Intrahippocampal injection of brain-derived neurotrophic factor increases anxiety-related, but not panic-related defensive responses: involvement of serotonin.

Changes in brain-derived neurotrophic factor (BDNF)-mediated signaling in the hippocampus have been implicated in the etiology of depression and in the mode of action of antidepressant drugs. There is also evidence from animal studies to suggest that BDNF-induced changes in the hippocampus may play a role in another stress-related pathology: anxiety. However, it is still unknown whether this neurotrophin plays a differential role in defensive responses associated with distinguished subtypes of anxiety disorders found in the clinic, such as generalized anxiety and panic disorder. In the present study, we investigated the effect of an acute BDNF injection into the rat dorsal hippocampus (DH) on inhibitory avoidance acquisition and escape expression measured in the elevated T-maze (ETM). We also assessed whether serotonergic neurotransmission may account for such effects. Intra-DH BDNF injection (200 pg) facilitated inhibitory avoidance in ETM. BDNF was equally anxiogenic in the light/dark transition test. Preadministration of the 5-HT1A receptor antagonist WAY-100635 fully counteracted the anxiogenic effect of BDNF in both tests. Intra-DH midazolam administration (10 nmol) impaired avoidance acquisition in ETM, suggesting an anxiolytic effect. Therefore, in the DH, facilitation of BDNF signaling seems to enhance 5-HT1A receptor-mediated neurotransmission to exert an anxiogenic effect associated with generalized anxiety.

Role of 5-HT1A and 5-HT2C receptors of the dorsal periaqueductal gray in the anxiety- and panic-modulating effects of antidepressants in rats.

Antidepressant drugs are first-line treatment for panic disorder. Facilitation of 5-HT1A receptor-mediated neurotransmission in the dorsal periaqueductal gray (dPAG), a key panic-associated area, has been implicated in the panicolytic effect of the selective serotonin reuptake inhibitor fluoxetine. However, it is still unknown whether this mechanism accounts for the antipanic effect of other classes of antidepressants drugs (ADs) and whether the 5-HT interaction with 5-HT2C receptors in this midbrain area (which increases anxiety) is implicated in the anxiogenic effect caused by short-term treatment with ADs. The results showed that previous injection of the 5-HT1A receptor antagonist WAY-100635 in the dPAG blocked the panicolytic-like effect caused by chronic systemic administration of the tricyclic AD imipramine in male Wistar rats tested in the elevated T-maze. Neither chronic treatment with imipramine nor fluoxetine changed the expression of 5-HT1A receptors in the dPAG. Treatment with these ADs also failed to significantly change ERK1/2 (extracellular-signal regulated kinase) phosphorylation level in this midbrain area. Blockade of 5-HT2C receptors in the dPAG with the 5-HT2C receptor antagonist SB-242084 did not change the anxiogenic effect caused by a single acute injection of fluoxetine or imipramine in the Vogel conflict test. These results reinforce the view that the facilitation of 5-HT1A receptor-mediated neurotransmission in the dPAG is a common mechanism involved in the panicolytic effect caused by chronic administration of ADs. On the other hand, the anxiogenic effect observed after short-term treatment with these drugs does not depend on 5-HT2C receptors located in the dPAG.

Panicolytic-like effect of BDNF in the rat dorsal periaqueductal grey matter: the role of 5-HT and GABA.

