Review

Stress, sensitive periods and maturational events in adolescent depression

Susan L. Andersen and Martin H. Teicher
Department of Psychiatry, McLean Hospital and Harvard Medical School, 115 Mill Street, Belmont, MA 02478, USA

In this paper, we provide an overview of how the maturation of specic brain regions and stress exposure during windows of vulnerability initiate a series of events that render adolescents exceptionally susceptible to the development of depression. This stress-incubation/corticolimbic development cascade provides a means of understanding why depression emerges with such force and frequency in adolescence. The development of the prefrontal cortex, hippocampus, amygdala and ventral striatum is described from a translational perspective as they relate to stress exposure, onset, pathogenesis and gender differences in depression. Adolescent depression is a serious recurrent brain-based disorder. Understanding the genesis and neurobiological basis is important in the development of more effective intervention strategies to treat or prevent the disorder. Introduction The overriding issue of this review is to understand why depression emerges with such force and frequency in adolescence, particularly in young women. Conceivably, a host of psychosocial factors can render adolescents especially vulnerable, but our focus will be on neurobiological factors. In particular, we will examine the interplay of genetic, maturational and experiential factors affecting mood using a translational perspective that melds clinical and basic laboratory ndings. Basic epidemiology of adolescent depression Major depression disorder is a common and serious disorder of adolescence [1]. Lifetime prevalence increases dramatically from 1% of the population under age 12 to $17%25% of the population by the end of adolescence [2]. The greatest surge in newly emergent cases occurs between 15 and 18 years [3]. Cross-sectional and prospective epidemiological studies indicate that anxiety disorders often precede the emergence of depression and identify children at risk for developing depression [4,5]. It is possible that early anxiety and later depression share a common genetic basis with a different developmental time course [5]. One study found that adolescent anxiety typically preceded onset of insomnia, whereas episodes of depression followed bouts of insomnia [6], suggesting that sleep disturbance might serve, in some instances, as a mediating link.
Corresponding author: Andersen, S.L. (sandersen@mclean.harvard.edu).

Adolescent major depression disorder typically follows a recurrent episodic course, with episodes averaging 79 months in those seeking treatment [1]. Adolescent onset is associated with a more chronic, severe and disabling form of depression, higher rates of family history and more suicide attempts than depression that rst emerges in adulthood [7]. Children and adolescents are more likely to present with irritability without overt sadness than adults [1]. Depressed adolescents show more signs of anhedonia, hypersomnia, decreased ability to think and concentrate, melancholia and suicidality than depressed children [8] and greater disturbance in circadian restactivity rhythms [9]. According to the National Center for Health Statistics, the incidence of suicide unfolds with age: suicide before the age of 10 is rare, increases 100-fold between 10 and 14 years, and rises an additional 10 times between the ages of 15 and 19. Prevalence rates for depression are twice as high in females as males. This ratio is not apparent in childhood, but emerges by $14 years of age [10]. Onset of depression is temporally linked to menarche [11], suggesting a hormonal mechanism. Females experience a marked increase in a subtype of depression associated with anxiety, sleep/appetite disturbances and fatigue [12]. They can also experience more body image dissatisfaction, feelings of failure, concentration problems and work difculties [13]. By contrast, depressed boys are more anhedonic, and have greater diurnal variation in mood and energy [13]. Etiology of depression Although the specic etiology of major depression remains unknown, both heredity and early experience are critical determinants. Further, maturational events might potentially increase prevalence or trigger episodes (see Box 1). It is important to distinguish between specic etiological factors (e.g. genetic polymorphisms, childhood adversity) that increase risk in selective recipients and universal phenomena (e.g. puberty) that exert moderating effects on the entire population. Genes and adversity Genetic factors accounted for 40.4% of the variance in risk of major depression in one study of adolescent female twins (n = 3416) [14]. Non-shared environmental effects accounted for the remaining 59.6%. Similarly, large-scale epidemiological studies indicate that exposure to early
183

0166-2236/$ see front matter 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.tins.2008.01.004 Available online 10 March 2008

