It is important to note that CRF may

It is important to note that CRF1 may be activated during acute stress and early phases of anxiety disorders (Coric et al., 2010, Ising et al., 2007), as well the administration of Astressin 2B into the lateral septum did not have an effect on anxiety-like behavior in low-stress conditions but had an anxiolytic-like effect when animals were submitted to restraint stress (Henry et al., 2006). Thus, it is possible to suggest that the TI modulation in guinea pigs sharing similar mechanisms with acute stress and spontaneous anxiety rather than with chronic states where stable anxiety levels have been established. Reinforcing this hypothesis, the urocortin 1-cointaining neurons of Edinger-Westphal nucleus were only activated in response to acute restraint stress (de Andrade et al., 2014) as well this nucleus has shown high densities of FOS-immunoreactivity hyPerFUsion™ high-fidelity PCR Kit induced by TI response in guinea pigs (Vieira et al., 2011). In line with this view, Coric et al. (2010) have shown that CRF1 receptor antagonist did not have significant anxiolytic properties in a placebo controlled trial in generalized anxiety disorders. Further, in a preclinical model of anxiety disorders CRF antagonists blocked stress-mediated amygdala alterations in the early stages of plasticity, but were ineffective once full plasticity and a chronic anxiety state was established (Rainnie et al., 2004). Consistent with our data, a recent study has shown that transient CRF2 activation in neurons located in the lateral septum promotes persistent anxious behavior, probably due to the positive regulation of corticosteroid levels (Anthony et al., 2014). So, we speculate it is possible that blockade of CRF1 and CRF2 in the BLA or CeA promoted an alteration of corticosteroid levels along with reduction of innate fear, which, in turn, reduced the TI response in guinea pigs. Indeed, TI response susceptibility is positively correlated with corticosterone plasma levels suggesting the involvement of the pituitary-adrenocortical axis in modulation of this behavior (Carli et al., 1979). In contrast to our results related to CRF2, Skórzewska and colleagues (Skórzewska et al., 2011) reported that intracerebroventricular administration of CRF2-selective antagonists (antisauvagine-30 or Astressin 2B) increased a conditioned fear freezing response and the conditioned fear-elevated concentration of serum corticosterone. These authors suggested that a selective blockade of CRF2 receptors could facilitate the conduction of signals via CRF1 receptors, and may contribute to the enhancement of anxiety-like responses. From this point of view, our results are distinct, possibly because the TI response is an emotional behavior critical for survival in the environment (Klemm, 2001, Ratner, 1967); therefore, it is elicited in high-stress situations, and under these conditions, CRF2 receptors are essential (Henry et al., 2006). In line with the present data, a previous report has shown that overexpression of the CRF2 receptor in the bed nucleus of the stria terminalis improves posttraumatic stress disorder-like symptoms (Elharrar et al., 2013). This result suggests that the CRF2 receptor promotes anxiety-like responses rather than reduces anxiety-related behavior. So, blockade of the CRF2 receptor in the BLA or CeA probably reduced fear and anxiety, as suggested by the decrease in the TI response. From the same perspective, previous studies by Sajdyk and colleagues (Sajdyk et al., 1999) and Spiga and colleagues (Spiga et al., 2006) have shown that the administration of CRF or urocortin 1 (a selective agonist for CRF1 and CRF2 receptors but with a higher affinity for the CRF1 receptor) into the BLA produced an anxiogenic effect, as evaluated by a social interaction test in rats; this effect was reversed by the antagonism of CRF1 receptors. Again, intracerebroventricular or systemic injections of non-peptidic CRF1 antagonists (CP-154,526, antalarmin or DMP696) reduced defensive behavior in the elevated plus maze test and light/dark test [for review Bale and Vale (2004) and Carrasco and Van De Kar (2003)]. However, intra-septal, intracerebroventricular or systemic injections of CRF1 agonists attenuated defensive behavior (Radulovic et al., 1999). The study of Iemolo et al. (2013) strongly support the anxiolytic effect of CeA CRF1 blockage. They reported that a selective CRF1 antagonist (R121919) in CeA blocked the anxiogenic effect induced by palatable food withdrawn in a model for food intake in rats (Iemolo et al., 2013). Further, another finding showed that the administration of CRF2 receptor antagonists (antisauvagine-30) in the BLA was not effective in reducing fear responses in aversive conditioning (Hubbard et al., 2007). However, the administration of Astressin 2B into the medial amygdala decreased T-maze avoidance latencies, an anxiolytic-like effect (Alves et al., 2016). Together, these results support the involvement of CRF1 and CRF2 receptors in the modulation of fear and anxiety behaviors, but the ways in which these responses are modulated should be clarified. It is possible that the differences in behavioral effects are related to the independent neural circuits that mediate the distinct ethologic models correlated with defensive responses (Lowry and Moore, 2006).