Neurobiology of Generalized Anxiety Disorder

GENERALIZED ANXIETY DISORDER

0193-953X/Ol $15.00

+ .OO

NEUROBIOLOGY OF GENERALIZED ANXIETY DISORDER Praveen V. Jetty, MD, MRCPsych, Dennis S. Charney, MD, and Andrew W. Goddard, MD

Generalized anxiety disorder (GAD) is a relatively new diagnostic entity first defined as a distinct category in the DSM-III.7 Two major epidemiologic studies, the National Survey of Mental Health and WellBeing2and the National Comorbidity Survey12ohave shown an incidence of 3.6% and 3.1% per year, respectively. Based on the National Comorbidity Survey data in the United States alone, more than 9 million people are afflicted with GAD at some point during their lifetimes. Also, GAD is one of the most common psychiatric disorders seen by primary care physicians. GAD was once considered by many clinicians to be a relatively mild condition, but new scientific data have challenged the perception that GAD is mild and minimally disabling. GAD has been found to be associated with various adverse social consequences, such as being more prevalent in lower socioeconomic groups, being poorly paid, being on disability, and experiencing more marital instability and divorce than the general population.12 GAD has a lifetime comorbidity rate of 62% with major depressive disorder (MDD), 39.5% with dysthymia, 37.6% with alcoholism, and 34.4% with social phobia. Comorbid GAD seems to be associated with an even greater level of functional impairment than From the Substance Abuse Program, Health South Metro West Hospital, Fairfield, Alabama (PVJ); the Mood and Anxiety Disorders Research Program and Experimental Therapeutics and Pathophysiology Branch, National Institute of Mental Health, Bethesda, Maryland (DSC); and the Anxiety Disorders Program, Department of Psychiatry, Yale University School of Medicine and Connecticut Mental Health Center, New Haven, Connecticut (AWG) THE PSYCHIATRIC CLINICS OF NORTH AMERICA VOLUME 24 * NUMBER 1 * MARCH 2001

75

76

JETTY et a1

GAD alone. Follow-up studies have shown a substantial impairment and even poorer outcome among people with GAD than among those with lZ5Thus, it is evident that GAD is neither an infrequent panic disorder.lZ1, nor mild condition but an important public health problem that merits additional research on treatment and mechanism of illness. The diagnostic concept of GAD, however, is somewhat controversial. Some investigators have suggested that GAD is a residual or prodroma1 phase of MDD,9z,98 or a “forme fruste” of MDD.61Other investigators have observed that it is attributable to other anxiety disorders, such as social phobias1 or panic di~0rder.l’~ Still others have conceptualized GAD as a dimension of personality function, a generalized anxious temperament that, when pronounced, constitutes psychopathology,4 whereas other investigators have conceptualized GAD as a part of a general neurotic A well-designed 5-year longitudinal follow-up study, however, showed significant diagnostic stability for GAD, supporting the reliability of the diagnosis.lZ1Thus additional research must be carried out to clarify the diagnostic boundaries of GAD so that meaningful conclusions may be drawn from research studies. This article provides a concise picture of preclinical and clinical research pertinent to the neurobiology of GAD. Research on functional neuroanatomy and fear circuitry; animal models relevant to GAD; research on the genetics of GAD; and neurochemistry, neurophysiology, and functional imaging studies relevant to GAD are discussed. Finally, a crucial discussion of several biological models of GAD is presented, and paths for future research are suggested. GENETIC STUDIES

Vulnerability to anxiety disorders may be determined in part by genetic predisposition. Genes may influence the expression of key neurotransmitters, thereby modifying behaviors. Because GAD is a relatively new diagnostic entity, and there have been major changes in the diagnostic criteria, it is not surprising that there is disagreement over the genetic findings from work of different researchers. There does seem to be a modest genetic contribution for the development of GAD, however, as evidenced by the research described herein. The various family studies of GAD that have been conducted generally point to GAD being a familial syndrome. One research found that 19.5% of first-degree relatives of GAD probands had developed the disorder compared with only 3.5% of control subjects’ families, with a corresponding relative risk of 5.6, which is large compared with other anxiety disorders. Other investigatorslo3diagnosed GAD in 22% of first-degree relatives of 33 probands with anxiety disorders, 13 of whom had GAD, but other investigatorss0found no evidence of inheritance of GAD in their family study. Other studies have suggested that GAD and MDD cosegregate in fa mi lie^.^, 118 Data from epidemiologic surveys also have shown a high rate of comorbidity of this disorder.

NEUROBIOLOGY OF GENERALIZED ANXIETY DISORDER

77

Although family studies may discriminate the impact of individualspecific environmental factors from those of genes and the family environment, they are unable to make the important distinction between the latter two factors. To help make this distinction, twin and adoption studies are required. Twin studies of GAD include one conducted by Torgersen et a1,l1l who studied anxiety disorders in 12 monozygotic and 20 dizygotic twin pairs in Finland. None of the monozygotic and only one of the dizygotic twins was concordant for GAD. Another group8 found that 21.5% of 63 monozygotic twin pairs were concordant for GAD compared with 13.5% of 81 dizygotic pairs, whereas another studylo4showed that three of five monozygotic and one of seven dizygotic twin pairs were concordant for GAD. Large cohorts60, 98 were used to study GAD and depression in the Virginia twin study and a sample of twins from the Swedish twin registry. In both samples, the same genetic factors accounted for both disorders. GAD and depression were genetic, but an index monozygotic twin with either diagnosis was as likely to have a twin with the other disorder as with the same disorder. A follow-up study by Kendler et aF2 showed additional support for the hypothesis that genetic factors are relatively nonspecific in their impact on symptoms of depression and GAD. They suggested that GAD has approximately a 30% heritability rate, but the study has been criticized on some methodologic shortcomings, with the diagnostic criteria being loosened to gain a larger sample size (e.g., DSM-111-R 6-month duration criterion for GAD being changed to 1 month), but this study is the largest to date, with 2352 female twin subjects. The findings suggest that the genetic liability between the two disorders is identical and that which disorder that develops is determined entirely by the environment. Because familial environment contributed to neither disorder, the environmental influences must be considered unique. These findings are echoed by a family study of cotransmission of depression anxiety disorders and alcoholism:* which showed that more than half the variance in liability for GAD and depression may be accounted for by extrafamilial influences and therefore not caused by a shared environment. In a study of male twins to test for distinction between subjects with GAD and panic disorder, Scherrer et alloofound that the lifetime co-occurrence of GAD and panic disorder could not be explained by family environmental influences. Thus, although evidence from some studies show that MDD and GAD share the same genotype, the expression of these disorders seems to depend on environmental influences unique to each patient. Molecular genetic studies of the anxiety disorders are still in their infancy. Some of the promising approaches used are molecular genetics of personality characteristics, linkage studies in animal behavioral models, and candidate gene studies in transgenic animals. It is hoped that these techniques will lead to a better understanding of the anxiety disorders at a molecular level. Several genome searches have been started for panic but so far there have been none for GAD. One GAD study focused on the serotonin (5-HT) transporter gene lo-

cated on chromosome 17q.91Significantly higher proportions of the GAD group than controls had the Stin2.12 allele. The odds ratio was 3.61, but this result is not only specific to GAD but also seen in obsessivecompulsive disorder and MDD. The same groupg0 also studied a polymorphism on the catechol-0-methyl transferuse gene but found no statistically significant difference between subjects and controls. Several strategies, such as systematic genome scanning and candidate-gene strategies, may be used for understanding linkage. The receptors in the yaminobutyric acid (GABA), norepinephrine (NE), corticotropin-releasing factor (CRF), and 5-HT systems would be considered prime candidates for the candidate-gene strategy. Despite many weaknesses, candidategene studies are a reasonable strategy in that the approach is hypothesis based, and, if a gene is found using this method, money and time would be saved. Genetic studies of GAD consistently indicate that genes contribute to the development of GAD. Although family and twin study data suggest this, they also demonstrate that the influence of genes is modest compared with classic Mendelian diseases or even the more heritable polygenic diseases. So clinicians must prepare to confront genes of small effect that most likely contribute incrementally to the threshold for GAD. FUNCTIONAL IMAGING STUDIES