Review
Box 1. Depression, anhedonia, reward and the dopamine system
The mesolimbic system, composed of the dopaminergic cell bodies in the ventral tegmental area that project to the nucleus accumbens, plays a prominent role in reward processing of drugs of abuse and natural rewards, including food, sex and social interactions [78]. As reviewed elsewhere [78], the nucleus accumbens receives inputs from multiple brain regions, including the cortex, hypothalamus and amygdala. Reductions in dopamine neurotransmission within the ventral striatum (nucleus accumbens/extended amygdala) have been observed in depression [79]. These changes are likely to influence the primary symptoms of depression that are related to low positive affectivity: reduced energy, anhedonia, loss of libido, loss of appetite, social withdrawal and psychomotor retardation [59,78]. Treatment with antidepressants with some dopamine (and noradrenergic) activity, such as bupropion, restores lost positive affect. The reduction in positive affectivity might be a normal process that occurs during adolescence [60]. To this end, extracellular levels of dopamine in the dorsal and ventral striatum are lower during adolescence relative to other ages [80]. By contrast, dopaminergic activity in the prefrontal cortex peaks during adolescence [81]. Here, dopamine within the adolescent cortex aids in the processing of contextual stimuli [76] and enhances responsiveness to stressful events [73]. As connections between the cortex and accumbens continue to mature [76], the prefrontal cortex increases its modulation of accumbens activity. Together, this might further render the adolescent uniquely vulnerable to contextual events. Whether increased cortical activity during adolescence is related to reduced accumbens activity through a direct or indirect pathway remains to be determined. On one hand, stimulation of the cortical D1 receptor enhances contextual salience through connections to the accumbens [76], as exemplified by increased drug-seeking to drugassociated environments. However, excessive D1 activation reduces cortical activity and takes it off-line [82], possibly leading to reduced motivation and poor decision making. The majority of research on the adolescent dopamine system has focused on appetitive processes, with little emphasis on the relationship between dopamine and aversives. Although this remains a fruitful area of research, it is tangible that adolescent changes within the mesocorticolimbic dopamine system increase susceptibility to stress-related depression and anhedonia during this vulnerable period.

Trends in Neurosciences Vol.31 No.4

genetic factors, the severity, timing and nature of the exposure and the presence of moderating factors, such as degree of parental support or involvement. Maturational events Figure 1 provides a comparison of trajectories of gray matter development in key regions associated with depression from childhood through early adulthood. Three sets of developmental factors operate in the genesis of adolescent depression. The rst are typical adolescent changes in brain maturation, including anatomical and functional rearrangements, sensitivity to gonadal and adrenal hormones, and increased psychosocial pressures. The second are early windows of vulnerability, or sensitive periods, when specic regions of the developing brain might be most susceptible to environmental inuences that have the potential to increase risk for depression. The third are the key maturational changes taking place during adolescence that lead to the overt expression of the disorder in individuals with an underlying predisposition. The developing brain undergoes a period of overproduction and pruning of synapses and signaling mechanisms between childhood, adolescence and young adulthood [17]. Windows of vulnerability potentially occur during periods of very rapid development, and synaptic pruning during adolescence might unmask underlying predispositions. Regional differences in the trajectory of synaptic development, programming of neurotrophic factor levels, connectivity between brain regions, rates of myelination and increased expression of glucocorticoid receptors potentially result in brain region-specic windows of vulnerability at different ages (Figure 2) and an overall increased sensitivity to depression onset during adolescence. The process is also sexually dimorphic, and males typically overproduce synapses and signaling mechanisms to a greater extent than females [18,19]. On average, gray matter density in the cortex peaks 12 years earlier in girls than boys (11.2 versus 12.6 years). Fetal exposure to gonadal hormones exerts organizing effects responsible for sex differences in brain development [20]. Pubertal exposure activates hormones that modulate the development of the prefrontal cortex, amygdala and hypothalamus. Evidence for the role of gonadal hormones in this process in humans is still relatively sparse, but inferred from a nding in males with congenital adrenal hypoplasia (resulting in increased prenatal testosterone) and in individuals with XXY genotype [21]. Rodent studies suggest that estrogen suppresses neuronal overproduction in the female prefrontal cortex [22,23], whereas rising levels of testosterone aid in pruning of dendrites within the adolescent male amygdala [24]. Hence, adolescence is associated with sexually dimorphic pruning of synapses and signaling mechanisms in brain regions implicated in depression. The emergence of depression during adolescence might result, in part, from either insufcient overproduction [25] or enhanced pruning of these brain regions. Estrogenic effects might further exacerbate these processes. Early studies that examined the role of pubertal hormones in depression failed to show a signicant relationship between rising gonadal steroids and depressive

