SSRIs and conditioned fear

Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 1810–1819

Contents lists available at SciVerse ScienceDirect

Progress in Neuro-Psychopharmacology & Biological Psychiatry journal homepage: www.elsevier.com/locate/pnp

Review article

SSRIs and conditioned fear Takeshi Inoue ⁎, Yuji Kitaichi, Tsukasa Koyama Department of Psychiatry, Hokkaido University Graduate School of Medicine, North 15, West 7, Kita-ku, Sapporo 060-8638, Japan

a r t i c l e

i n f o

Article history: Received 5 February 2011 Received in revised form 27 August 2011 Accepted 2 September 2011 Available online 8 September 2011 Keywords: Amygdala Conditioned freezing Contextual conditioned fear Hippocampus Selective serotonin reuptake inhibitor Serotonin (5-HT)

a b s t r a c t Among drugs that act on serotonergic neurotransmission, selective serotonin (5-HT) reuptake inhibitors (SSRIs) are now the gold standard for the treatment of anxiety disorders. The precise mechanisms of the anxiolytic actions of SSRIs are unclear. We reviewed the literature related to the effects of SSRIs and the neurochemical changes of 5-HT in conditioned fear. Acute SSRIs and 5-HT1A receptor agonists reduced the acquisition and expression of contextual conditioned fear. Chronic SSRI administration enhanced anxiolyticlike effects. Microinjection studies revealed the amygdala as the target brain region of both classes of serotonergic drugs, and the hippocampus as the target of 5-HT1A receptor agonists. These findings highlight the contribution of post-synaptic 5-HT receptors, especially 5-HT1A receptors, to the anxiolytic-like effects of serotonergic drugs. These results support the new 5-HT hypothesis of fear/anxiety: the facilitation of 5-HT neurotransmission ameliorates fear/anxiety. Furthermore, these behavioral data provide a new explanation of neurochemical adaptations to contextual conditioned fear: increased 5-HT transmission seems to decrease, not increase, fear. © 2011 Elsevier Inc. All rights reserved.

Contents 1. 2. 3. 4. 5.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Neural circuitry of conditioned fear and 5-HT innervation of the amygdala . . . . . . . . . . . Serotonergic neurotransmission during conditioned fear . . . . . . . . . . . . . . . . . . . Methodological issues of behavioral pharmacology on conditioned fear. . . . . . . . . . . . . Behavioral effects of systemic administration of serotonergic drugs on conditioned freezing . . . 5.1. Benzodiazepines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Selective 5-HT1A receptor agonists . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3. SSRIs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4. Other drugs that influence 5-HT neurotransmission . . . . . . . . . . . . . . . . . . 6. Interaction between SSRIs and other psychotropic drugs: implications for augmentation therapy. 7. Serotonergic drugs and fear-potentiated startle . . . . . . . . . . . . . . . . . . . . . . . . 8. Microinjection studies of SSRIs and conditioned fear . . . . . . . . . . . . . . . . . . . . . 9. Target brain sites of the effect of SSRI on conditioned fear: c-Fos study . . . . . . . . . . . . . 10. Summary and outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix A. Three processes of conditioned fear in relation to behavioral pharmacology . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Abbreviations: GAD, generalized anxiety disorder; 5-HT, 5-hydroxytryptamine; ITC, intercalated cell cluster; MAOI, monoamine oxidase inhibitor; NA, noradrenaline; OCD, obsessive–compulsive disorder; PD, panic disorder; PTSD, post-traumatic stress disorder; SAD, social anxiety disorder; SNRI, serotonin–noradrenaline reuptake inhibitor; SSRI, selective serotonin reuptake inhibitor; TCA, tricyclic antidepressant. ⁎ Corresponding author at: Department of Psychiary, Hokkaiido University Graduate School of Medicine, North 15, West 7, Sapporo 060-8638, Japan. Tel.: +81 11 706 5160; fax: +81 11 706 5081. E-mail address: [email protected] (T. Inoue). 0278-5846/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.pnpbp.2011.09.002

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

1810 1811 1811 1812 1813 1813 1813 1815 1815 1815 1816 1816 1817 1818 1818 1818

1. Introduction Conditioned fear is based on Pavlovian aversive conditioning. A neutral stimulus (a tone, figure, light, or context) is presented with an aversive stimulus such as pain or unpleasant sounds. Repeated pairings make a neutral stimulus more aversive, representing the acquisition of conditioned fear. After such acquisition, the presentation of a neutral stimulus alone provokes the expression of conditioned fear,

T. Inoue et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 1810–1819

but repeated presentation without the aversive stimulus diminishes the effects, signifying the extinction of conditioned fear. Since the 1990s, several studies have elucidated the neural circuitry, neurotransmitters, and transcription factors involved in conditioned fear. A particularly impressive experiment by Watson and Rayner (1920) demonstrated how Little Albert was made to fear a white rat paired with a loud sound, and this conditioned fear was generalized to several objects. Conditioned fear is the theoretical background of an experimental neurosis. The extinction process has been applied to the treatment of neurosis (now designated as anxiety disorder) as systematic desensitization since the 1950s by Wolpe (1969). The psychopharmacology of anxiety disorders has advanced considerably since the 1980s; selective serotonin reuptake inhibitors (SSRIs), monoamine oxidase inhibitors (MAOIs), and serotonin (5HT) 1A agonists are now commonly prescribed (Bandelow et al., 2008; Erikkson and Humble, 1990; Zohar and Westenberg, 2000) (Table 1). SSRIs have the broadest indications for anxiety disorders of various types (Bandelow et al., 2008). Because they do not produce dependency, cognitive impairment, or sedation like their predecessors the benzodiazepines, SSRIs are now the first-line drugs for the treatment of anxiety disorders (Bandelow et al., 2008). However, the anxiolytic mechanism of SSRIs has not been elucidated until recently. The classic hypothesis of 5-HT function in anxiety was proposed in the 1970s and posited that 5-HT systems promote anxiety and that suppression of these systems diminishes anxiety (Handley and McBlane, 1993; Traber and Glaser, 1987). However, new evidence suggests that this hypothesis is apparently contrary to the pharmacological action of SSRIs. Since the 1990s, there have been several reports that acute treatment with SSRIs or 5-HT1A agonists is anxiolytic in a conditioned fear model, although some inconsistencies exist among studies (Borsini et al., 2002; Inoue et al., 2000). It is extremely important to elucidate the mechanism of anxiolytic action because SSRIs were ineffective or anxiogenic in other animal models of anxiety (Borsini et al., 2002). Conditioned fear in rats or mice has an additional advantage in that the pharmacological action on fear can be understood in relation to neurochemical effects, which can be easily measured in this model (Inoue et al., 2000). Here, we review the behavioral pharmacology of SSRIs and other serotonergic drugs in conditioned fear and the 5-HT-related neurochemistry of conditioned fear. 2. Neural circuitry of conditioned fear and 5-HT innervation of the amygdala Damage to various brain regions interferes with the acquisition and expression of conditioned fear. Several lesion studies have clarified the roles of specific brain regions. As reviewed by LeDoux Table 1 Summary of clinical controlled studies of anxiety disorders (Bandelow et al., 2008; Inoue et al., 2000).

Benzodiazepine SSRI SNRI TCA 5-HT1A agonist 5-HT2 antagonist 5-HTP MAOI β-blocker NA reuptake inhibitor