A wealth of evidence suggests a role for brain-derived neurotrophic factor (BDNF) and its receptor tropomyosin-related kinase B (TrkB) in the aetiology of depression and in the mode of action of antidepressant drugs. Less clear is the involvement of this neurotrophin in other stress-related pathologies such as anxiety disorders. The dorsal periaqueductal grey matter (DPAG), a midbrain area rich in BDNF and TrkB receptor mRNAs and proteins, has been considered a key structure in the pathophysiology of panic disorder. In this study we investigated the effect of intra-DPAG injection of BDNF in a proposed animal model of panic: the escape response evoked by the electrical stimulation of the same midbrain area. To this end, the intensity of electrical current that needed to be applied to DPAG to evoke escape behaviour was measured before and after microinjection of BDNF. We also assessed whether 5-HT- or GABA-related mechanisms may account for the putative behavioural/autonomic effects of the neurotrophin. BDNF (0.05, 0.1, 0.2 ng) dose-dependently inhibited escape performance, suggesting a panicolytic-like effect. Local microinjection of K252a, an antagonist of TrkB receptors, or bicuculline, a GABAA receptor antagonist, blocked this effect. Intra-DPAG administration of WAY-100635 or ketanserin, respectively 5-HT1A and 5-HT2A/2C receptor antagonists, did not alter BDNF's effects on escape. Bicuculline also blocked the inhibitory effect of BDNF on mean arterial pressure increase caused by electrical stimulation of DPAG. Therefore, in the DPAG, BDNF-TrkB signalling interacts with the GABAergic system to cause a panicolytic-like effect.

Facilitation of 5-HT1A-mediated neurotransmission in dorsal periaqueductal grey matter accounts for the panicolytic-like effect of chronic fluoxetine.

Chronic administration of antidepressants such as fluoxetine and imipramine increases the responsiveness of 5-HT(1A) receptors in dorsal periaqueductal grey matter (DPAG), a midbrain area consistently implicated in the pathogenesis of panic disorder. This effect has been related to the clinically relevant anti-panic action of these drugs. In this study we determined whether long-term administration of fluoxetine also affects 5-HT efflux in DPAG. As a comparison, the effect of chronic treatment with the anxiolytic 5-HT(1A) receptor agonist buspirone on DPAG 5-HT levels was assessed. We also investigated whether the inhibitory effect of chronic fluoxetine on escape behaviour in the rat elevated T-maze, considered as a panicolytic-like effect, is counteracted by intra-DPAG injection of the 5-HT(1A) receptor antagonist WAY 100635. Male Wistar rats were treated (1 or 21 d, i.p.) with fluoxetine, buspirone or vehicle, once daily. After treatment, 5-HT in DPAG was measured by in-vivo microdialysis coupled to HPLC. In another study, rats treated (21 d, i.p.) with either fluoxetine or vehicle also received intra-DPAG injection of WAY 100635 or saline 10 min before being tested in the elevated T-maze. Chronic, but not acute, administration of fluoxetine significantly raised extracellular levels of 5-HT in DPAG. Long-term treatment with buspirone was ineffective. In the elevated T-maze, intra-DPAG injection of WAY 100635 fully blocked the anti-escape effect of chronic administration of fluoxetine. Therefore, chronic fluoxetine facilitates 5-HT(1A)-mediated neurotransmission within DPAG and this effect accounts for the panicolytic-like effect of this antidepressant in the elevated T-maze.

Alprazolam potentiates the antiaversive effect induced by the activation of 5-HT(1A) and 5-HT (2A) receptors in the rat dorsal periaqueductal gray.