adversity (i.e. childhood abuse, parental loss or chaotic household) account for 54% of the population attributable risk for current depression and 67% of the risk for suicide attempts [15]. Most importantly, genes and early experience interact. Caspi et al. [16] found that exposure to severe childhood maltreatment (between 3 and 11 years of age) doubled the risk of major depression in individuals with two copies of the short allele promoter polymorphism of the gene encoding the serotonin transporter (5-HTT). By contrast, childhood maltreatment produced no increase in risk in individuals with two copies of the long allele polymorphism. In short, many cases of depression might arise from the conuence of genetic predisposition interacting with environmental experiences that occur during a specic window of vulnerability. This, in turn, sets up a cascade of events that unfold over the course of maturation. Early stress exposure, however, is not specically linked to the development of depression, but is also a predisposing factor that can lead to the emergence of posttraumatic stress disorder (PTSD), substance abuse, personality disorders and aggression. The outcome will likely depend on
184

Review

Trends in Neurosciences

Vol.31 No.4

Figure 1. Brain development proceeds in a series of stages that vary across regions. Shown here are different gray matter changes of cortical and subcortical brain regions across age in humans; graph based on longitudinal data from Refs [32,73,74], with permission from Jay Giedd. Hippocampal volumes are 85% of adult values by the first few years of life. Amygdala data are from a cross-sectional analysis of n = 99 subjects, collapsed across gender [75]. Comparatively, this process occurs in all mammalian species, including humans, primates and rats at relatively the same developmental milestones, including the onset of puberty and higher-level cognition.

symptoms [26]. Recently, however, a population-based study demonstrated a linear relationship between depressive symptoms and estrogen (or estrogen + testosterone) in a representative sample of girls (n = 973) from 9 to 13 years of age. Gonadal hormones are likely to mediate their effects indirectly through GABA, 5-HT and/or dopamine systems (e.g. [27,28]), both of which are involved in depression and anxiety. Stress-related hormones, the glucocorticoids (GC) and mineralocorticoids, play an important role in sculpting the adolescent brain (reviewed in Ref. [29]). They aid in programming adaptive function for survival through longterm potentiation and/or synaptic selection. This function might occur via epigenetic mechanisms or through the regulated expression of several genes, including that encoding BDNF. By contrast, glucocorticoids can adversely affect the brain in two major ways. Maladaptive secretion of these hormones can alter trajectories of development that are predisposing to the emergence of psychopathology. Abundant preclinical evidence shows that increased glucocorticoid secretion reduces neurogenesis and synaptogenesis, especially in the hippocampus. Further, glucocorticoids can directly affect brain function, provoking emotional lability [30]. Mood is regulated by interactions between cortical and limbic regions (Figure 2). As these pathways mature during adolescence, they are impacted by exposure to gonadal and adrenal hormones. Episodes of melancholic depression are often associated with increased secretion of glucocorticoids that might potentially suppress hippocampal neurogenesis. Selective serotonin reuptake inhibitors (SSRIs) increase hippocampal neurogenesis in animal models of depression in parallel with symptom improvement [31]. These drugs also

block progressive loss of hippocampal volume associated with duration of depression. However, whether increased glucocorticoid secretion causes or is the consequence of emotional dysregulation has been difcult to show [32]. Adolescents can be particularly vulnerable to stress as a consequence of heightened glucocorticoid receptor expression in the cortex [29] and more prolonged and exacerbated corticosterone response following acute stress [33]. Adolescent female rats have a greater corticosteroid response to acute stress than males [33], which might explain their increased susceptibility to learned helplessness following acute stress [34]. Increased vulnerability to stress might be a key factor in rising rates of depression during adolescence, as an acute stressor can precipitate the onset of depressive episodes in individuals primed to develop the disorder. Interestingly, male adolescent rats are particularly susceptible to repeated episodes of social stress [34]. This vulnerability appears to wane by adulthood (R. Romeo, personal communication). Effects of early stress on brain development evidence for sensitive periods Early stress stemming from childhood adversity is a predisposing risk factor for the development of major depression [15], and depression is the most common adult sequela of early abuse [35]. Whether exposure to adversity leads to the eventual development of depression is likely to be determined by genetic susceptibilities [16], frequency, severity and multiplicity of the stressors [16,36], gender and timing of insult [37]. Exposure to physical or sexual abuse resulting in psychopathology has been associated with changes in brain structure (reviewed in Ref. [36]) that include attenuated left hemisphere maturation, dimin185