There have been few studies on functional imaging in GAD. Data from clinical imaging studies suggest the involvement of the occipital cortex in GAD. Buchsbaum et all8 found a significant decrease of occipital lobe metabolism after benzodiazepine administration in patients with GAD. Wu et allz studied 18 patients who met DSM-I11 criteria for GAD and were randomized to receive clorazepate or placebo during a series of activation and vigilance tasks. The study showed higher relative metabolic rates for subjects with GAD in parts of the occipital, temporal, and frontal lobe metabolism and cerebellum relative to normal controls and a decrease in basal ganglia metabolism. No right-left hippocampal asymmetry was found, as has been reported in panic disorder. Arousal tasks resulted in activation of the basal ganglia and right parietal metabolism. After benzodiazepine therapy, a reduction in glucose metabolism was found in the cortex, limbic system, and basal ganglia compared with the control subjects. The investigators concluded that the findings support a role for the potential involvement of the basal ganglia in GAD. Tiihonen et alllo conducted an investigation using MR imaging and single photon emission CT to assess central benzodiazepine receptor binding and distribution in subjects with GAD. Subjects with GAD had significantly decreased reduction in the density of these receptors in the left temporal pole. Results of various studies are summarized in Table 1. Neuroimaging studies in GAD are at an early stage. Unfortunately, there is a paucity of literature available on the subject, and the significance of findings to date is unclear but suggestive of brain changes in some

SPET, MR imaging

GAD, 18

GAD, 10; controls, 10

GAD, 18; controls, 15

Buschbaum et al, 198718

Tiihonen et al, 1997'1°

Wu et al, 1991lZ3

Decreased GMR in visual cortex during treatment; increased rGMR in BG and TH Benzodiazepine receptor binding was decreased in the Ltemporal pole among GAD patients; more homogeneous cerebral benzodiazepine receptor density distribution in patients with GAD Basal ganglia, temporal poles, cortex, CG resting rGMR GAD < NC cerebellum, L occ-cortex, R precentral frontal cortex resting GAD > NC

Global CBF insignificantly decreased in GAD patients COz-induced CBF increase GAD = NC

Findings

Involvement of basal ganglia in GAD; ? clue to ruminations in GAD, similar to the basal ganglia-frontal lobe circuits in OCD

CBF inversely correlated with trait anxiety ? CBF change inversely correlated with state anxiety Occipital lobe involvement in GAD; ? related to hypervigilance Trait abnormality that leads to abnormal stimulus processing in GAD

Inferences

BG = basal ganglia; CBF = cerebral blood flow; rCBF = regional cerebral blood flow; CG = cingulate cyrus; GMR = glucose metabolic rate; L = left; NC = normal control; PET = positron emission tomography; Occ = occipital; R = right; rGMR = regional glucose metabolic rate; OCD = obsessive-compulsive disorder; GAD = generalized anxiety disorder; SPET = single photon emission tomography; MR = magnetic resonance.

PET, GMR, rGMR

PET, GMR, rGMR

GAD, 13; controls, 13

Mathew and Wilson, 198876

Xenon inhalation, CBF, rCBF Xenon inhalation, CBF, rCBF

GAD 9, controls, 9

No. Subjects

Mathew et al, 198277

Study

Scan Type (Function Measured)

Table 1. FUNCTIONAL IMAGING STUDIES IN GENERALIZED ANXIETY DISORDER

80

J E W et a1

neuroanatomic regions that are relevant in arousal and anxiety. With increasing sophistication of technology, additional studies likely will clarify some of these ambiguities. NEUROCHEMISTRY y-Aminobutyric Acid Neuronal System

GABA is the main inhibitory neurotransmitter in the CNS and is widely distributed in all regions of the brain. The benzodiazepine receptors and GABAA receptors are part of the same macromolecular complex. These receptors have different binding sites but are functionally coupled and regulate themselves in an allosteric manner.70,74, lz6 The hypothesis that alteration of function of this complex may have a significant role in the pathophysiology of GAD is suggested by some lines of evidence from animal and studies, as discussed here. In one study, 30 animals exposed to long-term, inescapable stress in the form of cold swim or foot shock showed a decrease in benzodiazepine receptor binding in frontal cortex, hippocampus, and hypothalamus-areas associated with the neural circuitry of fear and anxiety. In another studyE y2 subunit knockout mice showed enhanced behavioral inhibition toward natural aversive stimuli and heightened responsiveness in trace fear conditioning and ambiguous cue discrimination. Synaptic clustering of GABAA receptors in mice heterozygous for the y2 subunit was reduced, mainly in hippocampus and cerebral cortex. This model suggests GABAA receptor dysfunction as a potential causal predisposition to anxiety disorders, such as GAD. The human literature also supports a role for benzodiazepine dysfunction in anxiogenesis. In one studyz9 normal subjects experienced severe anxiety reactions to benzodiazepine inverse agonists, such as pcarbolines. Another showed that benzodiazepines were clinically effective in the treatment of patients with GAD. Patients with GAD have been found to have abnormally low levels of peripheral lymphocyte benzodiazepine receptors (PBR), which normalize after treatment with benzodiazepinesrM,97 suggesting that GAD might be associated with an abnormal decrease of PBRs. This finding also is seen in obsessivecompulsive disorder but not in panic disorder, suggesting that changes in lymphocyte PBR may distinguish between different types of anxiety states. The decrease in the PBRs in patients with GAD is paralleled by a concomitant decrease in the relative content of mRNA encoding PBR, suggesting that the rate of synthesis of these receptors also decreases during active illness, which reverses with successful treatment. Recent research indicates that PBRs may modulate the central GABAA receptor function, possibly by regulating steroidogenesis in peripheral tissues and in the brain, suggesting a mechanism by which peripheral abnormalities may affect the CNS in GAD and other disorders. In studies of PBRs located on monocytes of patients with GAD, the ability of the

NEUROBIOLOGY OF GENERALIZED ANXIETY DISORDER

81

monocytes to migrate toward chemoattracting benzodiazepines is completely abolished.99This chemotaxis still is impaired after their treatment with diazepam, with good clinical recovery and a normalization of receptor density. The investigators explained this finding on the basis that treatment reversed the PBR decrease but that the receptor still may be desensitized because of an impairment in the receptor-transducer coupling mechanisms and an increase of the anxiogenic endogenous ligand, such as the diazepam binding inhibitor in GAD. Additional studies combining imaging and molecular genetics may clarify the role of the GABA-benzodiazepine system in GAD. Noradrenergic System

Although the role of the NE system in acute and chronic animal models of stress has been well documented,2O NE also is associated with neural mechanisms, such as sensitizatiorP and fear conditioning, which are associated with stress. The role of NE in the pathophysiology of GAD is unclear because the data from the studies in this area are mixed. Patients with GAD show good response to noradrenergic drugs, such as imipramine and the lo2 Platelet norepinephrine 5-HT reuptake inhibitor (SSRI) ~enlafaxine.~~, monoamine oxidase activity is increased in patients with GAD.78Using skin-conductance after stress, some studies have shown a hyporesponsive and prolonged autonomic response in subjects with GAD.51Other investigatorslol compared NE function in patients with GAD and in those with MDD and healthy controls and found that plasma levels of glycol were increased NE and free 3-methoxy-4-hydroxyphenylethelene in the GAD group and that the number of a,-adrenoreceptors decreased. They concluded that NE activity is increased in subjects with GAD and that higher levels of catecholamines may lead to a decrease of the presynaptic a,-adrenoreceptors, but other studies77,86 did not report any difference between patients with GAD and normal controls. Another studybl found no significant differences between controls and patients with GAD in the levels of catechol-0-methyl transferase, dopamine phydroxylase, and monoamine oxidase. An additional study37 showed that subjects with GAD had increased levels of another NE metabolite, vanillylmandelic acid, in urine. A blunted growth-hormone response to clonidine has been described by some investigators’ and may be caused by presynaptic autoreceptor hypersensitivity or postsynaptic hyposensitivity. Another focus of study on the NE system in GAD has been on the inhibitory a,-adrenergic receptor. Inhibition of these receptors presynaptically results in increased NE activity and anxiety behaviors in animals. Other investigatorsz2used an a,-adrenergic antagonist to study difference in response and found that subjects with panic disorder or posttraumatic stress disorder (PTSD) were abnormally sensitive to generating anxiety responses but that patients with GAD were not. So, overall, the results of studies of the NE system in GAD have been inconsistent.