RATIONALE: Serotonin in the dorsal periaqueductal gray (DPAG) through the activation of 5-HT(1A) and 5-HT(2A) receptors inhibits escape, a defensive behavior associated with panic attacks. Long-term treatment with antipanic drugs that nonselectively or selectively blocks the reuptake of serotonin (e.g., imipramine and fluoxetine, respectively) enhances the inhibitory effect on escape caused by intra-DPAG injection of 5-HT(1A) and 5-HT(2A) receptor agonists. It has been proposed that these compounds exert their effect on panic by facilitating 5-HT-mediated neurotransmission in the DPAG. OBJECTIVES: The objective of this study was to investigate whether facilitation of 5-HT neurotransmission in the DPAG is also observed after treatment with alprazolam, a pharmacologically distinct antipanic drug that acts primarily as a high potency benzodiazepine receptor agonist. MATERIALS AND METHODS: Male Wistar rats, subchronically (3-6 days) or chronically (14-17 days) treated with alprazolam (2 and 4 mg/kg, i.p.) were intra-DPAG injected with (+/-)-8-hydroxy-2-(di-n-propylamino)tetralin hydrobromide (8-OH-DPAT), (+/-)-1-(2,5-dimethoxy-4-iodophenyl) piperazine dihydrochloride (DOI), and midazolam, respectively, 5-HT(1A), 5-HT(2A/2C), and benzodiazepine receptor agonists. The intensity of electrical current that needed to be applied to the DPAG to evoke escape behavior was measured before and after the microinjection of these agonists. RESULTS: Intra-DPAG injection of the 5-HT agonists and midazolam increased the escape threshold in all groups of animals tested, indicating a panicolytic-like effect. The inhibitory effect of 8-OH-DPAT and DOI, but not midazolam, was significantly higher in animals receiving long-, but not short-term treatment with alprazolam. CONCLUSIONS: Alprazolam as antidepressants compounds facilitates 5-HT(1A)- and 5-HT(2A)-receptor-mediated neurotransmission in the DPAG, implicating this effect in the mode of action of different classes of antipanic drugs.

Cholecystokinin-2 receptors modulate freezing and escape behaviors evoked by the electrical stimulation of the rat dorsolateral periaqueductal gray.

Systemic injection of the cholecystokinin type 2 (CCK(2)) receptor agonist CCK-4 evokes panic attacks in humans and facilitates the expression of a panic-related defensive behavior, escape, in rats. Given the prominent role attributed to the dorsal periaqueductal gray (dPAG) in the pathophysiology of panic, this midbrain area has been assumed to be one of the key regions mediating these effects of CCK-4. However, only a few studies have directly investigated the role of dPAG CCK(2) receptors in the regulation of panic-related behaviors. Even more disappointingly, the results of these investigations have been far from conclusive. In the present study we further addressed this issue by evaluating the effect of the intra-dorsolateral periaqueductal gray (dlPAG) injection of CCK-4 on two panic-related defensive behaviors, freezing and escape, evoked in male Wistar rats by the electrical stimulation of the dlPAG. The effects of CCK-4 (0.005-0.5 microg/0.2 microl) were compared to those caused by the local microinjection of the CCK(2) receptor antagonist LY225910 (0.001-1.0 microg/0.2 microl). The results showed that whereas CCK-4 facilitated the expression of both freezing and escape behaviors, LY225910 had the opposite effect. Pretreatment with an ineffective dose of LY225910 prevented the panicogenic-like effect of CCK-4. These results strengthen the view that CCK(2) receptors located in the dlPAG are involved in the regulation of panic-related behaviors and may mediate the effect of CCK-4 on panic.

Effects of fluoxetine and buspirone on the panicolytic-like response induced by the activation of 5-HT1A and 5-HT2A receptors in the rat dorsal periaqueductal gray.

RATIONALE: Administration of 5-hydroxytryptamine (5-HT)1A and 5-HT2A receptor agonists into the dorsal periaqueductal gray (DPAG) inhibits escape, a defensive behavior associated with panic attacks. Long-term treatment with the antipanic compound imipramine enhances the DPAG 5-HT1A- and 5-HT2A-receptor-mediated inhibition of escape, implicating these receptors in the mode of action of panicolytic drugs. OBJECTIVES: In the present study, we investigated whether the inhibitory effect on escape elicited by the intra-DPAG injection of 5-HT1A and 5-HT2A receptor agonists is also enhanced after treatment with fluoxetine, another widely used antipanic drug. The effects of fluoxetine were compared to those of buspirone, an anxiolytic drug without major effect on panic disorder. METHODS: Male Wistar rats, subchronically (3-6 days) or chronically (21-24 days) treated with fluoxetine (10 mg/kg i.p.) or chronically treated with buspirone (0.3 mg/kg i.p.), were intra-DPAG injected with 5-HT (20 nmol), the 5-HT1A receptor agonist (+/-)-8-hydroxy-2-(di-n-propylamino)tetralin hydrobromide (8-OH-DPAT; 8 nmol) or the preferential 5-HT2A receptor agonist (+/-)-1-(2,5-dimethoxy-4-iodophenyl) piperazine dihydrochloride (DOI; 16 nmol). The intensity of electrical current that applied to the DPAG-evoked escape behavior was measured before and after the microinjection of these agonists. RESULTS: The electrical current necessary to produce escape was increased after the microinjection of the three 5-HT receptor agonists in all groups of animals tested. However, this panicolytic-like effect was significantly higher in animals receiving long-term treatment with fluoxetine. CONCLUSIONS: The results suggest that facilitation of the 5-HT1A- and 5-HT2A-receptor-mediated inhibition of DPAG neuronal activity is implicated in the beneficial effect of antidepressants in panic disorder.