Review

Trends in Neurosciences Vol.31 No.4

Figure 2. Stress-vulnerable brain circuits involved in emotional processing. In this review, we highlight the role of the hippocampus, amygdala, prefrontal cortex and nucleus accumbens in mood regulation during adolescence. Connectivity between brain regions might also influence the emergence of adolescent depression as corticolimbic circuits become functional. The green arrow shows the maturing connections between the amygdala and prefrontal cortex [72], whereas the blue arrow shows hippocampal-to-prefrontal cortex connections. The development of these connections is not well studied, nor are connections emanating from the prefrontal cortex to these regions (not shown), which are important for cognitive control over mood. Connections between the prefrontal cortex and nucleus accumbens (black) are still developing [76]. (Insets) The expression of glucocorticoid receptors changes across development in a regional-specific manner in the human hippocampus and frontal cortex (dorsolateral is shown here; adapted from Ref. [77]). Glucocorticoid receptors are abundant in the rat amygdala, but their density is not characterized in humans [29]. These receptors, along with gonadal steroids and genetic predisposition, play a role in the unique sensitive periods of vulnerability that exist for these regions.

ished size of the corpus callosum, reduced hippocampal volume in adults (but not in children), alterations in gray matter volume (GMV), symmetry and neuronal integrity of the frontal cortex and reduced size of the anterior cingulate cortex and caudate nucleus [38]. Not surprisingly, many of these regions are involved in emotional regulation, and similar abnormalities have been observed in individuals with depression [39,40]. Notably, this is a relatively new area of inquiry. To date, more than a dozen cross-sectional studies have reported morphological differences, but they
186

do not provide evidence for a cause-and-effect relationship. Only one small, longitudinal study (n = 15 subjects) shows that maltreatment-induced stress on cortisol levels and PTSD symptomatology correlates with changes in the trajectory of hippocampal development [41]. Here we emphasize structural changes that are associated with exposure to early stress. Several neurochemical and molecular changes likely accompany these morphometric differences and might serve to more directly tie early stress exposure to increased risk for depression.

Review
Preclinical studies have shown that early stress is associated with enduring decits in 5-HT turnover in the amygdala and nucleus accumbens [42], decreased density of dopamine transporters in the striatum and accumbens, alteration of GABAA receptor subunit expression in the prefrontal cortex, hippocampus and amygdala, and a2 receptor changes in the locus coeruleus and nucleus tractus solitarius [43,44]. At this stage, however, human imaging studies have almost exclusively depended on morphometry to identify regional alterations associated with developmental stress. Generally, early onset and longer duration of abuse are linked with greater morphological change [37], but this might be an oversimplication. An alternative hypothesis is that stress-susceptible brain regions have unique sensitive periods (or windows of vulnerability) to the effects of early stress [45]. Young adults who experienced childhood sexual abuse (CSA) between either 35 or 1113 years of age demonstrate maximal reductions in hipppocampal volume [46]. CSA between 9 and 10 years is associated with reduced corpus callosum size. By contrast, CSA between ages 14 and 16 years was associated with reduced frontal cortex GMV [46]. Preclinical studies in rats nd parallel changes [25,47]. Taken together, mammalian brain regions have unique windows of vulnerability to the effects of stress (Figure 3). These ndings raise the possibility that different abuserelated syndromes might be associated with particular ages of abuse and regional brain changes. For example, current symptoms of depression were associated with abuse between 3 and 6 years of age. PTSD symptoms were associated with abuse that occurred between 9 and 10 years [46]. Early exposure to childhood adversity might initiate a cascade of consequences leading to hippocampal morphological abnormalities and clinical signs of depression. Maturation and onset of depression Depression is the most extensively documented outcome of exposure to CSA in adults but is not a common occurrence in children [35]. This suggests that a substantial time lag, or incubation period, occurs between early stress exposure and eventual development of depression. In a prospective study, rates of depression were increased by physical abuse or neglect (n = 676), which lead to an earlier emergence of depression relative to controls (n = 520) [48]. In a separate cohort exposed to CSA but no other forms of abuse, the mean survival time from onset of abuse to emergence of depression was 11.47 years overall. There was a delay of 9.2 3.6 years in subjects who developed depression (62%). Notably, most rst episodes of depression occurred between 12 and 15 years of age. These ndings indicate that early stress not only increases risk of developing depression, it accelerates onset into early adolescence. Depression in individuals not exposed to abuse or neglect often emerges later in adulthood. However, $20% of these depressed individuals experienced their rst episode in mid to late adolescence. Hence, the respective proportion of depressed individuals with early adversity would be maximal in childhood and early adolescence