PD

GAD

SAD

+ + + + 0 0 + + +/0 0

+ + + + + +

+ + +

PTSD

OCD

+ + +

+ +

1811

(2000) and Davis (2002), the amygdala is critically involved in conditioned fear. Sensory inputs (e.g., tone as a conditioned stimulus) to the amygdala terminate mainly in the lateral nucleus (Fig. 1). Conditioning to the apparatus and other contextual cues that are present when the conditioned stimulus and unconditioned stimulus are paired, involves the representation of the context by the hippocampus, and communication between the basal and accessory basal nuclei of the amygdala. The central nucleus receives inputs from the lateral, basal, and accessory basal nuclei. In turn, it mediates the expression of conditioned fear responses elicited by both acoustic and contextual conditioned stimuli (Fig. 1) (LeDoux, 2000). The lateral, basal, and accessory basal nuclei are collectively designated as the basolateral amygdala. Moreover, the lateral division of the bed nucleus of the stria terminalis, which forms part of the lateral extended amygdala, receives projections from the basolateral amygdala and has direct projections to various anatomic areas that are likely involved in many symptoms of fear or anxiety. In fact, many effects previously attributed to the central nucleus might actually depend on the bed nucleus of the stria terminalis (Davis, 2002). The expression of conditioned fear is mediated by the neural circuitry shown in Fig. 1 (Wilensky et al., 2006). However, repeated exposure (expression) to a conditioned stimulus (cue or context) causes extinction of conditioned fear. The ventromedial prefrontal cortex plays a critical role in this process by suppressing activity in the amygdala through inhibition of the lateral/basal nucleus neurons and/or activation of the inhibitory intercalated cell cluster (ITC) (Fig. 2) (Ehrlich et al., 2009; Izumi et al., 2011; Sotres-Bayon et al., 2006). In the expression of conditioned fear, glutamatergic and GABAergic neurons in the basal nucleus of the amygdala, as well as ITC GABAergic neurons, are activated during the retrieval of conditioned fear (Izumi et al., 2011). Anatomical knowledge of the serotonergic system in the amygdala is important for understanding the effects of SSRIs on anxiety. The amygdala receives dense serotonergic innervation from the dorsal raphe nucleus (Fallon and Ciofi, 2000; Lowry et al., 2005) and contains several subtypes of 5-HT receptors (Radja et al., 1991; Wright et al., 1995). Whereas 5-HT2 and 5-HT3 agonists significantly increased the neuronal discharge rate in nearly all subdivisions of the amygdala, including the basal nucleus of the amygdala, a 5-HT1A agonist significantly inhibited the firing rate (Stein et al., 2000). The modulatory effect of 5HT on anxiety-related circuits has been noted. Although serotonergic drug microinjections or selective 5-HT lesions to the amygdala have been performed in several studies using various animal models, these showed conflicting results (anxiolytic vs. anxiogenic) (Lowry et al., 2005). 5-HT in the amygdala may play different roles in various animal models of anxiety (Borsini et al., 2002), suggesting that it is inappropriate to compare data between different animal models. The amygdala is innervated by the locus coeruleus, which synthesizes the majority of the brain's noradrenaline (Mueller and Cahill, 2010). Noradrenergic neurotransmission is a possible therapeutic target of some SSRIs (paroxetine and fluoxetine) (Beyer et al., 2002; Bymaster et al., 2002). The amygdala contains α1, α2, β1, and β2 adrenoceptors (Johnson et al., 1989; Unnerstall et al., 1985), and noradrenaline enhances intrinsic excitability in the amygdala. Noradrenaline strengthens the formation, consolidation and reconsolidation of emotional memory and the consolidation of extinction memory partly via β-receptor activation of the basal amygdala (Mueller and Cahill, 2010). 3. Serotonergic neurotransmission during conditioned fear

+ +/0

PD, panic disorder; GAD, generalized anxiety disorder; SAD, social anxiety disorder; OCD, obsessive–compulsive disorder; PTSD, post-traumatic stress disorder; SNRI, serotonin–noradrenaline reuptake inhibitor; TCA, tricyclic antidepressant; MAOI, monoamine oxidase inhibitor; 5-HT, serotonin; NA, noradrenaline. +, anxiolytic-like effect; 0, inactive.

The selective activation of dopamine metabolism by contextual conditioned fear in the antero-medial frontal cortex was first described over thirty years ago (Herman et al., 1982); fourteen years later, an in vivo microdialysis study revealed that the increased metabolism reflects the increased release of dopamine (Yoshioka et al., 1996). However, this regional selectivity is dependent on conditioning intensity; conditioned

1812

T. Inoue et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 1810–1819

CS (cue, tone)

CS (context)

Medial Geniculate Nucleus

Auditory Cortex

Hippocampus

Basal Nucleus Lateral Nucleus Accessory Basal Nucleus

Amygdala Central Nucleus

Periaqueductal Gray Matter

Lateral Hypothalamus

Paraventricular Nucleus of the Hypothalamus

Parabrachial Nucleus

Freezing

Increased Blood Pressure & Heart Rate

Increased ACTH

Increased Breathing

Fig. 1. Neural circuitry of the acquisition of conditioned fear.

fear after repeated conditioning by footshock, which is more intense in stress intensity than that after single conditioning, increases dopamine metabolism in more widespread regions of the brain (Inoue et al., 1994). Similarly, selective activation of 5-HT metabolism by contextual conditioned fear in the medial prefrontal cortex has been reported (Inoue et al., 1993); this 5-HT activation reflects an increased release of 5-HT (Table 2) (Hashimoto et al., 1999; Yoshioka et al., 1995). Like dopamine, this regional selectivity also depends on the intensity; repeated conditioning increases 5-HT metabolism to more areas, including the medial prefrontal cortex, nucleus accumbens, and amygdala (Table 2) (Inoue et al., 1994). Increases in extracellular 5-HT levels during contextual conditioned fear were reported in the medial prefrontal cortex after single and repeated contextual conditioning (Hashimoto et al., 1999; Yoshioka et al., 1995) and amygdala after repeated contextual conditioning (Kawahara et al., 1995) in a few in vivo microdialysis studies, but such increases have not been observed in other brain regions (Table 2). In line with the results of 5-HT metabolism studies, repeated conditioning was necessary to enhance extracellular 5-HT levels in the amygdala during contextual conditioned fear (Kawahara et al.,

4. Methodological issues of behavioral pharmacology on conditioned fear Conditioned fear produces various behavioral, endocrinological, and physiological changes in animals (Davis, 2002; LeDoux, 2000).

Ventromedial Prefrontal Cortex

CS (cue, tone)

Medial Geniculate Nucleus

1995). Tone-elicited conditioned fear also increased extracellular 5HT levels in the amygdala (Yokoyama et al., 2005). The simultaneous monitoring of extracellular 5-HT levels and freezing behavior induced by contextual conditioned fear showed that the 5-HT levels of the conditioned fear stress group did not increase when the animals exhibited freezing behavior; rather, their 5-HT levels increased when freezing was resolved (Hashimoto et al., 1999). Another group reported a similar observation: extracellular 5-HT levels increased gradually during conditioned fear and reached maximal levels after the anxiety-related behavior was ceased (Yoshioka et al., 1995). These findings suggest that increased 5-HT ameliorates fear or anxiety, which is contradictory to the classic hypothesis of 5-HT function. Additional studies using drug microinjection are necessary, as discussed in a later section.

CS (context)

Auditory Cortex

LN

Hippocampus

mITC

BN ABN

Amygdala Central Nucleus

Various Fear Responses (see Figure 1) LN, Lateral Nucleus; BN, Basal Nucleus; ABN, Accessory Basal Nucleus; mITC, medial intercalated cell cluster; dotted line, inhibitory; solid line, stimulatory Fig. 2. Neural circuitry of the extinction of conditioned fear.

T. Inoue et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 1810–1819

1813

Table 2 5-HT metabolism and release by contextual conditioned fear. Region

Metabolism

Reference

sCFS

rCFS

Medial prefrontal cortex Nucleus accumbens Striatum Paraventricular nucleus of the hypothalamus Amygdala Lateral hypothalamus Hippocampus

↑ → → → → → →

↑ (×10) ↑ (×10) → (× 10) → (× 10) ↑ (×10) → (× 10) → (× 10)

Medial prefrontal cortex Amygdala

Release revealed by in vivo microdialysis ↑ ↑ (×2) ↑ (elicited by tone as a cue) ↑ (×3)

Inoue Inoue Inoue Inoue Inoue Inoue Inoue

et et et et et et et

al. (1994) al. (1994) al. (1994) al. (1994) al. (1994) al. (1994) al. (1994)

Hashimoto et al. (1999); Yoshioka et al. (1995) Kawahara et al. (1995); Yokoyama et al. (2005)

sCFS, conditioned fear stress after a single conditioning session; rCFS, conditioned fear stress after repeated conditioning sessions (× times of conditioning); ↑, increased; →, unchanged.