Panic-modulating effects of alprazolam, moclobemide and sumatriptan in the rat elevated T-maze.

The elevated T-maze was developed to test the hypothesis that serotonin plays an opposing role in the regulation of defensive behaviors associated with anxiety and panic. Previous pharmacological exploitation of this test supports the association between inhibitory avoidance acquisition and escape expression with anxiety and fear/panic, respectively. In the present study, we extend the pharmacological validation of this test by investigating the effects of other putative or clinically effective anxiety- and panic-modulating drugs. The results showed that chronic, but not acute injection of the reversible monoamine oxidase-A inhibitor moclobemide (3, 10 and 30mg/kg) inhibited escape expression, indicating a panicolytic-like effect. The same effect was observed after either acute or chronic treatment with alprazolam (1, 2 and 4mg/kg), a high potency benzodiazepine. This drug also impaired inhibitory avoidance acquisition, suggesting an anxiolytic effect. On the other hand, subcutaneous administration of the 5-HT1D/1B receptor agonist sumatriptan (0.1, 0.5 and 2.5µg/kg) facilitated escape performance, indicating a panicogenic-like effect, while treatment with α-para-chlorophenylalanine (p-CPA; 4days i.p injections of 100mg/kg, or a single i.p injection of 300mg/kg), which caused marked 5-HT depletion in the amygdala and striatum, was without effect. Altogether, these results are in full agreement with the clinical effects of these compounds and offer further evidence that the elevated T-maze has broad predictive validity for the effects of anxiety- and panic-modulating drugs.

Inhibitory avoidance memory retention in the elevated T-maze is impaired after perivascular manipulation of the common carotid arteries.

Perivascular manipulation promoted by the positioning of a silicone collar around the common carotid arteries causes local inflammation and has been suggested as an animal model of atherosclerosis. This manipulation induces biochemical and morphological changes that are similar to those observed in the early stage of atherosclerosis in humans. Based on evidences showing that atherosclerosis is associated with cognitive deficits in humans, we presently investigated the temporal consequences of the bilateral positioning of silicone collars around the common carotid arteries (n = 15) on inhibitory avoidance memory retention in male Wistar rats tested in the elevated T-maze. The effects of this procedure were compared to those observed in sham-operated animals (n = 15) and to those observed in animals submitted to permanent bilateral occlusion of the common carotid arteries (n = 16). Additionally we studied the effects of the pretreatment with the non-selective anti-inflammatory drug indomethacin (n = 13) or the selective COX-2 inhibitor celecoxib (n = 12) and compared the effects to those of the pretreatment with vehicle (n = 11). The results showed that the silicone collar implants induced deficits in memory retention when animals were tested 2 and 4, but not 15 or 30, days after surgery. Permanent bilateral occlusion of the common carotid arteries impaired avoidance retention up to 30 days after surgery. Pretreatment with indomethacin (2 mg/kg/day) or celecoxib (5 mg/kg/day) post surgery and up to 3 days thereafter did not prevent memory deficits caused by silicone collar implants. Our data suggest that the prostanoids that participate in the inflammatory process triggered by the placement of the silicone collar do not seem responsible for the deficit in memory retention observed during the first days after collar placement.

