The consistent anti anxiety effects obtained
The consistent anti-anxiety effects obtained with intra-mPFC injections of CP 376395 strongly suggest a tonic role of CRF at CRF1 receptors located within this limbic Pioglitazone area in the modulation of anxiety in the mouse on the EPM. Intra-mPFC injection of CP 376395 led mice to explore the potentially aversive areas of the EPM to a greater degree. When given alone, all three doses (0.75, 1.5 and 3.0nmol) of this highly selective CRF1 antagonist attenuated spatiotemporal (%OE and %OT) and complementary (uHD, uSAP, OAEE) measures of anxiety without changing general locomotion (i.e., closed-arm entries). The present results are in line with previous findings showing an anxiolytic-like effect following the deletion of the CRF1 receptor gene in mice (Contarino et al., 1999, Smith et al., 1998, Timpl et al., 1998). Briefly, those authors showed that the lack of CRF1 receptor activity attenuates anxiety-like behavior of mice exposed to the EPM and light/dark box models. The hypothesis raised above, in that the role of endogenous CRF at CRF1 receptors located within the mPFC in the modulation of defensive behavior depends on the type of threatening situation confronting the animals, is strengthened by the lack of effect of local infusions of the acidic-astressin [Glu11,16]Ast, a CRF1 antagonist (Eckart et al., 2001), on the behavior of mice exposed to a predator (Pentkowski et al., 2013). Differently from what we showed in this study with the mouse EPM, those authors reported that intra-mPFC injections of 0.1 or 0.2μg of [Glu11,16]Ast did not change any behavioral measure in mice subjected to the rat exposure test. It is well established that CRF action is triggered by its binding to CRF1 or CRF2, two G protein-coupled receptors (GPCR). Perhaps the principal or best-known action mechanism following activation of both CRF receptors is through stimulatory G protein activation (GsP), which in turn stimulates the adenylyl cyclase-cAMP/PKA pathway, promoting phosphorylation of CREB (cAMP responsive element-binding protein) (Hauger et al., 2006), an important transcription factor (Tully, 1998). However, CRF1 and CRF2 receptor activation may reach CREB phosphorylation via other pathways. For instance, their activation stimulates the Gq protein pathway, leading to PKC activation and calcium release, both of which are involved in CREB phosphorylation (Silva et al., 1998). Moreover, CRF1 and CRF2 activation can lead to CREB phosphorylation through an extra-cAMP/PKA pathway, namely the MEK/MAPK pathway (Hillhouse and Grammatopoulos, 2006, Stern et al., 2011), as well as inducing activation of nitric oxide/cGMP, caspase enzyme, NFκB and Akt/PK (Hillhouse and Grammatopoulos, 2006). The role of these pathways in the effects of CRF on anxiety is still poorly known. Here, we investigated whether the cAMP/PKA pathway is involved in the anxiogenic-like effect produced by CRF injection into the mPFC of mice. To that end, we first investigated the effects of H-89 (1.25, 2.5 or 5.0nmol) injected alone into the mPFC of mice exposed to the EPM. Corroborating the results obtained with CP376395, the highest dose (5.0nmol) of this PKA inhibitor also attenuated conventional and complementary indices of anxiety, suggesting that the tonic anxiogenic-like role of CRF within the mPFC depends on intracellular cAMP/PKA activation. This hypothesis is strengthened by the blockade of the anxiogenic-like effects induced by intra-mPFC CRF with a prior local injection of 2.5nmol of H-89, a dose devoid of intrinsic effects on the behavior of mice exposed to the EPM. H-89 at 2.5nmol selectively blocked the anxiogenic effects produced by CRF injection into the mPFC of mice. The enhancement of open-arm aversion, characterized by reductions of percent of open-arm entries, percent of open-arm time, uHD and OAEE, and increase of pSAP in CRF-injected mice (Veh+CRF) was prevented by local infusion of H-89. Importantly, when given alone, H-89 (H-89+Veh) was incapable of changing locomotor activity (i.e., closed-arm entries), suggesting that the inhibition of PKA selectively attenuated anxiety-like behaviors induced by CRF in mice exposed to the EPM. Two intriguing individual results produced with intra-mPFC injections of 150pmol of CRF were observed in Experiments 1 and 4. While this dose of CRF tended to increase locomotor behavior (i.e. it tended to increase closed-arm entries) and left uSAP unchanged in Exp. 4, it decreased uSAP without altering closed-arm entries in Exp. 1. We do not have a clear explanation for these apparently inconsistent results. However, they might be related to the experimental procedure involving two intra-mPFC injections employed in Exp. 4. Taken together, the present results suggest that the mechanism of action by which CRF induces anxiety within the mPFC of Swiss mice involves the activation of Gs protein/adenylyl cyclase/cAMP/PKA cascade, probably leading to CREB phosphorylation. However, the present results do not exclude the possibility that CRF modulates emotional states via other signaling cascades such as PKC, MAPK and/or even NO production. Further studies would help to clarify these points.