However, freezing behavior is a more sensitive index of the stress intensity than either defecation or plasma corticosterone levels as it increases in proportion to shock intensity and the number of conditioning sessions (Fanselow, 1980; Inoue et al., 1994). Freezing behavior is defined as the absence of all observable movements of the skeleton and the vibrissae, except those related to respiration; it is elicited by conditioned stimuli, cue (e.g., a tone) or context (e.g., a shock box). Consequently, freezing behavior has been used in numerous studies to measure the pharmacological effects of various drugs. Most of these studies have examined contextual conditioned freezing; very few studies have examined cue (tone)-elicited conditioned freezing. Another sensitive parameter of conditioned fear is fear-potentiated startle; the amplitude of the acoustic startle reflex in the rat can be augmented by presenting the eliciting auditory startle stimulus in the presence of a cue (e.g., a light or tone) that has been paired with a shock (Davis et al., 1993). The rat is pre-conditioned with either an auditory or visual conditioned stimulus, and then the startle reflex is elicited by either a loud sound or an air-puff. The magnitude of fear-potentiated startle correlates highly with the amount of freezing measured in the same experimental situation (Davis et al., 1993). Fear-potentiated startle and freezing are conditioned responses that are elicited by a cue or context and are similar in several ways. Nevertheless, they should be distinguished in pharmacological terms, and are discussed in section 7. As explained in the Introduction, conditioned fear consists of three processes (Appendix A): First, the acquisition process is the pairing of a neutral stimulus and an aversive stimulus. Drug administration immediately preceding this process affects the formation of conditioned learning, fear memory, and pain when a footshock is used as an unconditioned stimulus. The second process is the expression of conditioned fear that has been acquired, for instance during exposure to a conditioned stimulus without an unconditioned stimulus. The experience causes various fear-related behaviors, including freezing and physiological signs. Drug administration immediately before this process affects fear and anxiety expression and motor activity, the latter of which should be differentiated from the former. The third process is the extinction process that takes place when repeated exposure to a conditioned stimulus occurs without an unconditioned stimulus. It does not involve forgetting or memory erasure but instead involves new learning that inhibits or overrides past learning (Sotres-Bayon et al., 2006). For that reason, drug administration during this process affects new learning. From a clinical perspective, an anxiolytic effect of a drug is evaluated by its effect on all three conditioned fear processes. Although a drug may be used most commonly during expression, its effect on the acquisition process is also important in considering the possibility of the clinical efficacy of a drug for preventing a vicious cycle of anticipatory anxiety leading to fearful cognition and anxiety symptoms in the feared situations. Finally, the effect

on the extinction process might be useful as treatment for resolution of situationally bound anxiety or fear. 5. Behavioral effects of systemic administration of serotonergic drugs on conditioned freezing Serotonergic drugs, such as selective 5-HT reuptake inhibitors, 5HT1A agonists and 5-HT antagonists, have been examined in terms of whether their systemic administration increases or decreases conditioned freezing in animal experiments (Table 3). Many factors can influence these effects: (1) the timing of drug administration, either before acquisition or expression; (2) the kind of conditioned stimulus, such as a cue (e.g., a tone) or context; (3) intervals between acquisition by an unconditioned stimulus (e.g., footshock) and expression; and (4) acute or repeated treatment. In addition, the species or strains used in studies might affect results. 5.1. Benzodiazepines As Table 3 shows, acute systemic treatment with classic anxiolytic drugs (benzodiazepines) inhibited both the acquisition and expression of contextual conditioned freezing without affecting pain sensitivity (Fanselow and Helmstetter, 1988) or motor activity (Miyamoto et al., 2000). State-dependent learning was not related to the effects of benzodiazepines on contextual conditioned freezing (Fanselow and Helmstetter, 1988). This observation is consistent with the results described in conditioned suppression of motility, which is another conditioned fear assessment method (Kitaichi et al., 1995). 5.2. Selective 5-HT1A receptor agonists Acute systemic treatment with selective 5-HT1A receptor agonists consistently reduced the expression of contextual conditioned freezing, irrespective of the specific type of agonist. In several studies, these effects on the acquisition of contextual conditioned freezing are also described (Table 3). The 5-HT1A receptor agonists did not affect motor activity (Inoue et al., 1996b; Li et al., 2001; Nakamura and Kurasawa, 2001; Nishikawa et al., 2007a) or pain sensitivity (Conti et al., 1990) at the effective dosages. Intervals (1 day vs. 14 days) between footshock (conditioning) and testing (expression) did not influence the effect of 5-HT1A receptor agonists on the expression of contextual conditioned freezing (Nishikawa et al., 2007a). Repeated treatment with 5-HT1A receptor agonists did not change the acute effect (Li et al., 2001). No systemic effect of a 5-HT1A receptor agonist on cue-elicited conditioned freezing has been reported, but the microinjection of a 5-HT1A receptor agonist into the median raphe nucleus inhibited the acquisition of light-elicited, but not tone-elicited, conditioned freezing (Avanzi et al., 2003).

1814

T. Inoue et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 1810–1819

Table 3 Behavioral effect of serotonergic drugs on conditioned freezing. Drugs

Process

Effective Route of Duration of dosage (mg/ administration treatment kg)

Period between Context footshock and or cue expression

Species Effect (effective dosage, mg/kg)

1.25–5

IP

Acute

Rat

↓ (2.5–5)

Conti et al. (1990)

0.1–1 0.125–1 2

SC IP IP

Acute Acute Acute

0 (immediately) 1 day 1 day 1 day

Context

Diazepam Diazepam Midazolam

Acquisition/ Expression Expression Expression Acquisition

Context Context Context

Rat Mouse Rat

↓ (1) ↓ (0.125) ↓ (2)

Midazolam

Expression

2

IP

Acute

1 day

Context

Rat

↓ (2)

Inoue et al. (1996b) Miyamoto et al. (2000) Fanselow and Helmstetter (1988) Fanselow and Helmstetter (1988)

Acquisition/ Expression Acquisition Expression Expression Expression

1.25–5

IP

Acute

Context

Rat

↓ (2.5)

Conti et al. (1990)

0.1–10 0.1–10 0.1–1 0.03–0.3

SC SC IP SC

Acute Acute Acute Acute

0 (immediately) 1 day 1 day 1 day 1 day

Context Context Context Context

Rat Rat Rat Mouse

↓ (1–10) ↓ (0.5–10) ↓ (1) ↓ (0.3)

Benzodiazepine Diazepam

5-HT1A agonist Buspirone Ipsapirone Ipsapirone Ipsapirone 8-OH-DPAT

References

Flesinoxan Flesinoxan Tandospirone Tandospirone SSRIs Citalopram Citalopram

Expression Expression Expression Expression

0.3–3 0.3 0.3–1 0.3–1

SC SC SC SC

Acute Repeated for 14 days Acute Acute

1 day 14 days 1 day 14 days

Context Context Context Context

Rat Rat Rat Rat

↓ (0.3–3) ↓* ↓ (1) ↓ (1)

Inoue et al. (1996c) Inoue et al. (1996b) Rittenhouse et al. (1992) Nakamura and Kurasawa (2001) Li et al. (2001) Li et al. (2001) Nishikawa et al. (2007a) Nishikawa et al. (2007a)

Acquisition Acquisition

1–10 10

SC IP

Acute Acute

1 day 1 day

Rat Rat

↓ (3–10) ↑ (10)

Inoue et al. (1996a) Burghardt et al. (2004)

Citalopram

Acquisition

10

IP

Repeated for 22 days

1 day

Rat

↓ (10)

Burghardt et al. (2004)

Citalopram Citalopram Citalopram Citalopram Citalopram Citalopram

Expression Expression Expression Expression Expression Expression

0.1–10 1–10 3–30 3–30 10 10

SC SC IP IP SC SC

1 day 1 day 1 day 14 days 1, 3, 7, 11 days 11 days

Rat Rat Rat Rat Rat Rat

Citalopram

Expression

10

IP

Acute Acute Acute Acute Acute Acute or Repeated for 7 days Acute

Context Cue (tone) Cue (tone) Context Context Context Context Context Context

Rat

Escitalopram Escitalopram Fluvoxamine Fluvoxamine Fluvoxamine

Expression Acquisition Expression Expression Expression

0.5–5 0.5–5 3–30 10–60 30

SC SC SC IP IP

Acute Acute Acute Acute Repeated for 14 days

1 day 1 day 1 day 1 day 14 days

Cue (tone) Context Context Context Context Context

↓ (1–10) Inoue et al. (1996b) ↓ (3–10) Hashimoto et al. (1996) ↓ (30) Nishikawa et al. (2007a) 0 Nishikawa et al. (2007a) ↓ (1 day only) Hashimoto et al. (2009) ↓ (10) (repeated Hashimoto et al. (2009) only) ↑ (10) Burghardt et al. (2007)

Fluvoxamine Fluvoxamine Fluvoxamine Paroxetine Paroxetine Fluoxetine

Expression Expression Expression Expression Expression Expression

30–60 30–60 5–20 5–20 5–20 1–10

IP IP IP IP IP IP

Acute Acute Acute Acute Acute Acute

1 day 14 days 1 day 1 day 14 days 1 day

Context Context Context Context Context Context

Rat Rat Mouse Rat Rat Mouse

↓ (1) ↑ (1) ↓ (30) ↓ (60) ↓ (30) (repeated only) ↓ (60) 0 ↓ (10–20) ↓ (10–20) 0 ↓ (10)