Dorsomedial hypothalamus serotonin 1A receptors mediate a panic-related response in the elevated T-maze.

The dorsomedial hypothalamus (DMH) has long been associated with the regulation of escape, a panic-related defensive response. Previous evidence has shown that the activation of serotonin (5-HT) 1A and 2A receptors impairs escape behavior induced by the electrical stimulation of the same region. In this study we further explore the relationship of the DMH with defense by investigating the effects of 5-HT1A activation on escape behavior generated in male Wistar rats by an ethologically based aversive stimuli, exposure to one of the open arms of the elevated T-maze (ETM). Aside from escape, the ETM also allows the measurement of inhibitory avoidance, a defensive response associated with generalized anxiety disorder. To evaluate locomotor activity, after ETM measurements animals were submitted to an open field. Results showed that intra-DMH administration of the 5-HT1A receptor agonist 8-OH-DPAT inhibited escape expression. Local administration of the 5-HT1A antagonist WAY-100635 by its own was ineffective, but blocked the panicolytic-like effect of 8-OH-DPAT. Chronic (21 days) systemic treatment with imipramine potentiated the anti-escape effect of 8-OH-DPAT. No significant effects of treatment with 8-OH-DPAT or imipramine on avoidance latencies or the number of lines crossed in the open field were found. These results indicate that 5-HT1A receptors within the DMH may play a phasic inhibitory role on ETM escape expression. As previously proposed, facilitation of 5-HT1A-mediated neurotransmission in the DMH may be involved in the mechanism of action of anti-panic compounds.

Influence of procedural variables on rat inhibitory avoidance and escape behaviors generated by the elevated T-maze.

A wealth of evidence indicates that changes in procedural parameters and/or environmental conditions may exert a remarkable influence on the basal expression of defensive behaviors in different animal tests of anxiety. The goal of the current study was to further investigate the influence of procedural factors upon inhibitory avoidance acquisition and escape expression of male Wistar rats exposed to the elevated T-maze. These responses have been related in terms of psychopathology to generalized anxiety and panic disorders, respectively. Our results showed that the expression of these behaviors is not affected by prior handling of the animals or by increasing the illumination level of the experimental room from 60 to 580lx. They also showed that enhancing the number of avoidance trials from 3 to 6 favors the acquisition of this behavior. Under this condition, both diazepam (2mg/kg) and clonazepam (1-4mg/kg) caused anxiolytic effects, but only the latter benzodiazepine impaired escape expression, a panicolytic-like effect. In animals exposed to the elevated T-maze whole apparatus 24h before the test, the anxiolytic effect of these drugs was canceled out, which is consistent with the one-trial tolerance phenomenon widely observed in the elevated plus-maze. This procedure, however, does not interfere with the anti-escape effect caused by clonazepam. These results suggest that a 6-trial avoidance learning protocol may be a useful measure for compensating possible individual differences in the acquisition of this defensive response and to improve drug detection in the test.

5-HT1A and 5-HT2A receptor control of a panic-like defensive response in the rat dorsomedial hypothalamic nucleus.

The dorsomedial nucleus of the hypothalamus (DMH) has long been implicated in the genesis/regulation of escape, a panic-related defensive behavior. In the dorsal periaqueductal gray matter (dPAG), another key panic-associated area, serotonin, through the activation of 5-HT1A and 5-HT2A receptors, exerts an inhibitory role on escape expression. This panicolytic-like effect is facilitated by chronic treatment with clinically effective antipanic drugs such as fluoxetine and imipramine. It is still unclear whether serotonin within the DMH plays a similar regulatory action. The results showed that intra-DMH injection of the 5-HT1A receptor agonist 8-OH-DPAT, the preferential 5-HT2A receptor agonist DOI, but not the 5-HT2C agonist MK-212, inhibited the escape reaction of male Wistar rats evoked by electrical stimulation of the DMH. Local microinjection of the 5-HT1A antagonist WAY-100635 or the preferential 5-HT2A antagonist ketanserin was ineffective. Whereas chronic (21 days) systemic treatment with imipramine potentiated the anti-escape effect of both 8-OH-DPAT and DOI, repeated administration of fluoxetine enhanced the effect of the latter agonist. The results indicate that 5-HT1A and 5-HT2A receptors within the DMH play a phasic inhibitory role upon escape expression, as previously reported in the dPAG. Facilitation of 5-HT-mediated neurotransmission in the DMH may be implicated in the mode of action of antipanic drugs.