Trends in Neurosciences

Vol.31 No.4

Figure 3. Effect of stress exposure on the size of the hippocampus and prefrontal cortex in young adults. Plotted are the effect sizes (Cohens d) for each comparison with 95% confidence intervals; effects in the negative direction indicate that the brain region was smaller in the stressed subjects relative to controls. (Top) Magnetic resonance imaging morphometric data in humans exposed to childhood sexual abuse (CSA) between 3 and 5 years of age (early) or exposed between 14 and 16 years of age (late). (Bottom) Synaptic density, as measured by synaptophysin optical density, in Sprague-Dawley rats exposed to maternal separation between postnatal days 2 and 20 (early) or a social stressor between 30 and 35 days of age. **p < 0.001. Data are adapted from Refs [25,46,47].

and begin to decline by mid to late adolescence. This is particularly important, as depressed individuals with a history of abuse or early loss can be less responsive to pharmacotherapy than depressed individuals without such histories [49]. Delayed effects of early stress on the hippocampus The role of the hippocampus in adolescent depression is highly likely, given its susceptibility to the effects of stress
187

Review
[50] and its role in mood regulation and antidepressant response [51]. Reduced hippocampal size has been observed in most studies of adults with major depression [39], and can vary as a function of episode number [39] or duration of time spent depressed but untreated [32]. We propose that the hippocampus contributes to the burden of depression by failing to provide an appropriate contextual response to affectively laden stimuli. This is consistent with ndings of perturbed encoding of fearful faces in depressed children and adolescents [52]. Few studies have examined hippocampal volume in adolescents with depression. However, a substantial left > right reduction in hippocampal volume was observed in older adolescents (mean 16.7 years) with major depression (n = 17 depressed versus n = 17 controls) [53]. Increased amygdala:hippocampal ratios were also found in pediatric depression (mean age 14 years; n = 23) versus controls (n = 23), although this difference might be more directly associated with degree of comorbid anxiety [54]. The relationship between exposure to early maltreatment and hippocampal volume is intriguing. Five studies (total n = 209) have reported reduced hippocampal volume in adults with history of exposure to childhood abuse [46,55], whereas three studies (total n = 186) failed to observe reduced hippocampal volume in children with PTSD and comparable abuse histories [56,57]. Animal models can increase our understanding of this phenomenon, as they show that exposure to early stress results in a 34%36% relative reduction in synaptic density in the hippocampus that emerged between puberty and early adulthood [25]. Thus, it is conceivable that exposure to early adversity during a window of vulnerability delays the expression of depression in susceptible individuals by setting into motion a series of events that affects processes regulating synaptic overproduction in the hippocampus. Late-maturing prefrontal cortex Multiple components of the prefrontal cortex (PFC) are involved in the emergence of depression as it matures during the periadolescent period. A primary role of the PFC is to modulate activity of limbic structures. The PFC has a protracted postnatal ontogeny and does not attain adult volume until the early 20s [58]. The complexity of the region and its developmental time course render it vulnerable, especially to stress during adolescence [46]. PFC development is also affected by early pathology in other brain regions such as the hippocampus or striatum. As the functional properties of the PFC emerge sequentially throughout development, early PFC pathology might be silent until the PFC would normally subsume control of the affected abilities. For these reasons, the PFC occupies a prominent place in theories of adolescent depression [59,60]. For example, the triadic model of Ernst et al. [60] postulates that adolescent depression emerges as limbic structures driving affect mature in advance of cortical structures providing regulatory control. However, this transient developmental mismatch does not account for the persistently elevated incidence of depression in adulthood. Major depression most likely emerges in some individuals as a consequence of abnormalities in PFC development.
188