Fluoxetine Fluoxetine Fluoxetine Fluoxetine

Expression Expression Acquisition Expression

10–20 7 10 10

IP IP IP IP

Acute or subchronic Repeated for 21 days Repeated for 14 days Acute

1 day 21 days 1 day 1 day

Context Context Context Cue (tone)

Rat Rat Rat Rat

↓ (10–20) 0 ↓ (10) ↑ (10)

Nishikawa et al. (2007a) Nishikawa et al. (2007a) Miyamoto et al. (2000) Nishikawa et al. (2007a) Nishikawa et al. (2007a) Nakamura and Kurasawa (2001) Santos et al. (2006) Spennato et al. (2008) Zhang et al. (2000) Burghardt et al. (2007)

SC

Acute

1 day

Context

Rat

↓ (30)

Hashimoto et al. (1996)

SC

Acute

1 day

Context

Rat

↓ (20)

Inoue et al. (1996b)

IP IP

Acute Acute

1 day 1 day

Context Context

Rat Rat

0 0

Hashimoto et al. (1997) Hashimoto et al. (1997)

SC

Acute

1 day

Context

Rat

0

Muraki et al. (2008)

SC

Acute

1 day

Context

Rat

0

Muraki et al. (2008)

SC

Acute

1 day

Context

Rat

0

Inoue et al. (1996a)

IP

Acute

1 day

Context

Mouse

0

SC SC

Acute Acute

1 day 1 day

Context Context

Rat Rat

0 0

Nakamura and Kurasawa (2001) Inoue et al. (1996b) Inoue et al. (1996b)

SNRI Milnacipran Expression 3–30 5-HT precursor L-5Expression 5–20 with hydroxytryptophan benserazide 5-HT1 antagonist (receptor subtypes) NAN-190 (1A) Expression 0.1–10 WAY100135 Expression 0.1 (1A) WAY100635 Expression 0.15 (1A) GR 127,935 Expression 4 (1B/1D) NAN-190 (1A) Acquisition 1 5-HT2 antagonist (receptor subtypes) Ketanserin (2A) Expression 0.1–1 Ketanserin (2A)

Expression Expression

0.1–5 5–20

1 day

Rat Rat Rat Rat Rat

Montezinho et al. (2010) Montezinho et al. (2010) Hashimoto et al. (1996) Li et al. (2001) Li et al. (2001)

T. Inoue et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 1810–1819

1815

Table 3 (continued) Drugs

ICI169,369 (2A, 2 C) ICI169,369 (2A, 2 C) 5-HT3 antagonist Tropisetron Others Mirtazapine Phenelzine

Process

Effective Route of Duration of dosage (mg/ administration treatment kg)

Period between Context footshock and or cue expression

Species Effect (effective dosage, mg/kg)

References

Acquisition

10

SC

Acute

1 day

Context

Rat

0

Inoue et al. (1996a)

Expression

0.01–0.1

IP

Acute

1 day

Context

Rat

↓(0.01–0.1)

Yoshioka et al. (1995)

Expression Expression

0.3–10 10–60

IP IP

Acute Acute

1 day 1 day

Context Context

Rat Rat

↓ (1–10) ↓ (30–60)

Kakui et al. (2009) Maki et al. (2000)

SC, subcutaneous; IP, intraperitoneal. ↓, decrease freezing; 0, no effect; ↑, increase freezing; *, repeated treatment did not enhance the effect of acute treatment. 1 day only, citalopram treatment only 1 day after conditioning had an inhibitory effect. Repeated only, repeated treatment had an inhibitory effect, but a single treatment did not.

5.3. SSRIs Most reports have described that various SSRIs reduce the expression of contextual conditioned freezing; only a few studies described effects of SSRIs on the acquisition of contextual conditioned freezing (Table 3). Systemic citalopram administration before training and before testing attenuated contextual conditioned freezing. Therefore, state-dependent learning was not related with the effects of SSRIs on contextual conditioned freezing (Hashimoto et al., 2009). In contrast to 5-HT1A receptor agonists, prolonging the interval between conditioning by footshock and exposure to conditioned fear stress diminishes the inhibitory effects of SSRIs of various kinds on the expression of contextual conditioned freezing (Hashimoto et al., 2009; Li et al., 2001; Nishikawa et al., 2007a). Repeated treatments with SSRIs restored their efficacy (Hashimoto et al., 2009) and further enhanced the effect of subeffective-dosage SSRIs, but such repeated treatments might not enhance the maximal effect of SSRIs (Li et al., 2001). Clinically, SSRIs exert an anxiolytic effect after chronic treatment (Bandelow et al., 2008). Therefore, the feasibility of establishing an animal model that has more precise predictive and face validities for anxiety disorder is suggested by prolonging the period between conditioning and exposure to conditioned fear stress. In other words, the long interval between conditioning by footshock and exposure to conditioned fear stress may improve predictive and face validities of contextual conditioned freezing as an animal model of anxiety disorders. However, this mechanism has not yet been elucidated. We suggest the possibility of both presynaptic and postsynaptic mechanisms: the increase of extracellular 5-HT or a marked reduction in the firing activity of 5-HT neurons in the raphe nucleus induced by SSRIs (Chaput et al., 1986) might be changed by prolonging the period between conditioning and exposure to conditioned fear stress. Alternatively, the responsiveness of postsynaptic 5-HT receptors might be modified. Results of earlier behavioral studies show long-lasting behavioral effects following a short session of inescapable footshock stress (Van Dijken et al., 1992) and reinforce the concept that recent fear memory after conditioning is less stable than remote fear memory (Sacchetti et al., 1999). In contrast to the effect on contextual conditioned freezing, acute SSRIs increased cue (tone)-elicited conditioned freezing in both the acquisition and expression of conditioned fear (Burghardt et al., 2004, 2007). An anxiogenic effect of SSRI was blocked by a 5-HT2C receptor antagonist, suggesting that SSRIs increase conditioned fear by stimulating 5-HT2C receptors (Burghardt et al., 2007). Chronic SSRI administration reduced the acquisition of cue (tone)-elicited conditioned freezing (Burghardt et al., 2004), but no effect on the expression has been reported for it. The reason for the discrepancy of SSRIs effects on contextual conditioned freezing versus cue (tone)-elicited conditioned freezing remains unknown. The effects of chronic SSRI administration on the acquisition of conditioned freezing were consistent between contextual and cue-elicited conditioned freezing: both

were reduced (Burghardt et al., 2004; Inoue et al., 1998). The neural circuitries underlying the two types of conditioning are not identical (LeDoux, 2000), and SSRIs might affect them differently. An acute worsening of conditioned fear might be a model of clinical initial worsening with SSRIs in some patients (Burghardt et al., 2007), but generally speaking, there is no evidence of SSRIs initially exacerbating anxiety (Zohar and Westenberg, 2000). 5.4. Other drugs that influence 5-HT neurotransmission Effects of other psychotropic drugs that affect 5-HT neurotransmission in contextual conditioned freezing have been investigated. Milnacipran (a serotonin–noradrenaline reuptake inhibitor) and 5hydroxytryptophan (a 5-HT precursor plus benserazide) reduce the expression of contextual conditioned freezing (Table 3). Phenelzine (a non-selective MAOI) and mirtazapine (a noradrenergic and specific serotonergic antidepressant) also reduce it (Table 3). The inhibitory effect of mirtazapine on contextual conditioned freezing was attenuated by the 5-HT1A antagonist WAY100635, indicating that mirtazapine's effect is in part mediated by 5-HT1A stimulation by increased 5HT release (Kakui et al., 2009). Although these drugs exert other actions on monoaminergic transmission, they also increase synaptic 5HT levels in the brain, which is likely to be the mechanism of action for decreased conditioned fear. Most antagonists for 5-HT1A, 5-HT1B/1D, and 5-HT2 receptor subtypes failed to change the acquisition or expression of contextual conditioned freezing, except for tropisetron, a 5-HT3 antagonist that reduced the expression of contextual conditioned freezing (Table 3). Interaction of these antagonists with SSRIs is described in the next section. In summary, the effects of selective 5-HT1A agonists, SSRIs, and other serotonergic drugs suggest that increased 5-HT neurotransmission reduces the acquisition and expression of contextual conditioned freezing. These observations are consistent with the clinical hypothesis for anxiety disorders that facilitation of 5-HT neurotransmission prevents anxiety (Erikkson and Humble, 1990). 6. Interaction between SSRIs and other psychotropic drugs: implications for augmentation therapy Treatment-resistant anxiety disorders have not been studied in depth. Switching from an SSRI to other classes of anxiolytics is not unusual (Bandelow et al., 2008). However, not all patients with anxiety disorders completely recover when these pharmacological strategies are employed. Using the contextual conditioned freezing model, novel therapies have been sought, as shown in Table 4. As expected, 5-HT1A receptor (autoreceptor) antagonists enhance the effect of low-dosage SSRIs on the expression of contextual