82

JE'ITYet a1

The available data are limited to draw meaningful conclusions, and future studies will need to address the issue of evaluating direct CNS indices of NE function. Also, with the changing taxonomy and diagnostic criteria, studies will have to use DSM-IV criteria to arrive at a coherent formulation. Serotonin

Another neurotransmitter associated with the pathogenesis of GAD is 5-HT. In animals, threatening situations seem to increase synaptic 5HT levels, and cortical and limbic regions may use this input to analyze and react to the situation.45In animal studies, 5-HT receptor subtypes 5HTlA, 5-HTZA,and 5-HT3 have been associated with fear behavior and consequently have been of most interest in human anxiety disorders, such as GAD. Mice bred without ~-HTIA receptors (i.e., 5-HTIAknockout mice) show decreased exploratory activity and increased fear of aversive environments, suggesting heightened anxiety.94The 5-HTIAreceptor agonists, such as buspirone, ipsapirone, and gepirone, which selectively decrease the firing rate of 5-HT neurons in animal models,lWhave been shown to be of help in treating GAD. Evidence for the possible role of 5-HT involvement in GAD comes from the following clinical studies. One study found decreased platelet paroxetine binding in patients with GAD.54Another study36showed a distinction between patients with GAD and those with panic disorder by showing that the urinary levels of the lysosomal enzyme N-acetyl-Pglusosaminidase were significantly higher in patients with GAD. Nacetyl-P-glusosaminidaselevels are thought to be an indirect marker for 5-HT activity. Another study37found that elevated urinary levels of the 5-HT metabolite 5-hydroxyindoleacetic acid predicted higher anxiety levels in patients with GAD, implying increased 5-HT metabolism in more anxious patients with GAD. On the other hand, one study found that the 5-HT synthesis inhibitor, PCPA, was anxiogenic in humans, implying an association between decreased 5-HT levels and anxiety. Hence, the relationship between 5-HT levels and anxiety has been inconsistent, so attention has focused on receptor subtypes. Studies have supported the role of 5-HTzc and 5-HTzAreceptor subtypes in GAD as reviewed by Kahn et al.57Also, patients with GAD have shown greater anger and anxiety responses to the mixed postsynaptic 5-HT agonistExaminantagonist meta-chlorophenylpiperazinethan control ing the differences in slow-wave sleep (which may be influenced by serotonergic activity) between patients with GAD and normal controls using ritanserin, some investigatorsz7found no significant differences, but the sample size was small, and although they did not find evidence to support the hypothesis that there is a generalized hypersensitivity of brain 5-HT2 receptors, they believed that their data could not exclude the presence of a regionally specific change in this receptor. Clinical data show efficacy of 5-HT, receptor partial agonists, such as buspirone,'06

NEUROBIOLOGY OF GENERALIZED ANXIETY DISORDER

83

and 5HT2 blockers, such as nefazodone, in GAD. Thus, although the general trend for the findings is that of 5-HT dysfunction in people with GAD, the question of hyperactivity or hypoactivity of 5-HT is unclear. Future studies (e.g., imaging studies of receptor subtypes and the 5-HT transporter) may elucidate the picture. Neuropeptides Cholecystokinin

Much research has associated neuropeptides with normal anxiety responses in animal models and in pathologic anxiety in humans. Of particular relevance is the cholecystokinin (CCK) system. CCK is one of the most abundant and widely distributed peptide neurotransmitters in the brain, and the CCK-B (brain) receptors are found with high densities in the hypothalamus, limbic system, basal ganglia, hippocampus, cortex, and brain stem, all of which have been associated with animal fear behaviors. There are several different CCK peptides, but CCK-4 and CCK-8 are of most interest in the study of anxiety." CCK-B is widely distributed in the brain and seems more directly involved in animal models of anxiety.47Observed responses to CCK-B receptor agonists include decreased exploratory activity in mice and rats, submissive and restless behavior in monkeys, and defensive attack in cats. Pretreatment with CCK-B antagonists, such as L-365,260, and CI-988 blocks the anxioCCK activates NE neurons in the locus genic activity of CCK agonists.*22 ceruleus by peripheral CCK receptors in vagal efferent pathways. CCK also interacts with the GABA system, and withdrawal from long-term diazepam administration is associated with an increase of CCK-8 binding in frontal cortex of rats. Endogenous CCK release enhances or amplifies stress responses that are 5-HT mediated. Researchers have tried to uncover the role of CCK in the modulation of anxiety and stress responses in human subjects. Studies have shown that intravenous injection of the CCK-B agonists CCK-4 and pentagastrin in subjects with panic disorder were panicogenic compared with control ~ubjects.'~ Another study15showed that intravenous pentagastrin induces higher rates of panic attacks in patients with GAD (71%) than in agematched and sex-matched control subjects (14%).One study3 explored the genetic basis of panic disorder and suggested a link between CCKB receptor gene polymorphism and panic disorder. The precise mechanism of CCK induced anxiety, however, is unclear. In response to the research associating CCK with anxiety and fear behaviors, anxiolytic drug development efforts have focused on the therapeutic potential of CCK-B receptor antagonists. Although preclinical studies have shown promise, studies in clinical trials in patients with GAD have proved disappointing,3 and the ability of CI-988 to block anxiety in lactateinduced panic and CCK-Pinduced panic has been limited, perhaps because of its poor bi~availability.~~ In another study, in patients with

84

JETTY et a1

GAD, the CCK-B antagonist CI-988 did not show any anxiolytic and neither did another CCK-B antagonist, L-365,260, in subjects with panic disorder.lo5Nevertheless, if agents from this class of compounds were found to be clinically effective in subjects with GAD, then they would have significant advantages over available pharmacotherapies in view of low abuse potential and rapid onset of action. Corticotropin-Releasing Factor

CRF is widely distributed in the brain, with highest concentrations found in the hypothalamus, where it is produced and secreted by the parvocellular neurons of the hypothalamic paraventricular nucleus. It is the major hypophysiotropic factor regulating basal and stress-induced release of adrenocorticotropic hormone, P-endorphin, and other proopiomelanocortin-derived pep tide^.^'^ Moderate and low levels of CRF also are present in cortical and limbic structures, respectively. The effects of CRF are mediated by two specific G-protein-coupled, seven-transmembrane domain receptors called CRF-1 and CRF-2. CRF-1 receptor expression is most abundant in neocortical, cerebellar, and limbic structures, whereas CRF-2 receptor expression is typically localized in subcortical structures, notably in the lateral septum and various hypothalamic areas. CRF-2a subtype is primarily expressed within the brain, whereas the subtype CRF-2P is found in the CNS and periphery.73CRF may contribute significantly to the behavioral responses to stress and the emotional behavior.ffi Intracerebroventricular administration of CRF in animals increases the concentrations of CRF in the CNS, produces physiologic and behavioral alterations virtually identical to those observed in laboratory animals in response to stress, including increases in heart rate and mean arterial pressure, suppression of exploratory behavior, induction of grooming, and reduction in feeding behavior. Additional actions include potentiation of acoustic startle responses, facilitation of fear conditioning, and enhancement of shock-induced freezing.@These effects do not occur after systemic administration of CRF and are not blocked by hypophysectomy, adrenelectomy, or pretreatment with dexamethasone, suggesting that these actions of CRF do not involve activation of the pituitary-adrenal axis but are mediated by CRF receptors present in the CNS. Several of these effects of CRF seem mediated by activation of the central NE system. Microinjection of CRF directly into the locus ceruleus of rats has been found to produce defensive withdrawal responses from a novel envir~nment.'~ Similarly, intra-amygdala infusion of CRF has been shown to produce anxiogenic behavior in the open field test and increase grooming in rats. Investigators used a transgenic mouse model overexpressing CRF and found that the mice exhibited a behavior state akin to anxiety," but in a study of C X F knockout mice, no differences between mutant and normal mice in 1997 were This may be because of compensation by other peptidergic mechanisms, but overall,