Trends in Neurosciences Vol.31 No.4

This neuropathogenic view is compatible with adult observations of decreased orbitofrontal volume in patients with remitted major depression [61], and autopsy studies showing a marked reduction in the density and size of neurons and glia in dorsolateral and orbital PFC [62]. Although few morphometric or functional imaging studies have been conducted in depressed adolescents, the ndings are generally consistent with those in adults. For example, reduced left subgenual cingulate volume was comparable between adolescents (n = 30) and middle-aged women (n = 18) with depression [63]. Similarly, reduced volume in right medial frontal gyrus and the anterior cingulate were observed in a preliminary analysis of depressed teens (total n = 16) (M.H.T. et al., unpublished). Depressed adolescents had increased resting relative cerebral blood volume (rCBV) in left orbital and dorsolateral PFC and right subgenual cingulate (BA 25) using T2 relaxometry. Depressive symptoms correlated with diminished rCBV in left and elevated rCBV in right dorsolateral PFC. Despite similarities between depressed adolescents and adults, elevated indices of right subgenual cingulate rCBV parallel ndings in treatment-refractory depressed adults [64]. This nding is consistent with the suboptimal response of depressed adolescents to many antidepressants (see Box 2). Exposure to childhood stress might predispose to depression over a long incubation period by altering hippocampal development. By contrast, exposure to stress during adolescence might precipitate depression over a shorter incubation period by directly affecting the PFC. Exposure to a stressful event between 14 and 16 years activates the PFC and is associated with an 8% synaptic loss by young adulthood [46]. Synaptic loss was also

Box 2. Efficacy of antidepressants and assessment tests

Childhood and adolescent depression appear to be less responsive to antidepressant treatment than adult depression. Fewer drugs exert significantly better effects than placebo in pediatric depression, and beneficial drugs can produce a weaker response (e.g. 61% versus 50% drug:placebo response to SSRIs in children versus 78%:44% response in adults) [83]. All antidepressants now carry, in the USA, black-box warnings of treatment-emergent suicidality in children, adolescents and young adults. Treatment is often indicated, as depression causes great suffering and can markedly interfere with academic and social development. Suicide is a leading cause of death in teenagers. Clinicians must weigh the importance of acute treatment of symptoms against the potential for unknown long-term side effects. Standard screening tests for antidepressant efficacy in animal models are inescapable shock and the forced swim test [84]. These tests are stress based and predicated on the learned helplessness model of Seligman and Maier [85]. Rats placed into inescapable stressful situations often show helplessness on subsequent exposure even when escape is possible. Antidepressants can reverse this effect. In adults, these tests have led to successful identification of antidepressant agents that have good clinical efficacy [84]. However, the use of these tests to identify new, effective agents for the treatment of child and adolescent depression is limited. There is a pressing need for antidepressant drugs with increased efficacy in children and adolescents. Specifically screening for novel agents with potential utility in pediatric depression might eventually be necessary, given the suboptimal response seen so far to drugs designed to target depression in adults.