1816

T. Inoue et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 1810–1819

conditioned freezing; although the blockade of 5-HT1B/1D receptor, another autoreceptor of 5-HT neurons, did not affect it or did not augment the effect of the co-administered 5-HT1A antagonist (Muraki et al., 2008). Earlier in vivo microdialysis studies reported conflicting data (Gobert et al., 1997; Sharp et al., 1997). GR 127,935, a 5HT1B/1D receptor antagonist, enhanced SSRI-induced extracellular 5HT to a level similar to that elicited by WAY 100,635, a 5-HT1A receptor antagonist (Gobert et al., 1997). However, negative data was also reported (Sharp et al., 1997). The combination of GR 127,935 and WAY 100,635 produced a greater increase of SSRI-induced extracellular 5-HT levels than WAY 100,635 did in naïve rats (Gobert et al., 1997; Sharp et al., 1997). However, these earlier in vivo microdialysis data do not fully explain the behavioral data on contextual conditioned freezing. Future studies must be undertaken to examine the effects of these drugs on extracellular 5-HT levels in the amygdala or medial prefrontal cortex in rats exposed to conditioned fear, instead of using only naïve rats. The interactions between SSRIs and many other subtypes of 5-HT receptors and the impact of these interactions on contextual conditioned freezing have not been reported. Co-administration of a subeffective or effective 5-HT1A receptor agonist enhanced the SSRI-induced inhibition of the expression of contextual conditioned freezing (Li et al., 2001; Nishikawa et al., 2007a). Chronic pretreatment with a 5-HT1A receptor agonist also enhanced expression (Li et al., 2001). Conversely, chronic pretreatment with an SSRI enhanced the inhibitory effect of a 5-HT1A receptor agonist (Li et al., 2001). The mechanism underlying this cross-sensitization-like effect might be due in part to the desensitization of presynaptic 5-HT1A receptor following chronic treatment with an SSRI and a 5-HT1A receptor agonist. Clinically, lithium augmentation of antidepressants is an important treatment strategy for refractory major depression (Bauer and Döpfmer, 1999). Subchronic lithium treatment enhanced the inhibitory effects of low-dose and high-dose citalopram (an SSRI), MKC-242 (a selective 5HT1A receptor agonist) and clorgyline (an MAOI) on the expression of contextual conditioned freezing (Kitaichi et al., 2006; Muraki et al., 1999). Subchronic lithium treatment increases extracellular 5-HT levels in the brain. This might provide an additive anxiolytic-like effect to serotonergic drugs (Kitaichi et al., 2006; Muraki et al., 2001). These findings suggest that adding lithium to an existing drug regimen may be a viable option for treatment-resistant anxiety disorders. Reboxetine and atomoxetine, selective noradrenaline reuptake inhibitors, interfere with the anxiolytic-like effects of SSRIs (Inoue et al., 2006; Montezinho et al., 2010). Reboxetine has the additional ability of exerting an anxiogenic effect on the expression of contextual conditioned freezing (Inoue et al., 2006). Moreover, pretreatment with the benzodiazepine anxiolytic drug diazepam unexpectedly suppressed the anxiolytic-like effect of an SSRI on the expression of contextual conditioned freezing (Miyamoto et al., 2000). Clinical observations are necessary to confirm these findings in animal models. In summary, co-administration of acute 5-HT1A receptor antagonist, chronic 5-HT1A receptor agonist, or subchronic lithium enhances the acute anxiolytic-like effect of an SSRI on the expression of conditioned fear, presumably via the enhanced availability of presynaptic 5-HT. In contrast, treatment with an acute noradrenaline reuptake inhibitor cancels the acute anxiolytic-like effect of an SSRI on the expression of conditioned fear via its direct anxiogenic effect. 7. Serotonergic drugs and fear-potentiated startle The fear-potentiated startle paradigm has been extensively utilized by Davis and colleagues to elucidate the neural and pharmacological mechanisms involved in conditioned fear (for review, see Davis et al., 1993). Two classes of typical anxiolytic drugs, selective 5-HT1A receptor agonists and benzodiazepines, decrease fear-potentiated startle without altering baseline levels of startle, suggesting an anxiolytic-like action (Davis et al., 1993; Joordens et al., 1996). Neither imipramine, which inhibits 5-HT and noradrenaline reuptake,

or SSRIs affect fear-potentiated startle in animals (Cassella and Davis, 1985; Joordens et al., 1996). 5-HT3, but not 5-HT2, receptor antagonists decrease fear-potentiated startle (Davis et al., 1993). Thus, serotonergic drugs have different effects on fear-potentiated startle and contextual conditioned freezing. It is noteworthy that the former is cue (light)-elicited conditioned fear, but the latter is a contextual one, thereby indicating that neural circuitries involved in these two paradigms might differ. A great advantage of fear-potentiated startle over contextual conditioned freezing is that it can be measured reliably in humans when the eyeblink component of startle is elicited at a time when a person is anticipating a shock (Davis et al., 1993). Therefore, findings from animals can be confirmed clinically. Acute treatment with citalopram, an SSRI, increased phasic fear-potentiated startle and sustained potentiated startle in healthy humans, which correspond to cue (color)-elicited and contextual fear, respectively (Grillon et al., 2007). Chronic treatment with citalopram reduced contextual anxiety, which, in this study, was anticipatory anxiety-potentiated startle (Grillon et al., 2009). Fear-potentiated startle was also investigated in juvenile rhesus monkeys. The results of that study showed that oral diazepam, but not oral buspirone, reduced fear-potentiated startle (Winslow et al., 2007). In addition to species differences, pharmacokinetic differences can account for the discrepancies between monkeys and rats. Subcutaneous buspirone reduced fear-potentiated startle because it is reportedly metabolized slower in contrast to oral administration (Nishikawa et al., 2007b). 8. Microinjection studies of SSRIs and conditioned fear As discussed in earlier sections, 5-HT neurotransmission is apparently increased during contextual conditioned fear. Stimulation of 5-HT neurotransmission reduces conditioned fear when freezing is used as an index of fear. This parallelism might prove the anxiolytic role of 5HT, but two opposing hypotheses have been suggested: (1) 5-HT1A receptor agonists inhibit the raphe nucleus; this inhibition of 5-HT neurotransmission might reduce fear or anxiety; and (2) the clinical anxiolytic effects of SSRIs necessitate repeated treatment; receptor desensitization might be involved in the anxiolytic effect. To clarify the SSRI mechanism of action on contextual conditioned fear, microinjections of SSRIs into the putative target brain regions should be performed to verify the brain regions in which SSRIs act as anxiolytics. Our study showed that a single microinjection of the SSRI citalopram into the basolateral amygdala reduced the expression of contextual conditioned freezing, but injection into the medial prefrontal cortex or mediodorsal nucleus of the thalamus did not affect freezing behavior (Inoue et al., 2004). This finding suggests that the anxiolyticlike effect of an SSRI in contextual conditioned freezing is mediated by increased 5-HT in the amygdala. Furthermore, we reported that a single microinjection of the selective 5-HT1A receptor agonist flesinoxan into the basolateral amygdala and hippocampus, but not into the medial prefrontal cortex, reduced the expression of contextual conditioned freezing (Li et al., 2006). Therefore, although this study does not completely disprove the possibility of the contribution of presynaptic 5-HT1A receptors, we suggest that flesinoxan exerts its anxiolytic-like actions during fear conditioning through stimulation of the postsynaptic 5-HT1A receptors in the hippocampus and amygdala. The direct demonstration by drug microinjection provides more evidence to support the hypothesis that increased 5-HT in the terminal areas ameliorates fear or anxiety. A major limitation of regional specificity should be noted; although the injection cannula was inserted into the basolateral amygdala in the studies described above, intracerebral drug injection in a 0.5 μl volume can diffuse one or more millimeters from the infusion site (Wise and Hoffman, 1992). Therefore, it is possible that the drugs diffused into the adjacent brain regions, such as the central nucleus of the amygdala. Accordingly, the results of microinjection

T. Inoue et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 1810–1819

1817

Table 4 Interaction between serotonergic drugs and possible augmentation therapy on contextual conditioned freezing. Pre- or co-treated drugs

Process

Period between Acute challenge Species Effect of pretreated References footshock and drugs (dose, mg/ drugs on challenge expression kg) drugsa

