Several lines of evidence suggest that
Several lines of evidence suggest that the anxiety provoking behavioral effects of CRF involve CRF1 receptors. CRF1 knockout mice show increased anxiolytic-like activity in the elevated plus maze and a tendency to enter the illuminated region of a light–dark box , . In addition, studies using antisense oligonucleotides directed against the CRF1 Auranofin reported anxiolytic-like effects in several different tests including the elevated plus maze, the open field, and the defensive-withdrawal tests , , . Finally, treatment with antagonists that selectively block CRF1 receptor action was shown to promote anxiolytic responses in the elevated plus maze, the light–dark box, the mouse defense test battery, and the fear-potentiated startle test , , , .
In contrast, the role of CRF2 receptors in anxiety behavior is less clear. For example, behavioral differences were noted in three recent CRF2 knockout studies. Using open-field center exploration as a measure, one study reported that CRF2 knockout mice showed increased center activity , another found a decrease in center scores  and a third study showed no significant changes in center exploration . In addition, using the elevated plus maze test, one study found no effects of CRF2 knockout on anxiety behavior , whereas two other studies reported a decrease in open-arm exploration which suggests an increase in anxiety , . In other work incorporating an antisense approach, two studies reported that central infusion of CRF2 antisense oligonucleotides had no effect on anxiety behavior in rats exposed to the elevated plus maze or defensive-withdrawal test , . However, these negative results should be interpreted cautiously because in one study the level of CRF2 receptor reduction was ∼20% and the animals exhibited a significant weight loss  and in the other report no anatomical details were provided on the amount of CRF2 receptor knockdown . A high affinity CRF2 peptide antagonist, antisauvagine-30 (anti-SVG-30) , has recently been tested in mice exposed to the elevated plus maze. However, whereas one study reported that i.c.v. infusion of anti-SVG-30 produced an increase in anxiety behavior , another study reported that infusion of the CRF2 antagonist into the lateral septum blocked the effects of immobilization stress on open arm exploration . In that latter study, intraseptal delivery of anti-SVG-30 was also shown to have no effects on open-arm exploration when mice were tested in the absence of prior stress. These seemingly conflicting behavioral results may be due, in part, to differences in regional brain antagonism of CRF2 receptors stemming from i.c.v. infusions versus site specific administration. The former procedure may allow the peptide to antagonize CRF2 receptors in the lateral septum as well as other brain CRF2 receptor sites. It is also possible that CRF2 receptors may play a prominent role in the expression of stress-induced anxiety behavior.
Therefore, to provide additional information on the role of CRF2 receptors in anxiety, we conducted dose–response studies using anti-SVG-30 and examined the behavior of rats in the conditioned anxiety, the elevated plus maze, and the defensive-withdrawal test. The conditioned anxiety test assesses the effects of prior foot shock stress on facilitating the occurrence of freezing, a measure of fear and anxiety. The elevated plus maze and defensive-withdrawal tests evaluate the spontaneous or unconditioned anxiety promoting effects of an elevated open runway or a large illuminated open field. These three animal models of anxiety are frequently used to evaluate the actions of CRF and other neuropeptide and neurotransmitter compounds , , , .
Discussion The results demonstrate that anti-SVG-30 produced a dose-dependent reduction in the duration of conditioned freezing, increased the level of elevated open arm exploration, and increased the amount of time spent in an open field. Hence, antagonism of CRF2 receptors produces a consistent anxiolytic-like behavioral profile in a number of different animal anxiety models. In addition, a behaviorally effective dose of anti-SVG-30 (10.0 μg) had no significant effects on locomotor activity. Although higher doses of anti-SVG-30 were not tested for their potential intrinsic behavioral effects, current results suggest that antagonism of CRF2 receptors, at a dose effective in attenuating freezing and increasing behavioral exploration in opened or unprotected areas, is not due to an antagonist-induced increase in motor behavior.