5-HT2C receptor regulation of defensive responses in the rat dorsal periaqueductal gray.

Activation of 5-HT2C receptors in limbic structures such as the amygdala and hippocampus increases anxiety. Indirect evidence obtained with non-selective 5-HT2C-interacting drugs suggests that the same may occur in the dPAG, a brainstem region consistently implicated in the genesis/regulation of panic attacks. In this study we used more selective agonists and antagonists to unveil the role played by dPAG 5-HT2C receptors in the regulation of anxiety- and panic-related defensive behaviors. Our results showed that intra-dPAG microinjection of the endogenous agonist 5-HT (20 nmol) or the 5-HT2C receptor agonists MK-212 (1 and 10 nmol) and RO-600175 (40 nmol) significantly increased inhibitory avoidance acquisition in rats tested in the elevated T-maze, suggesting an anxiogenic effect. 5-HT, but not the two 5-HT2C receptor agonists, inhibited escape performance. In the elevated T-maze, inhibitory avoidance and escape responses have been related to generalized anxiety and panic attacks, respectively. The behavioral effects caused by 5-HT and MK-212 were fully blocked by previous local microinjection of the 5-HT2C receptor antagonist SB-242084. Intra-dPAG injection of MK-212 also failed to affect escape expression in another test relating this behavior to panic, the electrical stimulation of the dPAG. Overall, the results indicate that 5-HT2C receptors in the dPAG are preferentially involved in the regulation of defensive behaviors related to anxiety, but not panic. This finding extends to the dPAG the prominent role that has been attributed to 5-HT2C receptors in anxiety generation.

Serotonin-2A receptor regulation of panic-like behavior in the rat dorsal periaqueductal gray matter: the role of GABA.

RATIONALE: Electrical stimulation of the dorsal periaqueductal gray (dPAG) evokes escape, a defensive response associated with panic attacks. Stimulation of 5-HT1A or 5-HT2A receptors in this midbrain area equally inhibits escape performance, even though at the molecular level these receptors cause opposite effects, i.e., activation of the former hyperpolarizes the cell membrane, while the latter excites it. A proposal has been made that 5-HT2A receptor agonists exert their inhibitory effect on escape by activating GABAergic interneurons located in the dPAG. OBJECTIVES: In the present study, we evaluated this hypothesis by investigating whether previous intra-dPAG administration of the GABAA receptor antagonist bicuculline blocks the anti-escape effect caused by the local injection of different 5-HT2A/2C receptor agonists. RESULTS: Intra-dPAG administration of 5-HT, the preferential 5-HT2A receptor agonist DOI, the nonselective 5-HT2C receptor agonist mCPP or the 5-HT2C receptor agonist RO 60-0175 significantly inhibited the escape reaction induced by electrical stimulation of the same brain area. In all cases, this panicolytic-like effect was blocked by previous microinjection of bicuculline. This GABAA antagonist, however, failed to antagonize the anti-escape effect caused by the 5-HT1A receptor agonist 8-OH-DPAT. The inhibitory effect caused by DOI, RO 60-0175, and mCPP was also blocked by previous intra-dPAG injection of the preferential 5-HT2A receptor antagonist ketanserin. Pre-administration of the 5-HT2C receptor antagonist SB-242084 in the dPAG did not block the anti-escape effect of RO 60-0175. CONCLUSIONS: Stimulation of 5-HT2A but not 5-HT2C receptors in the dPAG causes a panicolytic-like effect that is mediated by facilitation of GABAergic neurotransmission.