NEUROBIOLOGY OF GENERALIZED ANXIETY DISORDER

85

CRF seems to have a significant in anxiety-related and stress-related states. Clinical data suggesting a role for CRF in anxiety disorders has been accumulating for many years. Cerebrospinal fluid (CSF) levels of CRF have been shown to be elevated in patients suffering with obsessive-compulsive disorder6 and PTSD but not panic disorder.56Baseline CSF studies of CRF levels, however, have not shown significant differences between control subjects and patients with GAD, panic disorder, or obsessive-compulsive suggesting no tonic hypersecretion in people with these disorders. CRF may be episodically hypersecreted and may initiate fear responses in some contexts, however. In this regard, KOOB proposed a model that stress activates two types of CRF-NE interactions. Stress may activate CRF release in the region of the locus ceruleus, activating it and releasing NE in forebrain terminal projections, which, in turn, stimulates the release of CRF. Accordingly, a powerful feed-forward system, akin to kindlingG may be triggered by episodic or chronic stress. The past few years have seen important advances in the understanding of CRF and its mechanisms of action in modulating responses to stress. The finding that CRF stimulation increases anxiety-related behaviors in various animal models suggests that agents acting at CRF receptors may have therapeutic value in anxiety disorders. Industry is actively pursuing the development of nonpeptide and lipophilic CRF receptor antagonists as novel anxiolytics, and data on their efficacy should be available Neuropeptide-Y (NPY) is one of the most abundant peptides in the body. There are at least three NPY receptors, classified as Y1-3.117 High densities of Y, and Y, receptors are found widely in the CNS. The presence of NPY and its receptors in brain regions that are activated during stress (e.g., the amygdala and hypothalamus) has provided a rationale for studying NPY and related peptides in animal models of anxiety. Several studies in rats have shown that intracerebroventricular injections of NPY produce a behavioral profile consistent with an anxiolyticlike action in various anxiety models. One study17 showed that the anxiolytic activity is comparable to that of chlordiazepoxide. A series of studies have shown that intracerebroventricular infusion of high-affinity Y, agonists, including (G1ylZ1,G1+, LYS'~,Pro70, Leu"O)-NPY, yielded anxiolytic activity. Y2 receptor analogues have been found to be inactive in anxiety models in many studies, leading to the conclusion that the anxiolytic-like effects of NPY may be mediated primarily by activation of the Y, receptors. A few studies have shown that NPY might be involved in human anxiety but not in GAD. For example, one study119found that the lowest CSF concentrations of NPY in depressed patients were among those who had the most severe anxiety; another study13showed higher plasma NPY-like immunoreactivity in patients with panic disorder compared with healthy controls. Other investigators,lo7however, did not find sig-

86

JElTYet a1

nificant differences between controls and subjects with panic disorder or social phobia at basal and during stress stimulation. The reasons for these discrepancies are unknown but underscore the need for additional study of NPY in anxiety disorders in general. The robust anxiolytic effects observed with Y, agonists in preclinical research suggest that these compounds may become an alternative to benzodiazepines for the treatment of anxiety disorders, but additional work on searching for selective nonpeptide Y, receptor agonists is needed to potentially provide new drugs for the management of anxiety disorders. Tachykinins

Tachykinins are a group of neuropeptides that include substance-P, neurokinin-A, and neurokinin-B. The biological effects of tachykinins are mediated by the NK,, receptors. NK, and NK, receptors are widely distributed in the CNS and are found in significant density in brain regions traditionally associated with control of fear and anxiety, such as the amygdala, hypothalamus, and the periaqueductal gray. Substance-P, when injected in picomolar concentrations, is anxiogenic and anxiolytic depending on the dose and the brain region in which it is injected. Studies using a range of NK, antagonists have indicated that they possess anxiolytic activity, albeit some of them weakly, but studies on NK2 receptor antagonists in animal models have invariably shown anxiolytic activity, which is robust. The most studied drugs in this group are GR-15989711and SR-48968, which show anxiolytic activity similar to that of benzodiazepines but do not produce behavioral suppression at higher doses. Although the anxiolytic effects of NK, receptor antagonists are compelling, these effects have been obtained only in exploration tests. Research is needed using these in conflict paradigms to compare their efficacy to classic anxiolytics. Other Peptide Systems in Anxiety

The natriuretic peptide system consists of the atrial, brain, and Ctype. Although several preclinical studies have shown that natriuretic peptides display anxiolytic activity, the potential of these compounds must be evaluated in more preclinical paradigms and in clinical anxiety disorders. One of the primary behavioral effects of uncontrollable stress is analgesia, which results from the release of endogenous opioids. Most clinical research on opiates in treating anxiety disorders has focused on PTSD. Most studies support a hypothesis of increased release of endogenous opiates with stress in PTSD. This hypothesis also has been supported by a finding of elevated levels of P-endorphin in CSF in PTSD. Opioids are powerful suppressors of the NE system. They decrease hyperarousal in many patients with PTSD.I6 Opioids may be used as a treatment paradigm because increased NE activity is considered one of

NEUROBIOLOGY OF GENERALIZED ANXIETY DISORDER

87

the causes of anxiety. There has been a lack of studies in other anxiety disorders, and this area requires research.

Glutamate

The glutamate receptors mediate excitatory neurotransmission in the brain. Glutamate neurotransmission also is important in neuronal plasticity, as exemplified by long-term potentiation in the hippocampus, a mechanism of relevance in the pathophysiology of anxiety.87Stress activates cortical and limbic glutaminergic systems.67CNS circuitry mediating response to stress is heavily dependent on glutaminergic pathways. Stress-related animal models of depression have shown an increase in N-methyl-d-aspartate (NMDA) NR, subunit gene expression in the ventral tegmental area and regionally selective increases in NMDA binding or function.l" Chronic administration of NMDA antagonists and glycine-B partial agonists reduces behavioral deficits in animal models of anxiety and depression.112NMDA antagonists prevent fear conditioning and have direct anxiolytic activity.%Stress seems to stimulate glutamate release in the hippocampus, in part because of the effects of g l ~ ~ o ~ o r t i ~ oStudies i d s . ~ in ~ healthy human subjects suggest that NMDA antagonists produce disturbances in identity and perception resembling dissociation. Lamotrigine, a drug that reduces glutamate release in humans, attenuates the dissociative effects of ketamine in humans. The efficacy of NMDA antagonists in the treatment of human anxiety disorders has not been explored. Industry has already begun clinical testing of metabotropic glutamate receptor agonists at the GLU2/3 receptor in GAD and other anxiety disorders. Research using the startle paradigm, which is influenced by NMDA receptors, will be of interest because startle reactivity may be a vulnerability marker for the development of anxiety disorders. Research is needed to determine whether the inhibition of glutamate release by drugs such as lamotrigine or metabotropic agonists will help people with anxiety disorders.

Neuroactive Steroids

Neurosteroids are synthesized in the CNS and peripheral nervous system, particularly but not exclusively in myelinating glial cells, from cholesterol or steroidal precursors. They include pregnenolone and dehydroepiandrosterone (DHEA) and their sulphates and metabolites. These compounds may act as allosteric modulators of neurotransmitter receptors.'O Allopregnanolone is of great significance for psychiatric research because it binds with high affinity at nanomolar concentrations to GABA, receptors and potently facilitates GABA action at these receptors.44DHEA has been reported to act as a functional antiglucocorticoid. In animal studies, chronic exposure to DHEA-S impairs contextual fear

88

JETTY et a1

conditioning 24 hours after conditioning but not immediately after conditioning, which is similar to adrenalectomy. In mice, DHEA and DHEAS show anxiolytic activity in the plus-maze and the investigators in this study believed that neurosteroids were involved in the termination of stress responses. Also, allopregnanolone had anxiolytic activity, and the central nucleus of the amygdala is the key region involved in the mechanism? Acute foot-shock stress and carbon dioxide inhalation elicit a time-dependent increase in progesterone, allopregnalanolone, and allotetrahydrodeoxycorticosterone in rat brain and plasma and antagonism with abecarnil, an anxiolytic P-carboline derivative, blocks this increase. Other investigators have studied DHEA-%cortisol ratio values in subjects with panic disorder and found that subjects with panic disorder seemed to have a markedly increased DHEA-S-cortisol ratio No studies have been done in value compared with control human subjects with GAD, and this may be an area for future research in view of the possibility of developing novel treatments.