Dosage (mg/ Route of Duration of kg) administration treatment

5-HT receptor antagonists (receptor subtype) ICI169,369 (2A, 2 C) Acquisition 10b

SC

Acute

1 day

Citalopram (10c) Rat

b

SC

Acute

1 day

Citalopram (10 ) Rat

SC

Acute

1 day

SC

Acute

1 day

SC

Acute

1 day

Citalopram (3d, 30c) Citalopram (3d, 30c) Citalopram (3d)

NAN-190 (1A)

Acquisition 1

GR 127,935

Expression

4b

WAY 100,635

Expression

0.15b b

b

GR 127,935 + WAY 100,635 NAN-190 (1A)

Expression

4 + 0.15

Expression

0.1–0.25

b

(+)WAY100135 (1A)

Expression

c

Rat Rat Rat

d

0

Inoue et al. (1996a) 0 Inoue et al. (1996a) 0 Muraki et al. (2008) Enhance (3), 0 (30) Muraki et al. (2008) Enhance (3) Muraki et al. = WAY alone (2008) Enhance (10) Hashimoto et al. (1997) Enhance (1) Hashimoto et al. (1997)

IP

Acute

1 day

Citalopram (10 ) Rat

0.1b

IP

Acute

1 day

Citalopram (1d)

Rat

Expression

0.3d

SC

Enhance (30)

Expression

0.3

SC

Fluvoxamine (30d) Fluvoxamine (30d)

Rat

Tandospirone

Repeated for 13 days Acute

14 days

d

Rat

Enhance (30)

Tandospirone

Expression

0.3d

SC

Acute

14 days

Citalopram (10d) Rat

Enhance (10)

Tandospirone

Expression

0.3d

SC

Acute

14 days

Paroxetine (5d)

Enhance (5)

SSRI Fluvoxamine

Expression

30d

IP

Repeated for 13 days

14 days

Flesinoxan (0.3c) Rat

Enhance (0.3)

Li et al. (2001)

Others Diazepam

Expression

IP 0.125–1 (only 0.25c) 10 SC (anxiogenic) 3b IP

Acute

1 day

Cancel

Acute

1 day

Fluvoxamine Mouse (20c) Citalopram (10c) Rat

Acute

1 day

IP

Acute

1 day

IP

11 and 12 days before test Subchronic for 7 days Subchronic for 7 days Subchronic for 7 days

1 day

Miyamoto et al. (2000) Inoue et al. (2006) Montezinho et al. (2010) Montezinho et al. (2010) Inoue et al. (1996b)

5-HT1A receptor agonists Flesinoxan

Reboxetine (noradrenaline Expression reuptake inhibitor) Atomoxetine (noradrenaline Expression reuptake inhibitor) Atomoxetine Acquisition 3b

b

p-Chloroamphetamine (5-HT toxin)

Expression

10 × 2

Lithium

Expression

Lithium

Expression

0.2%Li2CO3 in dietb 0.2%Li2CO3 in dietb 0.2%Li2CO3 in dietb

Lithium

Expression

O O O

14 days

7 days 7 days 7 days

Rat

Cancel

Li et al. (2001) Nishikawa et al. (2007a) Nishikawa et al. (2007a) Nishikawa et al. (2007a)

Escitalopram (1c) Escitalopram (1, anxiogenic) Ipsapirone (1c)

Rat

Cancel

Rat

Cancel

Rat

0

Citalopram (3d, 30c) MKC-242 (1d)

Rat

d

Rat

Enhance (3 and 30) Muraki et al. (1999) Enhance (1) Muraki et al. (1999) Enhance (10) Kitaichi et al. (2006)

Clorgyline (10 )

Rat

SC, subcutaneous; IP, intraperitoneal; O, oral. Enhance, pre- or co-treated drugs enhanced the inhibitory effect of an acute challenge drug on freezing. Cancel, pre- or co-treated drugs canceled the inhibitory effect of an acute challenge drug on freezing. 0, no effect. = WAY alone; the combination of GR 127,935 and WAY 100,635 did not enhance the effect of WAY 100,635 alone on freezing. a Dosage of challenge drug (mg/kg). b Ineffective drug. c Effective dose. d Subeffective dose.

studies permit no definitive conclusion about the site of action of drugs within the amygdala. Additional experimental approaches will therefore be necessary to draw a firm conclusion about the amygdalar target subnucleus of serotonergic drugs, which is described in the subsequent section. 9. Target brain sites of the effect of SSRI on conditioned fear: c-Fos study The proto-oncogene c-fos is expressed rapidly and transiently in response to increased neuronal activity in the central nervous system following many experimental manipulations. Therefore, it has been

used as a marker of physiological activity (Davis et al., 1993). An early study showed that contextual conditioned fear increased c-fos mRNA levels in the amygdala (Campeau et al., 1991). Immunohistochemical studies have been used to evaluate the expression of c-Fos protein in several discrete brain regions in response to stress and treatment (Beck and Fibiger, 1995). The number of c-Fos-like immunoreactive neurons is analyzed as c-Fos expression. Contextual conditioned fear increased c-Fos expression in several brain regions, but systemic administration of the SSRI citalopram attenuated it in only three brain regions, one of which was the basal nucleus of the amygdala (Izumi et al., 2006). Simultaneously, citalopram produced a dose-dependent reduction in contextual conditioned freezing (Izumi

1818

T. Inoue et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 1810–1819

et al., 2006). The effects of diazepam on c-Fos expression induced by contextual conditioned fear differed from those of citalopram; diazepam did not reduce c-Fos expression elevated by contextual conditioned fear in the basal nucleus of the amygdala. Rather, it was increased in the central nucleus of the amygdala (Beck and Fibiger, 1995), suggesting a mechanistic difference of anxiolytic action between benzodiazepines and SSRIs. More recently, we reported that most c-Fos positive cells induced by conditioned fear in the basal nucleus of the amygdala are glutamatergic, but not GABAergic (Izumi et al., 2011). In conclusion, the basal nucleus of the amygdala is the likely region in which SSRIs exert an anxiolytic-like effect via inhibitory modulation of this brain region. 10. Summary and outlook Research over the past two decades has provided greater insight into the role of 5-HT in the physiology and pharmacology of conditioned fear. However, several points remain to be clarified: the difference between cue-elicited conditioned fear and contextual conditioned fear, the diminishment of an anxiolytic-like effect of SSRI by prolonging the period between conditioning and exposure to conditioned fear, and the nature of the 5-HT that mediate the effect of SSRIs. In addition, the molecular mechanism of conditioned fear and SSRIs in relation to the subnuclei of the amygdala and other brain regions should be investigated more thoroughly. Appendix A. Three processes of conditioned fear in relation to behavioral pharmacology 1. Acquisition: drug administration before fear conditioning, e.g., cue or context with footshock as an unconditioned stimulus (pretraining administration) 2. Expression: drug administration before re-exposure to cue or context without an unconditioned stimulus (pre-testing administration) 3. Extinction: drug administration during repeated exposure sessions to cue or context without an unconditioned stimulus References Avanzi V, Silva RC, Macedo CE, Brandao ML. 5-HT mechanisms of median raphe nucleus in the conditioned freezing caused by light/foot-shock association. Physiol Behav 2003;78:471–7. Bandelow B, Zohar J, Hollander E, Kasper S, Moller HJ, Allgulander C, et al. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for the pharmacological treatment of anxiety, obsessive-compulsive and post-traumatic stress disorders — first revision. World J Biol Psychiatry 2008;9:248–312. Bauer M, Döpfmer S. Lithium augmentation in treatment-resistant depression: metaanalysis of placebo-controlled studies. J Clin Psychopharmacol 1999;19:427–34. Beck CH, Fibiger HC. Conditioned fear-induced changes in behavior and in the expression of the immediate early gene c-fos: with and without diazepam pretreatment. J Neurosci 1995;15:709–20. Beyer CE, Boikess S, Luo B, Dawson LA. Comparison of the effects of antidepressants on norepinephrine and serotonin concentrations in the rat frontal cortex: an in-vivo microdialysis study. J Psychopharmacol 2002;16:297–304. Borsini F, Podhorna J, Marazziti D. Do animal models of anxiety predict anxiolytic-like effects of antidepressants? Psychopharmacology (Berl) 2002;163:121–41. Burghardt NS, Sullivan GM, McEwen BS, Gorman JM, LeDoux JE. The selective serotonin reuptake inhibitor citalopram increases fear after acute treatment but reduces fear with chronic treatment: a comparison with tianeptine. Biol Psychiatry 2004;55: 1171–8. Burghardt NS, Bush DEA, McEwen BS, LeDoux JE. Acute selective serotonin reuptake inhibitors increase conditioned fear expression: blockade with a 5-HT2C receptor antagonist. Biol Psychiatry 2007;62:1111–8. Bymaster FP, Zhang W, Carter PA, Shaw J, Chernet E, Phebus L, et al. Fluoxetine, but not other selective serotonin uptake inhibitors, increases norepinephrine and dopamine extracellular levels in prefrontal cortex. Psychopharmacology (Berl) 2002;160:353–61. Campeau S, Hayward MD, Hope BT, Rosen JB, Nestler EJ, Davis M. Induction of the c-fos proto-oncogene in rat amygdala during unconditioned and conditioned fear. Brain Res 1991;565:349–52. Cassella JV, Davis M. Fear-enhanced acoustic startle is not attenuated by acute or chronic imipramine treatment in rats. Psychopharmacology (Berl) 1985;87: 278–82.