Review
observed in an adolescent model of social stress in rats, and depressive symptoms manifested immediately [47]. Depressive episodes in adolescents often occur within a year of exposure to one or more signicantly stressful life events [65], and might be the result of stress-induced alterations in prefrontal development. The amygdala: fear, irritability and anxiety The amygdala has become an increasingly important component of theories of depression, based largely on imaging studies of mood regulation. The amygdala is over-responsive to fearful stimuli in depression [66], and insufciently regulated by PFC controls [67]. This leads to an excessive and persistent degree of negative affectivity, which is a signicant but nonspecic feature of depression. Alteration in the genes encoding 5-HTT and 5HT1 is associated with augmented amygdala response to emotional stimuli [68], which might underlie observed decreases in amygdala volume [69]. Onset of depression in adolescence is often preceded by social anxiety disorder [4], which might be related to overactivation of the amygdala. Interestingly, social anxiety is typically an early-onset adolescent disorder, with the rare new case occurring after age 25. Early emergence of social anxiety can result from immature integration of cortical and limbic components into expression of affective states [70]. With increasing maturation there is an enhanced degree of cortical involvement in processing response to fearful faces [71], which provides a potential means of checking or correcting an excessive amygdaloid response. This is supported by preclinical studies showing delayed development of connectivity between basolateral amygdala and medial PFC [72]. Conclusions Recent studies have emphasized the importance of gene x environment in the genesis of depression. Time is another crucial factor, both in terms of windows of vulnerability when brain regions might be maximally sensitive to environmental inuences and in the cascade of maturational events that lead to the unfolding of depression. The hippocampus and PFC can have early and late windows of vulnerability, and serve respectively as targets for the predisposing and precipitating effects of stress. Alterations in the amygdala and nucleus accumbens might contribute, in turn, to the negative and positive affectivity symptoms seen in depression. Together, this model provides a neurobiological framework that potentially accounts for the emergence of depression and gender differences during adolescence.
Acknowledgements
This work was supported, in part, by awards from NARSAD (2001, 2002, 2005), NIDA (RO1DA-017846), NIMH (RO1MH-66222) and the Simches and Rosenberg Families.

Trends in Neurosciences

Vol.31 No.4

References
1 Emslie, G.J. et al. (2005) Continuation and maintenance therapy of early-onset major depressive disorder. Paediatr. Drugs 7, 203217 2 Kessler, R.C. et al. (2001) Mood disorders in children and adolescents: an epidemiologic perspective. Biol. Psychiatry 49, 10021014

3 Hankin, B.L. et al. (1998) Development of depression from preadolescence to young adulthood: emerging gender differences in a 10-year longitudinal study. J. Abnorm. Psychol. 107, 128140 4 Beesdo, K. et al. (2007) Incidence of social anxiety disorder and the consistent risk for secondary depression in the rst three decades of life. Arch. Gen. Psychiatry 64, 903912 5 Williamson, D.E. et al. (2005) A genetic epidemiologic perspective on comorbidity of depression and anxiety. Child Adolesc. Psychiatr. Clin. N. Am. 14, 707726 6 Johnson, E.O. et al. (2006) The association of insomnia with anxiety disorders and depression: exploration of the direction of risk. J. Psychiatr. Res. 40, 700708 7 Zisook, S. et al. (2007) Preadult onset vs. adult onset of major depressive disorder: a replication study. Acta Psychiatr. Scand. 115, 196205 8 Sorensen, M.J. et al. (2005) Age and gender differences in depressive symptomatology and comorbidity: an incident sample of psychiatrically admitted children. J. Affect. Disord. 84, 8591 9 Teicher, M.H. et al. (1993) Locomotor activity in depressed children and adolescents: I. Circadian dysregulation. J. Am. Acad. Child Adolesc. Psychiatry 32, 760769 10 Wade, T.J. et al. (2002) Emergence of gender differences in depression during adolescence: national panel results from three countries. J. Am. Acad. Child Adolesc. Psychiatry 41, 190198 11 Patton, G.C. et al. (1996) Menarche and the onset of depression and anxiety in Victoria, Australia. J. Epidemiol. Community Health 50, 661666 12 Silverstein, B. (1999) Gender difference in the prevalence of clinical depression: the role played by depression associated with somatic symptoms. Am. J. Psychiatry 156, 480482 13 Bennett, D.S. et al. (2005) Gender differences in adolescent depression: do symptoms differ for boys and girls? J. Affect. Disord. 89, 3544 14 Glowinski, A.L. et al. (2003) Genetic epidemiology of self-reported lifetime DSM-IV major depressive disorder in a population-based twin sample of female adolescents. J. Child Psychol. Psychiatry 44, 988996 15 Chapman, D.P. et al. (2004) Adverse childhood experiences and the risk of depressive disorders in adulthood. J. Affect. Disord. 82, 217225 16 Caspi, A. et al. (2003) Inuence of life stress on depression: moderation by a polymorphism in the 5-HTT gene. Science 301, 386389 17 Andersen, S.L. (2003) Trajectories of brain development: point of vulnerability or window of opportunity? Neurosci. Biobehav. Rev. 27, 318 18 Giedd, J.N. et al. (1997) Sexual dimorphism of the developing human brain. Prog. Neuropsychopharmacol. Biol. Psychiatry 21, 11851201 19 Andersen, S.L. et al. (2002) Pubertal changes in gonadal hormones do not underlie adolescent dopamine receptor overproduction. Psychoneuroendocrinology 27, 683691 20 Lenroot, R.K. et al. (2007) Sexual dimorphism of brain developmental trajectories during childhood and adolescence. Neuroimage 36, 1065 1073 21 Giedd, J.N. et al. (2006) Puberty-related inuences on brain development. Mol. Cell. Endocrinol. 254255, 154162 22 Juraska, J.M. and Markham, J.A. (2004) The cellular basis for volume changes in the rat cortex during puberty: white and gray matter. Ann. N. Y. Acad. Sci. 1021, 431435 23 Markham, J.A. et al. (2007) Neuron number decreases in the rat ventral, but not dorsal, medial prefrontal cortex between adolescence and adulthood. Neuroscience 144, 961968 24 Zehr, J.L. et al. (2006) Dendritic pruning of the medial amygdala during pubertal development of the male Syrian hamster. J. Neurobiol. 66, 578590 25 Andersen, S.L. and Teicher, M.H. (2004) Delayed effects of early stress on hippocampal development. Neuropsychopharmacology 29, 1988 1993 26 Angold, A. and Costello, E.J. (2006) Puberty and depression. Child Adolesc. Psychiatr. Clin. N. Am. 15, 919937 27 Robichaud, M. and Debonnel, G. (2005) Oestrogen and testosterone modulate the ring activity of dorsal raphe nucleus serotonergic neurones in both male and female rats. J. Neuroendocrinol. 17, 179185 28 Birzniece, V. et al. (2006) Neuroactive steroid effects on cognitive functions with a focus on the serotonin and GABA systems. Brain Res. Rev. 51, 212239