Lactate and Carbon Dioxide Challenge Paradigms

Pharmacologic challenge strategies are considered an increasingly important investigational method in studying anxiety disorders. Carbon dioxide-induced and lactate-induced anxiety have been studied with a view to distinguishing GAD from panic disorder. Studies found that many patients with panic disorder have panic attacks while rebreathing 5% carbon dioxide, whereas those with GAD did n ~ t . ~ Other ~ , investi~ ~ , ~ ~ gators found, however, that when 35% carbon dioxide was inhaled (single-inhalation) by subjects with GAD, they had less anxiety and fewer panic attacks than did patients with panic disorder, but that both groups had similar increases in somatic symptoms.116Another study evaluated the response of patients with GAD to the administration of sodium lactatez4and showed patients with GAD panicked at a lower rate (11Y0 v 41%) after lactate infusion than did patients with panic disorder but that patients with GAD reported significantly more anxiety symptoms than did normal controls. Another study showed lactateinduced panic rates of 26% for patients with panic disorder compared with 13% of those with GAD.68Rapee et a195reported that 15 minutes of breathing 5.5% carbon dioxide caused patients with panic disorder to develop a greater subjective response than in patients with GAD (49% v 21%), who, in turn, had a greater'response than did control subjects (O%).95 A history of panic attacks may be an important factor in determining anxiety sensitivity to challenge paradigms. Overall, the results of these challenge studies demonstrate that GAD and panic disorder are discrete disorders but that they share a common sensitivity to some physiologic stressors. These findings also imply that both of these disorders might derive from a dysregulation of the self-preserving response such as that elaborated in Klein's suffocation alarm theory.

NEUROBIOLOGY OF GENERALIZED ANXIETY DISORDER

89

NEUROBIOLOGICAL MODELS

The large body of work previously reviewed has contributed to the generation of hypotheses concerning the pathogenesis of GAD and other anxiety disorders. Biological and psychosocial factors seem to contribute to the development of GAD. Various biological models of anxiety have been proposed that provide a framework for understanding the phenomenology and neurobiology of GAD. An elegant model of human anxiety has been articulated by Gray,” based on preclinical data focusing on states of ”behavioral inhibition” seen in animals facing threat. The primary anatomic components of this behavioral inhibition system include the septohippocampal areas, locus ceruleus, and median raphe nuclei. Also, the prefrontal cortex may modulate septohippocampal activity. In this model, the septohippocampal system assesses the potential threat value of stimuli, and, when appropriate, activates the behavioral inhibition circuit, which, in turn, increases the monitoring of sensory stimuli for additional evidence of danger and suppresses motor activity. Gray postulates that theta electric rhythms transmitted from the septum to the hippocampus suppress hippocampal activation, acting as an “all-clear” signal, which also finds validation in clinical studies.108Increased NE output to the septum inhibits theta transmission, causing the hippocampus to maintain a state of heightened activity. Also, serotonergic stimulation of the septohippocampal area further activates this system. Under conditions of acute stress, 5-HT and NE activity are increased and amplify septohippocampal activity. Chronic activation of the behavioral-inhibition system may produce a chronic state of fear in animals. This consequent state of hypervigilance and increased arousal is analogous to the chronic anxiety in humans with GAD. In this model, antianxiety medications are said to exert their effects by the reduction of the NE and 5-HT inputs into the septohippocampal region. Benzodiazepine anxiolytics theoretically would reduce function in these systems by acting at the presynaptic GABA, receptors in the raphe nuclei, in the locus ceruleus, and postsynaptically in hippocampal formation. The strengths of the Gray model in explaining the pathogenesis GAD include the fact that it is well grounded in the preclinical literature. Also, it has direct clinical implications in relation to the psychological concept of behavioral inhibition, which has been identified as a risk factor for adult anxiety disorders.5O Behavioral inhibition would be expected to be associated with decreased behavioral activity and relatively decreased activity in the basal ganglia. This prediction has been observed in human studies of moderate anxiety states (e.g., GAD) that show a generally enhanced level of cortical activity and decreased basal ganglia activity,18,123 in contrast to severe anxiety states (e.g., panic), which tend to enhance basal ganglia activity to support the fight-or-flight response. These data may be caused by alterations in focal cerebral metabolism or may be explained by changes in regional cerebral blood flow secondary to moderate anxiety.55 Some human studies also have shown an in-

90

IETTY et a1

creased cortisol output in states of moderate anxiety, with no change or a decrease of plasma cortisol levels in panic consistent with the prediction of chronic NE hyperactivity suggested by the Gray model. Taken together, these human data provide support for Gray’s model. A weakness of this model is that some of the effective treatments for GAD, such as buspirone and the SSRTs, result in a net increase serotonergic function in the long term, apparently contradicting the idea that decreasing 5-HT input would decrease clinical anxiety. Also, in terms of its face validity, Gray’s model seems to account best for situational anxiety and exacerbations of a chronic illness brought on by stress. How the model accounts for the maintenance of a chronic anxiety state is unclear. A developmental-vulnerabilitymodel was proposed for conceptualizing the etiopathogenesis of the anxiety In this model, adverse experiences early in life predispose individuals to anxiety and mood disorders in adult life. A genetic predisposition, coupled with early stress in crucial phases of development, may result in a phenotype that is neurobiologically vulnerable to stress and may lower an individual’s threshold for developing anxiety or depression on additional stress exposure. The model is based on results from preclinical studies, which show that stress early in life results in persistent central CRF hyperactivity and increased stress reactivity in adulthood.72,93 The strengths of the Nemeroff model (i.e., the developmental vulnerability model) include its attempt to describe the neurobiological impact of early life stressors. It fits well with preclinical research linking sensitization to stressors with an increase of central CRF and NE function65and with research showing that chronic stress by CRF may alter hippocampal structure and funct i ~ n . ~A*potential limitation of this model is its nonspecificity with regard to GAD because the model may be equally well applied to the pathogenesis of MDD and PTSD. Another model focuses on inherited abnormalities in neurotransmitter systems, which express personality traits that may manifest as GAD.23 It conceptualizes personality as a combination of three heritable temperament dimensions-novelty seeking, harm avoidance, and reward dependence-which may be determined by neurotransmitter activity. Harm avoidance has been proposed to be associated with a high level of 5-HT activity and with anxiety. A high level of harm avoidance may predispose individuals to feel that they are in danger and to worry constantly, which may lead to generalized anxiety. Studies attempting to link candidate genes in these three neurotransmitter systems and behavioral traits have provided some empiric support for this model. Some research has suggested a link between the 5-HT transporter polymorphism and harm avoidance58and in research using prolactin response to 5-HT1, agonists, which suggests that serotonergic activity and harm avoidance are positively ~ o r r e l a t e dInvestigators .~~ have worked on the 5-HT transporter system,48,71 and observed that a short variant of the transporter occurring because of a polymorphism shows decreased transcription of the gene, reduced synthesis of the transporter molecule, and decreased uptake of 5-HT in vitro. Individuals who have the short variant have higher neuroticism scores on the NEO-PI (a commonly

