«Distribution Agreement In presenting this thesis or dissertation as a partial fulfillment of the requirements for an advanced degree from Emory ...»
1.1 Neurodevelopmental Basis of Anxiety Disorders 1.1.1 Emergence of Anxiety Disorders in Children and Adolescents Psychiatric disorders are increasingly considered developmental phenomena, largely due to high prevalence in juveniles. Of all adults with a mental disorder, over 50% were first diagnosed before 15 years of age (Kim-Cohen et al., 2003). In terms of cases of specific disorders, an estimated 50% of anxiety disorders emerge by age 11, 90% of impulse control disorders by age 19, and 25% of mood disorder cases by age 18. The most prevalent psychiatric disorders in children are, by most accounts, anxiety disorders, due to their early onset and high lifetime prevalence of nearly 30% (Kessler et al., 2005). For example, around 15% of children and adolescents are diagnosed with specific phobias, nearly 10% with social phobias, and around 5% with separation anxiety disorder (Beesdo-Baum and Knappe, 2012).
Anxiety disorders in children pose a significant societal burden, due not only to excessive expression of fear that detracts from quality of life and can require treatment and care, but also to the effects on other health outcomes. Although there are a number of appropriate, age-specific fears that normally manifest during temporal windows throughout development (Gullone, 2000), the expression of excessive or age-inappropriate fears can occur throughout most of life. While the symptoms may change with age, the fearful responses experienced in those with anxiety disorders often continue into adulthood. Furthermore, anxiety disorders in youths contribute to the emergence of secondary psychological complications that include other anxiety disorders, depressive disorders, and substance abuse (Beesdo-Baum and Knappe, 2012). Because of high rates of homo- and heterotypic continuity for anxiety disorders, there is great need for treatments of juvenile anxiety that can intervene and potentially halt the progression of psychiatric illness.
The early emergence of anxiety disorders and the limited treatment options targeted to juvenile anxiety disorders highlight the need for research targeted specifically at their pathogenesis in early life (Cartwright-Hatton et al., 2006). The current conceptualization of anxiety in juveniles is based almost exclusively on models of adult anxiety disorders. Reframing anxiety disorders as developmental phenomena with distinct pathology in adults and children will be a critical step toward understanding their etiology (Cartwright-Hatton, 2006). Many potentially promising treatments for juvenile anxiety exist, but treatment efficacy is difficult to evaluate because of a limited understanding of the pathophysiology of anxiety disorders in the immature brain (Reinblatt and Riddle, 2007). These limitations stem from a lack of basic research targeted to the development of emotion; the normative development of brain regions that process emotion serves as a platform for the actions of genetic and environmental factors that promote emergence of neurodevelopmental disorders.
1.1.2 Early Life Sensitivity to Stress One well-documented risk factor for anxiety disorders is exposure to stress. Although the response to an acute stressor is often adaptive, enabling an organism to mobilize an appropriate reaction to the stressor, exposure to chronic or particularly potent stressors is capable of permanently influencing an organism’s stress response and emotional behavior. Anxiety disorders can be conceptualized as hyper-activation of certain components of the stress response, and chronic stress has been proposed to predicate anxiety disorders through sensitization to subsequent stressors (Chrousos and Gold, 1992; McEwen, 2004, 2007).
An association between stress and emotional dysfunction is present at all ages, but the effects of stress are particularly robust when exposure occurs during development. The timing of stress exposure is a critical factor, along with duration and intensity, in determining the effects on the brain and emotional outcomes (for review, see Lupien et al., 2009). Early life stress (ELS) is a well-documented risk factor for anxiety disorders, as well as depression, substance abuse, attention deficit hyperactivity, and autism spectrum disorders (MacMillan et al., 2001; Welberg and Seckl, 2001; Nemeroff, 2004; Moffitt et al., 2007; Kinney et al., 2008; Ronald et al., 2010;
Wang et al., 2013). Studies in numerous animal models, including rodents and primates, confirm the potency of ELS in promoting emotional deficits and abnormal stress reactivity later in life (Sanchez et al., 2001; Teicher et al., 2003). In line with the allostatic load model, that the cumulative burden of stress over one’s lifetime drives dysfunctional stress reactivity and psychiatric illness, the later-life stress response is influenced by ELS (Graham et al., 1999;
McEwen, 2004; Heim et al., 2008; Juster et al., 2010).
184.108.40.206 Effects of Early Life Stress Animal studies have proven extremely useful in characterizing the effects of ELS, particularly in recognizing the variability and complexity of those effects. Many such studies have recapitulated a number of long-term behavioral deficits caused by ELS in humans; these studies have also identified a variety of neurological and physiological effects discussed in detail below (Maccari and Morley-Fletcher, 2007; Weinstock, 2008).
