br Conclusions br Acknowledgments This work was supported
Acknowledgments This work was supported by a grant from NIH from the National Institutes on Arthritis and Musculoskeletal and Skin Diseases [AR056092].
Introduction Prior exposure to stressors has been consistently associated with the manifestation and/or exacerbation of psychiatric disorders, including anxiety disorders (Arborelius et al., 1999, Binder and Nemeroff, 2010). Corticotropin-releasing factor (CRF) is a hypothalamic neuropeptide that initiates and orchestrates stress hormonal, behavioral and autonomic responses (Reul and Holsboer, 2002, Vale et al., 1983). An increasing amount of evidence suggests that dysfunction in CRF-mediated neurotransmission/modulation may play an important role in some types of anxiety pathological conditions, particularly post-traumatic stress disorder (Binder and Nemeroff, 2010, Risbrough and Stein, 2006). CRF-modulating drugs, primarily those blocking CRF type 1 receptors (CRF1), have been considered as a putative therapeutic tool for these pathologies (Arborelius et al., 1999, Bailey et al., 2011, Valdez, 2006). More recently, evidence suggests that dysregulation of CRF signaling may also be implicated in panic disorder. For instance, polymorphisms in the CRF1, but not CRF2, gene have been associated with susceptibility to panic disorder in German and Japanese population samples (Ishitobi et al., 2012, Keck et al., 2008). There is also an association of polymorphisms in the CRF gene with inhibited temperament in children at risk for panic disorder (Smoller et al., 2005). Only few animal studies have so far addressed where in the 4E1RCat sale changes in CRF-mediated neurotransmission/modulation may influence panic-associated behavioral and autonomic responses. Sajdyk et al. (1999) reported that repeated intra-basolateral amygdala (BLA) injections of sub-anxiogenic doses of urocortin 1, a peptide that binds to both CRF1 and CRF2 receptors (Vaughan et al., 1995), induce synaptic/biochemical changes that render animals sensitive to sodium lactate. More specifically, in animals treated with this peptide, but not in sham-treated subjects, intravenous injection of lactate causes anxiogenic-like behavioral and cardiovascular responses. The importance of this finding to panic is that in panic disorder patients, but not in healthy volunteers, lactate evokes fear and autonomic responses similar as those experienced in a panic attack (Griez and Schruers, 1998). Therefore, it seems likely that a persistent subthreshold stimulation of CRF receptors in the BLA may predispose individuals to panic-evoking stimuli. Of importance to the present investigation, two studies (Carvalho-Netto et al., 2007, Litvin et al., 2007) investigated the role played by CRF receptors of the dorsal periaqueductal gray (dPAG), which express both subtypes of CRF ligand sites (Merchenthaler, 1984, Swanson et al., 1983), in the regulation of panic-related defensive behaviors. This midbrain area, which comprises the dorsolateral (dlPAG) and dorsomedial (dmPAG) columns of the periaqueductal gray matter, has been critically involved in the pathogeny of panic disorder (Del-Ben and Graeff, 2009, Graeff, 2002, Lovick et al., 2000). In laboratory animals, electrical or chemical stimulation of the dPAG evokes vigorous escape behavior that is accompanied by marked autonomic reactions similar to those observed when animals are confronted with predators (Jenck et al., 1995, Schenberg, 2010). Electrical stimulation of the dPAG in patients undergoing neurosurgery resulted in marked autonomic and sensory changes, accompanied by strong aversive emotional reactions (e.g. feelings of fear, terror and impending death) that resemble symptoms of a panic attack (Nashold et al., 1969). Using the mouse defense test battery, Carvalho-Netto et al. (2007) observed that intra-dPAG injection of CRF significantly increased the number of escape attempts made by mice when confronted with a predator, the rat. Based on evidence that this defensive response is facilitated and inhibited by panic-promoting (e.g. yohimbine) and -alleviating drugs (e.g. antidepressants), respectively (Blanchard et al., 1993, Blanchard et al., 2003), CRF\'s effect was interpreted as panicogenic. However, in a subsequent study using cortagine, the same group of researchers failed to replicate this finding (Litvin et al., 2007). Cortagine, differently from CRF, which binds to both CRF1 and CRF2 receptors (although with higher affinity for the former subtype), is a highly selective CRF1 agonist (Farrokhi et al., 2007). As pointed out by the authors (Litvin et al., 2007), the reasons for this discrepancy could be manifold, such as the use of insufficient doses of cortagine and/or differences in the sites of drug injection along the rostro-caudal plane of the dPAG. One intriguing possibility also raised is that the panicogenic effect caused by CRF resulted from an interaction between CRF1 and CRF2 receptor-mediated mechanisms in the dPAG, a process not triggered by the selective CRF1 agonist cortagine.