NEUROBIOLOGY OF GENERALIZED ANXIETY DISORDER

91

used objective test of personality traits). K e n d l e r ’ ~model ~ ~ (discussed later) also provides some support for this observation. Although this model accounts for harm avoidance mediated by inhibiting 5-HT uptake, it does not account for the action of SSRIs, which also reduce anxiety by inhibiting 5-HT uptake, but the action of SSRIs is more complex than acutely increasing 5-HT in the synaptic cleft. Another limitation is how it would account for adult-onset GAD. The interaction of inheritance and the environment in the pathogenesis of GAD has been emphasized by other investigator^.^^ The KendleP group proposed that the genotype for GAD and MDD are similar if not identical but nonspecific in their impact on the expression of either disorder, which is determined entirely by the environment. Kendler observed that the common genetic diathesis was strongly linked to neuroticism, suggesting that this shared genetic factor tends to respond poorly to stress and therefore to experience frequent and intense episodes of distress and negative affect.6O. 61 Merikangas et aP2 found that more than half the variance in the liability for GAD and depression could be accounted for by extrafamilial influences and not by a shared environment. The limitation of the Kendler model is that it does not incorporate some of the current neurochemical and neuroanatomic data. Other groups (i.e., Charney et aP1) have focused on fear circuitry and the role of the amygdala in this circuitry in human anxiety. This model draws on the preclinical work of many investigators,28,31, 69 who have explicated the neural basis of fear conditioning and contextual conditioning. Neural processes, such as long-term potentiation in the central nucleus of the amygdala, may mediate the development of fear conditioning. Contextual conditioning involves the hippocampus to first create a representation of the environment using information received from the subiculum and the entorhinal cortex. Then the information is relayed through the basal or accessory basal nuclei of the amygdala and then to the central nucleus. The bed nucleus of the stria terminalis also may be involved in this process. Lesions in the basal nucleus of the amygdala attenuate contextual fear conditioning. In humans, the prefrontal cortex is necessary for the association of new sensory input with the memory of the type of emotional state usually associated with the type of situation in prior experience. The amygdala-prefrontal cortexlocus ceruleus interactions may be responsible for the establishment of the appropriate emotional valence in a given situation and thus are associated with pathologic fear and anxiety. Fear responses are implemented by the locus ceruleus, hippocampus, dorsal motor nucleus of the vagus, parabrachial nucleus, trigeminal nucleus, facial motor nucleus, striatum, and periaqueductal gray (the effector part of the circuit). The strength of this theory is its ability to integrate the numerous neural areas associated with fear and anxiety. GAD symptoms might be explained by contextual conditioning processes that occur in the extended amygdala and hippocampus. The Charney model accounts for several symptoms of GAD, including hyperarousal, increased motor tension, stress sensitivity, and avoidance behavior. Limitations of this

92

JETTY et a1

model include that it does not directly address which factors predispose one to a specific disorder and that it does not specify the effects of genetic influences. Additional research into the role of specific neural structures are required to address this issue. The amygdala-fear circuitry model of anxiety, though, offers a comprehensive model by which neural structures initiate and propagate anxiety disorders.

SUMMARY

On reviewing the literature on GAD and trying to summarize the various developments in the field of neurobiology of GAD, we see that a range of hypotheses try to explore and integrate the observations found into potentially meaningful theories. Abnormal serotonergic and GABAergic function occur in many patients with GAD. Functional imaging data have shown increased cortical activity and decreased basal ganglia activity in patients with GAD, which reverses with treatment, but it is apparent that no one theory is sufficiently comprehensive to propose a unitary hypothesis for the development of GAD and other anxiety disorders. GAD is a relatively new diagnosable condition, first introduced into the classification system of psychiatric disorders in 1980, and since then has undergone a series of changes in its conceptualization, with some investigators questioning the existence of the condition as a distinct entity. Any inferences that may be drawn from various studies must be guarded, and it is appropriate to compare studies using the same diagnostic criteria. Significant research has been done and may lead to exciting new discoveries in the treatment of anxiety disorders in general and GAD in particular. Gray’s model of behavioral inhibition, in which the septohippocampal system acts by assessing stimuli for the presence of danger and, when that is detected, activates the behavioralinhibition circuit, provides a neuroanatomic conceptualization that has been expanded by preclinical research. Some exciting work has been done on CRF and the concept of development, vulnerability, and kindling and some investigators have contributed to this area of interest. This concept supports the hypothesis that a genetic predisposition, coupled with early stress, in the crucial phases of development may result in a phenotype that is neurobiologically vulnerable to stress and may lower an individual’s threshold for developing anxiety or depression on additional stress exposure. The pharmaceutical industry is exploring treatment options using CRF antagonists, and qesearch on other neuropeptides, especially NPY, will be of interest. Research on neurosteroids also may bring the opportunity for pharmacologic treatment approaches. Future research on the startle reflex and on the NMDA and the metabotropic glutamate receptors is important. Future studies of a more homogenous patient population and using more sophisticated techniques, such as molecular genetic strategies and better imaging techniques, may answer some of the outstanding questions.

NEUROBIOLOGY OF GENERALIZED ANXIETY DISORDER

93

References 1. Abelson J L Blunted growth hormone response to clonidine in patients with generalized anxiety disorder. Arch Gen Psychiatry 48:157-162, 1991 2. Australian Bureau of Statistics: Mental health and well-being profile of adults, Australia, 1998. Canberra, Commonwealth of Australia, 1999 3. Adams JB, et a1 A double-blind, placebo-controlled study of a CCK-B receptor antagonist, CI-988,in patients with generalized anxiety disorder. J Clin Psychopharmacol 15:42&434, 1995 4. Akiskal HS: Toward a definition of generalized anxiety disorder as an anxious temperament type. Acta Psychiatr Scand Suppl 393:66-73, 1998 5. Akwa Y, et al: The amygdala mediates the anxiolytic-like effect of the neurosteroid allopregnanolone in rat. Behav Brain Res 106:119-125, 1999 6. Altemus M, et a1 Changes in cerebrospinal fluid neurochemistry during treatment of obsessive-compulsive disorder with clomipramine. Arch Gen Psychiatry 51:794-803, 1994 7. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, ed 3. Washington, DC, American Psychiatric Association, 1980 8. Andrews G, et al: The genetics of six neurotic disorders: A twin study. J Affect Disord 19:23-29, 1990 9. Angst J, et a1 Comorbidity of anxiety and depression in the Zurich cohort study of young adults. In Maser JD, Cloninger RC (eds): Comorbidity of Mood and Anxiety Disorders. Washington, DC, American Psychiatric Press, 1990 10. Baulieu EE: Neurosteroids: A novel function of the brain. Psychoneuroendocrinology 23963-987,1998 11. Beresford IJ, et al: GR159897 A potent non-peptide antagonist at tachykinin NK2 receptors. Eur J Pharmacol272:241-248, 1995 12. Blazer D, et al: Generalized anxiety disorder. in Psychiatric disorders in America: The epidemiologic Catchment Area Study, L. Bobus and D. Regier, Editors. 1991, Free Press: New York.180-203. 13. Boulenger JP, et a 1 Elevated plasma levels of neuropeptide Y in patients with panic disorder. Am J Psychiatry 153:114-116, 1996 14. Bradwejn J, et al: The cholecystokinin hypothesis of panic and anxiety disorders: A review. J Psychophannacol6345-351,1992 15. Brawman-Mintzer 0,et a1 Effects of the cholecystokinin agonist pentagastrin in patients with generalized anxiety disorder. Am J Psychiatry 154:700-702,1997 16. Bremner JD, Southwick SM, Chamey D S The neurobiology of posttraumatic stress disorder: An integration of animal and human research. In Saigh PA, Bremner J D Posttraumatic Stress Disorder: A Comprehensive Text. Boston, Allyn & Bacon, 1999, pp 103-143 17. Britton KT, et a1 Anxiolytic activity of NPY receptor agonists in the conflict test. Psychopharmacology 1326-13,1997 18. Buchsbaum MS, et al: Positron emission tomography assessment of effects of benzodiazepines on regional glucose metabolic rate in patients with anxiety disorder. Life Sci 402393-2400, 1987 19. Butler PD, et a 1 Corticotropin-releasing factor produces fear-enhancing and behavioral activating effects following infusion into the locus coeruleus. J Neurosci 10:17& 183, 1990 20. Chamey D S Anxiety disorders. In Chamey DS, Nestler EJ (eds): Neurobiology of Mental Illness. New York, Oxford University Press, 1999, pp 433-563 21. Chamey DS, Deutch A: A functional neuroanatomy of anxiety and fear: Implications for the pathophysiology and treatment of anxiety disorders. Crit Rev Neurobiol 10:41946,1996 22. Chamey DS, Woods SW, Heninger G R Noradrenergic function in generalized anxiety disorder: Effects of yohimbine in healthy subjects and patients with generalized anxiety disorder. Psychiatry Res 27173-182,1989 23. Cloninger C R A unified biopsychosocial theory of personality and its role in the development of anxiety states. Psychiatr Dev 4:167-226, 1986