Chaput Y, de Montigny C, Blier P. Effects of a selective 5-HT reuptake blocker, citalopram, on the sensitivity of 5-HT autoreceptors: electrophysiological studies in the rat brain. Naunyn Schmiedebergs Arch Pharmacol 1986;333:342–8. Conti LH, MacIver CR, Ferkany JW, Abreu ME. Footshock-induced freezing behavior in rats as a model for assessing anxiolytics. Psychopharmacology (Berl) 1990;102: 492–7. Davis M. Neural circuitry of anxiety and stress disorders. In: Davis KL, Charney D, Coyle JT, Nemeroff C, editors. Neuropsychopharmacology — 5th generation of progress. Philadelphia: Lippincott Williams & Wilkins; 2002. p. 931–51. Davis M, Falls WA, Campeau S, Kim M. Fear-potentiated startle: a neural and pharmacological analysis. Behav Brain Res 1993;58:175–98. Ehrlich I, Humeau Y, Grenier F, Ciocchi S, Herry C, Luthi A. Amygdala inhibitory circuits and the control of fear memory. Neuron 2009;62:757–71. Erikkson E, Humble M. Serotonin in psychiatric pathophysiology. A review of data from experimental and clinical research. In: Pohl R, Gershon S, editors. Progress in basic clinical pharmacology. Basel: Karger; 1990. p. 66-119. Fallon JH, Ciofi P. Distribution of monoamines within the amygdala. In: Aggleton J, editor. The amygdala: a functional analysis. 2nd ed. Oxford: Oxford University Press; 2000. p. 97-114. Fanselow MS. Conditioned and unconditional components of post-shock freezing. Pavlov J Biol Sci 1980;15:177–82. Fanselow MS, Helmstetter FJ. Conditional analgesia, defensive freezing, and benzodiazepines. Behav Neurosci 1988;102:233–43. Gobert A, Rivet JM, Cistarelli L, Millan MJ. Potentiation of the fluoxetine-induced increase in dialysate levels of serotonin (5-HT) in the frontal cortex of freely moving rats by combined blockade of 5-HT1A and 5-HT1B receptors with WAY 100,635 and GR 127,935. J Neurochem 1997;68:1159–63. Grillon C, Levenson J, Pine DS. A single dose of the selective serotonin reuptake inhibitor citalopram exacerbates anxiety in humans: a fear-potentiated startle study. Neuropsychopharmacology 2007;32:225–31. Grillon C, Chavis C, Covington MF, Pine DS. Two-week treatment with the selective serotonin reuptake inhibitor citalopram reduces contextual anxiety but not cued fear in healthy volunteers: a fear-potentiated startle study. Neuropsychopharmacology 2009;34:964–71. Handley SL, McBlane JW. 5HT drugs in animal models of anxiety. Psychopharmacology (Berl) 1993;112:13–20. Hashimoto S, Inoue T, Koyama T. Serotonin reuptake inhibitors reduce conditioned fear stress-induced freezing behavior in rats. Psychopharmacology (Berl) 1996;123:182–6. Hashimoto S, Inoue T, Koyama T. Effects of the co-administration of 5-HT1A receptor antagonists with an SSRI in conditioned fear stress-induced freezing behavior. Pharmacol Biochem Behav 1997;58:471–5. Hashimoto S, Inoue T, Koyama T. Effects of conditioned fear stress on serotonin neurotransmission and freezing behavior in rats. Eur J Pharmacol 1999;378:23–30. Hashimoto S, Inoue T, Muraki I, Koyama T. Effects of acute citalopram on the expression of conditioned freezing in naive versus chronic citalopram-treated rats. Prog Neuropsychopharmacol Biol Psychiatry 2009;33:113–7. Herman JP, Guillonneau D, Dantzer R, Scatton B, Semerdjian-Rouquier L, Le Moal M. Differential effects of inescapable footshocks and of stimuli previously paired with inescapable footshocks on dopamine turnover in cortical and limbic areas of the rat. Life Sci 1982;30:2207–14. Inoue T, Koyama T, Yamashita I. Effect of conditioned fear stress on serotonin metabolism in the rat brain. Pharmacol Biochem Behav 1993;44:371–4. Inoue T, Tsuchiya K, Koyama T. Regional changes in dopamine and serotonin activation with various intensity of physical and psychological stress in the rat brain. Pharmacol Biochem Behav 1994;49:911–20. Inoue T, Hashimoto S, Tsuchiya K, Izumi T, Ohmori T, Koyama T. Effect of citalopram, a selective serotonin reuptake inhibitor, on the acquisition of conditioned freezing. Eur J Pharmacol 1996a;311:1–6. Inoue T, Tsuchiya K, Koyama T. Serotonergic activation reduces defensive freezing in the conditioned fear paradigm. Pharmacol Biochem Behav 1996b;53:825–31. Inoue T, Tsuchiya K, Koyama T. Effects of typical and atypical antipsychotic drugs on freezing behavior induced by conditioned fear. Pharmacol Biochem Behav 1996c;55:195–201. Inoue T, Hashimoto S, Izumi T, Maki Y, Muraki I, Koyama T. Effect of repeated SSRI on the acquisition of conditioned freezing. 28th annual meeting of Japanese Society of Neuropsychopharmacology (Abstract), Tokyo; 1998. Inoue T, Hashimoto S, Izumi T, Maki Y, Muraki I, Li XB, et al. Serotonergic neurotransmission and defensive freezing in the conditioned fear stress paradigm. In: Hatano G, Okada N, Tanabe H, editors. Affective minds. Amsterdam: Elsevier Science; 2000. p. 91-100. Inoue T, Li XB, Abekawa T, Kitaichi Y, Izumi T, Nakagawa S, et al. Selective serotonin reuptake inhibitor reduces conditioned fear through its effect in the amygdala. Eur J Pharmacol 2004;497:311–6. Inoue T, Nakagawa S, Izumi T, Kitaichi Y, Koyama T. Effect of combined treatment with noradrenaline and serotonin reuptake inhibitors on conditioned freezing. Eur J Pharmacol 2006;540:91–5. Izumi T, Inoue T, Kitaichi Y, Nakagawa S, Koyama T. Target brain sites of the anxiolytic effect of citalopram, a selective serotonin reuptake inhibitor. Eur J Pharmacol 2006;534:129–32. Izumi T, Boku S, Shinmin W, Inoue T, Konno K, Yamaguchi T, et al. Retrieval of conditioned fear activates the basolateral and intercalated nucleus of amygdala. J Neurosci Res 2011;89:773–90. Johnson EW, Wolfe BB, Molinoff PB. Regulation of subtypes of beta-adrenergic receptors in rat brain following treatment with 6-hydroxydopamine. J Neurosci Official J Soc Neurosci 1989;9:2297–305. Joordens RJ, Hijzen TH, Peeters BW, Olivier B. Fear-potentiated startle response is remarkably similar in two laboratories. Psychopharmacology (Berl) 1996;126:104–9.