The specific effects of ELS are dependent on the developmental timing of stress exposure. While direct stress to a child is the prototypical form of ELS, the indirect stress exposure of a fetus, termed prenatal stress (PS), in some cases has more pronounced effects than early postnatal stress (Estanislau and Morato, 2005). A number of studies have linked stress exposure in pregnant mothers to a host of deficits in their offspring (Kofman, 2002; Van den Bergh et al., 2005; Talge et al., 2007), including anxiety (Weinstock, 2001; O'Connor et al., 2003;
Van den Bergh and Marcoen, 2004), depression (Weinstock, 2001; Huizink et al., 2007;
Markham and Koenig, 2011), schizophrenia (Koenig et al., 2002; Khashan et al., 2008), attention deficit hyperactivity disorder (Huizink et al., 2007; Ronald et al., 2010), and autism spectrum disorders (Kinney et al., 2008; Ronald et al., 2010), as well as developmental delays and altered stress reactivity (Huizink et al., 2003; Bergman et al., 2007; Glover et al., 2010).
To add another layer of complexity, the impact of PS varies with a number of factors, including the timing of stress exposure during pregnancy. For example, stress to mothers during early pregnancy alters stress reactivity and stress neurotransmitter and hormone systems in the brains of the offspring (Mueller and Bale, 2008). In contrast, stress towards the end of pregnancy causes developmental delays in offspring, influencing a number of physiological and behavioral milestones (Barlow et al., 1978).
The effects of PS also depend on the quality of the stressor. For example, PS that occurs at unpredictable intervals results in developmental delays in rats, while predictable PS does not (Fride and Weinstock, 1984). However, there may be an interaction between stressor predictability and timing on offspring outcomes. Another study in rats comparing predictable and unpredictable stress restricted to the last week of gestation found opposite results: that the predictable stressor elicited the most robust changes in anxiety-like behavior and stress reactivity (Richardson et al., 2006). Comparing studies with different stress paradigms is not straightforward, as stressor intensity controls the long-term effects. Mild PS in some cases has little impact on the offspring (Mabandla et al., 2008), but can also promote fearful behaviors later in life (Griffin et al., 2003; Dickerson et al., 2005). Together, these studies highlight the potent but complex effects of PS, which depends on the timing, potency, and predictability of the stressor during pregnancy.
The impact of PS also depends on the sex of the offspring (Darnaudery and Maccari, 2008). One study found early gestation stress altered behavioral and endocrine components of the stress response and emotional behavior in male, but not female, mouse offspring (Mueller and Bale, 2008). The effects of slightly later gestational stress, presented from E10 to E18, were also detected in male but not female rat offspring, manifesting as emotional and memory deficits at 4 and 6 weeks of age, respectively (Nishio et al., 2001). However, another study found that stress in developing rats, from embryonic day (E)10 to E19 (of the 22-23 day long gestation), increased anxiety-like behavior selectively in female offspring at 4 weeks of age (Baker et al., 2008).
Finally, restraint stress from E14 to E21 in rats resulted in a more pronounced increase in anxietylike behavior in female offspring, but spatial memory deficits selectively in male offspring at 5 weeks of age (Zagron and Weinstock, 2006). Interestingly, the placentas of male but not female offspring exhibited altered protein expression following early PS (Mueller and Bale, 2008). These findings suggest there are sex differences in the basic mechanisms by which PS alters neurodevelopment, but the effects on each sex vary depending on the age and type of stress exposure.
While the work summarized in this section suggests PS contributes solely to negative outcomes, contrasting reports support a role for some types of PS in promoting resiliency (see Section 1.4.
220.127.116.11 Maturation of the Stress Response In order to understand how ELS contributes to the pathogenesis of anxiety disorders, it is necessary to consider the normative and maladaptive function of stress systems. Anxiety disorders can be considered as improper mobilization of the cognitive aspects of the stress response (Charney and Deutch, 1996). The perception of a stressor, defined as an actual or hypothetical disturbance of an individual’s environment, leads to the shift of resources towards body systems necessary for immediate survival. The ‘stress response’ is initiated by activation of a variety of brain regions and involves release of myriad neurotransmitters and hormones.
Corticotropin-releasing factor (CRF) is a neurotransmitter that coordinates many of the behavioral, endocrine, and autonomic aspects of the stress response (Vale et al., 1981; Dunn and Berridge, 1990; Owens and Nemeroff, 1991; Smagin et al., 2001). The perception of a stressor triggers immediate release of CRF from the hypothalamus and extended amygdala, which activates the hypothalamic-pituitary-adrenal (HPA) axis. The HPA axis, in turn, stimulates many physiological components of the stress response, including release of catecholamine neurotransmitters and activation of the sympathetic nervous system (for review, see Joels and Baram, 2009). HPA axis activation induces the release from the adrenal glands of hormones of the glucocorticoid family (cortisol in humans, corticosterone in rodents). Glucocorticoids elicit peripheral effects contributing to the stress response, but also pass through the blood-brain barrier and activate receptors in the brain that provide negative feedback on the HPA axis (Zarrow et al., 1970; Joels and Baram, 2009). Dysfunctional CRF release and activation of the HPA axis have been implicated in the pathophysiology of both anxiety and depressive disorders (Arborelius et al., 1999; Yehuda, 2001; Nemeroff and Vale, 2005).