94

JETTY et a1

24. Cowley DS, et a1 Response to lactate infusion in generalized anxiety disorder. Biol Psychiatry 24:409414, 1988 25. Crestani F, et al: Decreased GABA,-receptor clustering results in enhanced anxiety and a bias for threat cues. Nat Neurosci. 2833-839, 1999 26. Crowe RR, et al: Candidate gene study of eight GABAAreceptor subunits in panic disorder. Am J Psychiatry 1541096--1100,1997 27. da Roza Davis JM, Sharpley AL, Cowen PJ: Slow wave sleep and 5-HT2 receptor sensitivity in generalised anxiety disorder: A pilot study with ritanserin. Psychopharmacology 108:387-389, 1992 28. Davis M: Neurobiology of fear responses: The role of the amygdala. J Neuropsychiatry Clin Neurosci 9:382-402,1997 29. Dorow R, et a1 Clinical perspectives of lb-carbolines from first studies in humans. Brain Res Bull 19319-326,1987 30. Drugan RC, et al: A pretest procedure reliably predicts performance in two animal models of inescapable stress. Pharmacol Biochemistry Behav 33649-654, 1989 31. Fanselow M S Neural organization of the defensive behavior system responsible for fear. Psychonom Bull Rev L429-438, 1994 32. Fava M, et al: Dehydroepiandrosterone-sulfate/cortisol ratio in panic disorder. Psychiatry Res 28:345-350,1989 33. Feighner JP: Overview of antidepressants currently used to treat anxiety disorders. J Clin Psychiatry 22:18-22,1999 34. Ferrarese C, et al: Decreased density of benzodiazepine receptors in lymphocytes of anxious patients: Reversal after chronic diazepam treatment. Acta Psychiatr Scand 82169-173,1990 35. Fossey MD, et al: Cerebrospinal fluid corticotropin-releasing factor concentrations in patients with anxiety disorders and normal comparison subjects. Biol Psychiatry 39:703-707. 1996 36. Garvey MJ, et al: A biological difference between panic disorder and generalized anxiety disorder. Biol Psychiatry 34:572-575, 1993 37. Garvey MJ, et al: Relationship of generalized anxiety symptoms to urinary 5-hydroxyindoleacetic acid and vanillylmandelic acid. Psychiatry Res 571-5,1995 38. Germine M, et al: Anger and anxiety responses to m-chlorophenylpiperazine in generalized anxiety disorder. Biol Psychiatry 32457461, 1992 39. Goddard AW, et a1 Effects of the CCK antagonist CI-988 on responses to mCPP in generalized anxiety disorder. Psychiatry Res 85525-240, 1999 40. Gorman JM, et a1 Ventilatory physiology of patients with panic disorder. Arch Gen Psychiatry 45:31-39, 1988 41. Gduld E,-Tanapat P: Stress and hippocampal neurogenesis. Biol Psychiatry 46:14721479, 1999 42. Gray JA: The neuropsychological basis of anxiety. 151:lO-37, 1988 43. Griebel G: Is there a future for neuropeptide receptor ligands in the treatment of anxiety disorders? Pharmacol Ther 82:1-61, 1999 44. Guidotti A, Costa E: Can the antidysphoric and anxiolytic profiles of selective serotonin reuptake inhibitors be related to their ability to increase brain 3 alpha, 5 alphatetrahydroprogesterone (allopregnanolone) availability? Biol Psychiatry 442365473, 1998 45. Handley SL: 5-HT pathways in anxiety and its treatment. Pharmacology and Therapeutics 66:103-148, 1995 46. Hansenne M, Ansseau M: Harm avoidance and serotonin. Biol Psycho1 51:77-81,1999 47. Harro J, Vasar E, Bradwejn J: CCK in animal and human researchon anxiety. TIPS 14944-249, 1993 48. Heils A, et al: Allelic variation of human serotonin transporter gene expression. J Neurochem 662621-2624, 1996 49. Heim C, Nemeroff CB: The impact of early adverse experiences on brain systems involved in the pathophysiology of anxiety and affective disorders. Biol Psychiatry 461509-1522, 1999 50. Hill SY, Kagan J: Behavioral inhibition and developmental risk: Response to commentary [letter]. J Am Acad Child Adolesc Psychiatry 39:271-272, 2000

NEUROBIOLOGY OF GENERALIZED ANXIETY DISORDER

95

51. Hoehn-Saric R, McLeod DR, Zimmerli WD: Somatic manifestations in women with generalized anxiety disorder: Psychophysiological responses to psychological stress. Arch Gen Psychiatry 46:1113-1119, 1989 52. Hollander E, et a1 Fenfluramine, cortisol, and anxiety [letter; comment]. Psychiatry Res 31211-213, 1990 53. Holt PE, Andrews G: Hyperventilation and anxiety in panic disorder, social phobia, GAD and normal controls. Behav Res Ther 27453460,1989 54. Iny LJ, et al: Studies of a neurochemical link between depression, anxiety, and stress from. Biol Psychiatry 36:281-291, 1994 55. Johnson MR, Lydiard RB The neurobiology of anxiety disorders. Psychiatr Clin North Am 18~681-725,1995 56. Jolkkonen J, et a 1 CSF corticotropin-releasing factor is not affected in panic disorder. Biol Psychiatry 33:136-138, 1993 57. Kahn RS, et al: Pituitary hormone responses to meta-chlorophenylpiperazinein panic disorder and healthy control subjects. Psychiatry Res 3795-34, 1991 58. Katsuragi S, et a1 Association between serotonin transporter gene polymorphism and anxiety-related traits. Biol Psychiatry 45:368-370, 1999 59. Kendler KS: Major depression and generalised anxiety disorder same genes, (partly) different environments: Revisited. Br J Psychiatry 168(suppl 30):68-75, 1996 60. Kendler KS, et a1 Generalized anxiety disorder in women: A population-based twin study [see comments]. Arch Gen Psychiatry 49:267-272, 1992 61. Kendler KS, et al: Major depression and generalized anxiety disorder: Same genes, (partly) different environments? Arch Gen Psychiatry 49:716-722, 1992 62. Kendler KS, et a1 The structure of the genetic and environmental risk factors for six major psychiatric disorders in women: Phobia, generalized anxiety disorder, panic disorder, bulimia, major depression, and alcoholism. Arch Gen Psychiatry 52374383,1995 63. Kennedy JL, et a1 Investigation of cholecystokinin system genes in panic disorder. Mol Psychiatry, 4286285, 1999 64. Khan A, et a1 Platelet MAO-B activity in anxiety and depression. Biol Psychiatry 21M7-849, 1986 65. Koob G F Corticotropin-releasing factor, norepinephrine, and stress. Biol Psychiatry 46:1167-1180, 1999 66. Koob GF, Gold LH: Molecular biological approaches in the behavioural pharmacology of anxiety and depression. Behav Pharmacol 8:652, 1997 67. Krystal JH, et a1 NMDA agonists and antagonists as probes of glutamatergic dysfunction and pharmacotherapies in neuropsychiatric disorders. Harvard Rev Psychiatry 7125-143, 1999 68. Lapierre Y, Knott V, Gray R Psychophysiological correlates of sodium lactate. Psychopharmacol Bull 20:50-57, 1984 69. LeDoux J: Fear and the brain: Where have we been, and where are we going [see comments]? Biol Psychiatry 44:1229-1238, 1998 70. Leonard BE: Commentary on the mode of action of benzodiazepines. J Psychiatr Res ~ ~ ( S U 1):193-207, PP~ 1993 71. Lesch KP, et al: Association of anxiety-related traits with a polymorphism in the serotonin transporter gene regulatory region [see comments]. Science 274:1527-1531, 1996 72. Liu D, et a1 Maternal care, hippocampal glucocorticoid receptors, and hypothalamicpituitary-adrenal responses to stress [see comments]. Science 2771659-1662, 1997 73. Lovenberg W ,et a1 CRF2 alpha and CRF2 beta receptor mRNAs are differentially distributed between the rat central nervous system and peripheral tissues. Endocrinology 136:4139-4142, 1995 74. Lueddens H, Korpi ER Biological function of GABAJbenzodiazepine receptor heterogeneity. J Psychiatr Res 29:77-94,1995 75. Lydiard RB, Neuropeptides and anxiety: Focus on cholecystokinin. Clin Chem 40:315318, 1994 76. Mathew RJ, Wilson W H Cerebral blood flow changes induced by C 0 2 in anxiety. Psychiatry Res 23:285-294, 1988