T. Inoue et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 1810–1819 Kakui N, Yokoyama F, Yamauchi M, Kitamura K, Imanishi T, Inoue T, et al. Anxiolyticlike profile of mirtazapine in rat conditioned fear stress model: functional significance of 5-hydroxytryptamine 1A receptor and alpha1-adrenergic receptor. Pharmacol Biochem Behav 2009;92:393–8. Kawahara H, Kawahara Y, Yoshida M, Yokoo H, Nishi M, Tanaka M. Serotonin release in the rat amygdala during exposure to emotional stress. In: Tanaka A, Curzon G, editors. Serotonin in the central nervous system and periphery. Amsterdam: Elsevier Science; 1995. p. 89–96. Kitaichi K, Minami Y, Amano M, Yamada K, Hasegawa T, Nabeshima T. The attenuation of suppression of motility by triazolam in the conditioned fear stress task is exacerbated by ethanol in mice. Life Sci 1995;57:743–53. Kitaichi Y, Inoue T, Nakagawa S, Izumi T, Koyama T. Effect of co-administration of subchronic lithium pretreatment and acute MAO inhibitors on extracellular monoamine levels and the expression of contextual conditioned fear in rats. Eur J Pharmacol 2006;532:236–45. LeDoux JE. The amygdala and emotion: a view through fear. In: Aggleton J, editor. The amygdala: a functional analysis. 2nd ed. Oxford: Oxford University Press; 2000. p. 289–310. Li XB, Inoue T, Hashimoto S, Koyama T. Effect of chronic administration of flesinoxan and fluvoxamine on freezing behavior induced by conditioned fear. Eur J Pharmacol 2001;425:43–50. Li XB, Inoue T, Abekawa T, Weng S, Nakagawa S, Izumi T, et al. 5-HT1A receptor agonist affects fear conditioning through stimulations of the postsynaptic 5-HT1A receptors in the hippocampus and amygdala. Eur J Pharmacol 2006;532:74–80. Lowry CA, Johnson PL, Hay-Schmidt A, Mikkelsen J, Shekhar A. Modulation of anxiety circuits by serotonergic systems. Stress 2005;8:233–46. Maki Y, Inoue T, Izumi T, Muraki I, Ito K, Kitaichi Y, et al. Monoamine oxidase inhibitors reduce conditioned fear stress-induced freezing behavior in rats. Eur J Pharmacol 2000;406:411–8. Miyamoto J, Tsuji M, Takeda H, Nawa H, Matsumiya T. Pretreatment with diazepam suppresses the reduction in defensive freezing behavior induced by fluvoxamine in the conditioned fear stress paradigm in mice. Eur J Pharmacol 2000;409:81–4. Montezinho LP, Miller S, Plath N, Jensen NH, Karlsson JJ, Witten L, et al. The effects of acute treatment with escitalopram on the different stages of contextual fear conditioning are reversed by atomoxetine. Psychopharmacology (Berl) 2010;212:131–43. Mueller D, Cahill SP. Noradrenergic modulation of extinction learning and exposure therapy. Behav Brain Res 2010;208:1-11. Muraki I, Inoue T, Hashimoto S, Izumi T, Ito K, Ohmori T, et al. Effect of subchronic lithium carbonate treatment on anxiolytic-like effect of citalopram and MKC-242 in conditioned fear stress in the rat. Eur J Pharmacol 1999;383:223–9. Muraki I, Inoue T, Hashimoto S, Izumi T, Ito K, Koyama T. Effect of subchronic lithium treatment on citalopram-induced increases in extracellular concentrations of serotonin in the medial prefrontal cortex. J Neurochem 2001;76:490–7. Muraki I, Inoue T, Koyama T. Effect of co-administration of the selective 5-HT1A receptor antagonist WAY 100,635 and selective 5-HT1B/1D receptor antagonist GR 127,935 on anxiolytic effect of citalopram in conditioned fear stress in the rat. Eur J Pharmacol 2008;586:171–8. Nakamura K, Kurasawa M. Anxiolytic effects of aniracetam in three different mouse models of anxiety and the underlying mechanism. Eur J Pharmacol 2001;420:33–43. Nishikawa H, Inoue T, Izumi T, Koyama T. Synergistic effects of tandospirone and selective serotonin reuptake inhibitors on the contextual conditioned fear stress response in rats. Eur Neuropsychopharmacol 2007a;17:643–50. Nishikawa H, Inoue T, Masui T, Izumi T, Koyama T. Effects of cytochrome P450 (CYP) 3A4 inhibitors on the anxiolytic action of tandospirone in rat contextual conditioned fear. Prog Neuropsychopharmacol Biol Psychiatry 2007b;31:926–31. Radja F, Laporte AM, Daval G, Verge D, Gozlan H, Hamon M. Autoradiography of serotonin receptor subtypes in the central nervous system. Neurochem Int 1991;18:1-15.

1819

Rittenhouse PA, Bakkum EA, O'Connor PA, Carnes M, Bethea CL, van de Kar LD. Comparison of neuroendocrine and behavioral effects of ipsapirone, a 5-HT1A agonist, in three stress paradigms: immobilization, forced swim and conditioned fear. Brain Res 1992;580:205–14. Sacchetti B, Lorenzini CA, Baldi E, Tassoni G, Bucherelli C. Auditory thalamus, dorsal hippocampus, basolateral amygdala, and perirhinal cortex role in the consolidation of conditioned freezing to context and to acoustic conditioned stimulus in the rat. J Neurosci 1999;19:9570–8. Santos JM, Martinez RC, Brandao ML. Effects of acute and subchronic treatments with fluoxetine and desipramine on the memory of fear in moderate and highintensity contextual conditioning. Eur J Pharmacol 2006;542:121–8. Sharp T, Umbers V, Gartside SE. Effect of a selective 5-HT reuptake inhibitor in combination with 5-HT1A and 5-HT1B receptor antagonists on extracellular 5-HT in rat frontal cortex in vivo. Br J Pharmacol 1997;121:941–6. Sotres-Bayon F, Cain CK, LeDoux JE. Brain mechanisms of fear extinction: historical perspectives on the contribution of prefrontal cortex. Biol Psychiatry 2006;60:329–36. Spennato G, Zerbib C, Mondadori C, Garcia R. Fluoxetine protects hippocampal plasticity during conditioned fear stress and prevents fear learning potentiation. Psychopharmacology (Berl) 2008;196:583–9. Stein C, Davidowa H, Albrecht D. 5-HT1A receptor-mediated inhibition and 5-HT2 as well as 5-HT3 receptor-mediated excitation in different subdivisions of the rat amygdala. Synapse 2000;38:328–37. Traber J, Glaser T. 5-HT1A receptor-related anxiolytics. Trends Pharmacol Sci 1987;8:432–7. Unnerstall JR, Fernandez I, Orensanz LM. The alpha-adrenergic receptor: radiohistochemical analysis of functional characteristics and biochemical differences. Pharmacol Biochem Behav 1985;22:859–74. Van Dijken HH, Van der Heyden JA, Mos J, Tilders FJ. Inescapable footshocks induce progressive and long-lasting behavioural changes in male rats. Physiol Behav 1992;51:787–94. Watson J, Rayner R. Conditioned emotional reactions. J Exp Psychol 1920;3:1-14. Wilensky AE, Schafe GE, Kristensen MP, LeDoux JE. Rethinking the fear circuit: the central nucleus of the amygdala is required for the acquisition, consolidation, and expression of Pavlovian fear conditioning. J Neurosci 2006;26:12,387–96. Winslow JT, Noble PL, Davis M. Modulation of fear-potentiated startle and vocalizations in juvenile rhesus monkeys by morphine, diazepam, and buspirone. Biol Psychiatry 2007;61:389–95. Wise RA, Hoffman DC. Localization of drug reward mechanisms by intracranial injections. Synapse 1992;10:247–63. Wolpe J. The practice of behavioral therapy. New York: Pergamon Press; 1969. Wright DE, Seroogy KB, Lundgren KH, Davis BM, Jennes L. Comparative localization of serotonin1A, 1C, and 2 receptor subtype mRNAs in rat brain. J Comp Neurol 1995;351:357–73. Yokoyama M, Suzuki E, Sato T, Maruta S, Watanabe S, Miyaoka H. Amygdalic levels of dopamine and serotonin rise upon exposure to conditioned fear stress without elevation of glutamate. Neurosci Lett 2005;379:37–41. Yoshioka M, Matsumoto M, Togashi H, Saito H. Effects of conditioned fear stress on 5HT release in the rat prefrontal cortex. Pharmacol Biochem Behav 1995;51:515–9. Yoshioka M, Matsumoto M, Togashi H, Saito H. Effect of conditioned fear stress on dopamine release in the rat prefrontal cortex. Neurosci Lett 1996;209:201–3. Zhang Y, Raap DK, Garcia F, Serres F, Ma Q, Battaglia G, et al. Long-term fluoxetine produces behavioral anxiolytic effects without inhibiting neuroendocrine responses to conditioned stress in rats. Brain Res 2000;855:58–66. Zohar J, Westenberg HG. Anxiety disorders: a review of tricyclic antidepressants and selective serotonin reuptake inhibitors. Acta Psychiatr Scand Suppl 2000;403: 39–49.