the AMPA-mediated EPSC is enhanced following application of the agonist. (b) This is also reflected in the group data by a significant increase in the AMPA-mediated EPSC (star, P<0.01), but not in the NMDA-mediated EPSC (ns, non-significant; n=12 each). (c) The AMPA/NMDA ratio was not significantly different. (d,e) The Avp1b agonist, d[Leu⁴,Lys⁸]-Avp, induces no significant change in paired-pulse facilitation. Pairs of stimulation pulses were delivered to the SC input to CA2 separated by an inter-pulse interval of 100 msec. Although 50nM d[Leu⁴,Lys⁸]-Avp enhanced the amplitude of synaptic currents in CA2, there was no change in paired-pulse facilitation; scaled example currents from a representative experiment are shown in (d) and group data are shown in (e); n=6 each. (f) In addition, responses-to-response variability in the amplitude of EPSCs was reduced following application of the Avpr1b agonist as indicated by an increase in the coefficient of variation (1/CV²) (n=9 each).

Figure 4

Avpr1b agonist-induced synaptic potentiation in CA2 is calcium dependent. (a) Proposed mechanism of Avpr1b-induced potentiation based on the idea that Avpr1b is coupled to Gq proteins and phospholipase C-dependent calcium increases. This mechanism is in contrast to the potentiation induced with antagonists of the CA2-enriched A1 adenosine receptor, which acts through a PKA-dependent pathway. (b) Avpr1b agonist-induced potentiation is blocked by application of 50 μM AP5, an inhibitor of NMDA receptors (blue circles; n=9) or by temporarily pausing stimulation during drug application (black triangles; n=7), indicating that synaptic glutamate release and NMDA receptors are required for the potentiation. For reference in all panels, the results of application of d[Leu⁴,Lys⁸]-Avp is indicated by the grey squares (n=9). (c) Similarly,loading cells with 15 mM BAPTA, a high affinity calcium chelator, blocked the potentiation induced by 50 nM d[Leu⁴,Lys⁸]-Avp (red triangles; n=5), indicating that this potentiation is Ca²⁺ dependent. In addition, 10 μM KN-62 (orange circles; n=7) and KN-93 (d; green triangles; n=6), inhibitors of CaMKII, included in the recording pipette also inhibited the potentiation. However, KN-92, an inactive analogue of KN-93 (brown circles; n=7), 20 μM PKI, an inhibitor of PKA, (magenta circles; n=8, e), or bicuculline, an inhibitor of GABA_A receptors (pink circles; n=7, f), failed to block the potentiation. Representative traces are shown as insets at the times indicated by 1 and 2 (calibration bars: 50 pA, 20 ms).