96

JETTYet a1

77. Mathew RJ, et al: Catecholamines and anxiety. Acta Psychiatr Scand 65:142--147,1982 78. Mathew RJ, et al: Catecholamines and monoamine oxidase activity in anxiety. Acta Psychiatr Scand 63245-252, 1981 79. Melchior CL, Ritzmann RF. Pregnenolone and pregnenolone sulfate, alone and with ethanol, in mice on the plus-maze. Pharmacol Biochem Behav 48:89%397,1994 80. Mendlewicz J, Papadimitriou G, Wimotte J: Family study of panic disorder: Comparison to generalized anxiety disorder, major depression and normal subjects. Psychiatr Genet 3:73-78, 1993 81. Mennin D, Heimberg RG, Jack MS: Generalized anxiety disorder in primary social phobia: Functional and treatment implications. Presented at the 18th National Conference of Anxiety Disorders Association of America. Boston, 1998 82. Merikangas KR, Risch NJ, Weissman MM: Comorbidity and co-transmission of alcoholism, anxiety and depression. Psychol Med 2469-80, 1994 83. Miczek KA: Genetic approaches to anxiety and depression. Behav Pharmacol 8:657658,1997 84. Miserendino MJ, et al: Blocking of acquisition but not expression of conditioned fearpotentiated startle by NMDA antagonists in the amygdala. Nature 345:716-718, 1990 85. Moghaddam B, et al: Glucocorticoids mediate the stress-induced extracellular accumulation of glutamate. Brain Res 655:251-254,1994 86. Munjack DJ, et a1 Generalized anxiety disorder: Some biochemical aspects. Psychiatry Res 32:35-43, 1990 87. Nakanishi S, Molecular diversity of glutamate receptors and implications for brain function. Science 258597403,1992 88. Nisenbaum LK, et al: Prior exposure to chronic stress results in enhanced synthesis and release of hippocampal norepinephrine in response to a novel stressor. J Neurosci 11:1478-1484,1991 89. Noyes R Jr, et al: A family study of generalized anxiety disorder. Am J Psychiatry 144:1019-1024, 1987 90. Ohara K, et al: No association between anxiety disorders and catechol-0-methyltransferase polymorphism. Psychiatry Res 80:145-148,1998 91. Ohara K, et al: A variable-number-tandem-repeat of the serotonin transporter gene and anxiety disorders. Prog Neuropsychopharmacol Biol Psychiatry 23:55-65, 1999 92. Piccinelli M Comorbidity of depression and generalized anxiety: Is there any distinct boundary? Curr Opin Psychiatry 11:57-60,1998 93. Plotsky PM, Meaney MJ: Early, postnatal experience alters hypothalamic corticotropin-releasing factor (CRF) mRNA, median eminence CRF content and stress induced release in adult rats. Mol Brain Res 18195-200, 1993 94. Ramboz S, et al: Serotonin receptor 1A knockout: An animal model of anxiety-related disorder [see comments]. Proc Acad Sci U S A 95:14476-14481, 1998 95. Rapee RM, et al: Response to hyperventilation and inhalation of 5.5% carbon dioxideenriched air across the DSM-111-R anxiety disorders. J Abnorm Psychol 101:538-552, 1992 96. Rickels K, Case WG, Schweizer E: The drug treatment of anxiety and panic disorder. Stress Med 4:231-239, 1988 97. Rocca P, et a1 Peripheral benzodiazepine receptor messenger RNA is decreased in lymphocytes of generalized anxiety disorder patients. Biol Psychiatry 43:767-773,1998 98. Roy MA, et al: A twin study of generalized anxiety disorder and major depression. Psychol Med 25:1037-1049, 1995 99. Sacerdote P, et al: Benzodiazepine-induced chemotaxis is impaired in monocytes from patients with generalized anxiety disorder. Psychoneuroendocrinology 2424S249, 1999 100. Scherrer JF, et al: Evidence for genetic influences common and specific to symptoms of generalized anxiety and panic. J Affect Disord 5725-35, 2000 101. Sevy S, et al: Noradrenergic function in generalized anxiety disorder, major depressive disorder, and healthy subjects. Biol Psychiatry 25341-152, 1989 102. Sheehan DV: Venlafaxine extended release (XR) in the treatment of generalized anxiety disorder. J Clin Psychiatry 22:23-28, 1999 103. Skre I, et al: A family study of anxiety disorders: Familial transmission and relation-

NEUROBIOLOGY OF GENERALIZED ANXIETY DISORDER

97

ship to mood disorder and psychoactive substance use disorder. Acta Psychiatr Scand 90:366-374, 1994 104. Skre I, et a1 A twin study of DSM-111-R anxiety disorders. Acta Psychiatr Scand 88:8592, 1993 105. Sramek JJ, et al: Pilot study of a CCKB antagonist in patients with panic disorder: Preliminary findings. Anxiety 1:141-143, 1994 106. Stahl S M Mixed depression and anxiety: Serotonin-1A receptors as a common pharmacologic link. J Clin Psychiatry 58(supp18):20-26, 1997 107. Stein MB, et al: Plasma neuropeptide Y in anxiety disorders: findings in panic disorder and social phobia. Psychiatry Res 59183-188, 1996 108. Suetsugi M, et al: Appearance of frontal midline theta activity in patients with generalized anxiety disorder. Neuropsychobiology 41:108-112, 2000 109. Taylor DP, et a1 Pharmacological and clinical effects of buspirone. Pharmacol Biochem Behav 23687-694,1985 110. Tiihonen J, et al: Cerebral benzodiazepine receptor binding and distribution in generalized anxiety disorder: A fractal analysis. Mol Psychiatry 2463471, 1997 111. Torgersen S Genetic aspects of anxiety disorders. Arch Gen Psychiatry 40:1085-1089, 1986 112. Trullas R, Jackson B, Skolnick Anxiolytic properties of l-aminocyclopropanecarboxylic acid, a ligand at strychnine-insensitive glycine receptors. Pharmacol Biochem Behav 34313-316, 1989 113. Tyrer P: Neurosis divisible? Lancet 1:685488, 1985 114. Uhde TW, et a1 Longitudinal course of panic disorder: Clinical and biological considerations. Prog Neuropsychopharmacol Biol Psychiatry 9:39-51, 1985 115. Vale W, et al: Characterization of a 41 residue ovine hypothalamic peptide that stimulates the secretion of corticotropin and beta-endorphin. Science 213:1394-1397, 1981 116. Verburg C, Griez E, Meijer J: A 35% carbon dioxide challenge in simple phobias. Acta Psychiatr Scand 90:420-423, 1994 117. Wahlestedt C, et al: Modulation of anxiety and neuropeptide Y-Y1 receptors by antisense oligodeoxynucleotides. Science 259:528-531, 1993 118. Weissman MM. The epidemiology of anxiety disorders: rates, risks and familial patterns. J Psychiatr Res 22(suppl 1):99-114, 1988 119. Widerlov E, et al: Neuropeptide Y and peptide YY as possible cerebrospinal fluid markers for major depression and schizophrenia, respectively. J Psychiatr Res 22:6979. 1988 120. Wittchen H U DSM-111-R generalized anxiety disorder in the national comorbidity survey. Arch Gen Psychiatry 51:355-364, 1994 121. Woodman CL, et al: A 5-year follow-up study of generalized anxiety disorder and panic disorder. J Nerv Ment Dis 1873-9, 1999 122. Woodruff GN, Hughes J: Cholecystokinin antagonists. Annu Rev Pharmacol Toxic01 31:469-501, 1991 123. Wu JC, et al: PET in generalized anxiety disorder. Biol Psychiatry 29:1181-1199, 1991 124. Yoneda Y, Han D, Ogita K, Preferential induction by stress of the N-methyl-d-aspartate recognition domain in discrete structures of rat brain. J Neurochem 63:1863-1871,1994 125. Yonkers KA, et a1 Phenomenology and course of generalised anxiety disorder. Br J Psychiatry 168308-313, 1996 126. Zorumski CF, Isenberg KE, Insights into the structure and function of GABA-benzodiazepine receptors: Ion channels and psychiatry. Am J Psychiatry 148:162-173, 1991

Address reprint requests to Praveen V. Jetty, MD, MRCPsych Substance Abuse Program Health South Metro West Hospital Fairfield, AL 35064 e-mail [email protected]