189

Review
29 Pryce, C.R. Postnatal ontogeny of expression of the corticosteroid receptor genes in mammalian brains: inter-species and intra-species differences. Brain Res. Rev. (in press) 30 Wei, Q. et al. (2004) Glucocorticoid receptor overexpression in forebrain: a mouse model of increased emotional lability. Proc. Natl. Acad. Sci. U. S. A. 101, 1185111856 31 Gross, C. and Hen, R. (2004) Genetic and environmental factors interact to inuence anxiety. Neurotox. Res. 6, 493501 32 Sheline, Y.I. et al. (2003) Untreated depression and hippocampal volume loss. Am. J. Psychiatry 160, 15161518 33 Romeo, R.D. and McEwen, B.S. (2006) Stress and the adolescent brain. Ann. N. Y. Acad. Sci. 1094, 202214 34 Leussis, M.P. and Andersen, S.L. (2008) Is adolescence a sensitive period for depression? Behavioral and neuroanatomical ndings from a social stress model. Synapse 62, 2230 35 Putnam, F.W. (2003) Ten-year research update review: child sexual abuse. J. Am. Acad. Child Adolesc. Psychiatry 42, 269278 36 Teicher, M.H. et al. (2006) Sticks, stones, and hurtful words: relative effects of various forms of childhood maltreatment. Am. J. Psychiatry 163, 9931000 37 De Bellis, M.D. et al. (1999) Developmental traumatology. Part II: Brain development. Biol. Psychiatry 45, 12711284 38 Cohen, R.A. et al. (2006) Early life stress and morphometry of the adult anterior cingulate cortex and caudate nuclei. Biol. Psychiatry 59, 975 982 39 Videbech, P. and Ravnkilde, B. (2004) Hippocampal volume and depression: a meta-analysis of MRI studies. Am. J. Psychiatry 161, 19571966 40 Ballmaier, M. et al. Mapping callosal morphology in early- and lateonset elderly depression: an index of distinct changes in cortical connectivity. Neuropsychopharmacology (in press) 41 Carrion, V.G. et al. (2007) Stress predicts brain changes in children: a pilot longitudinal study on youth stress, posttraumatic stress disorder, and the hippocampus. Pediatrics 119, 509516 42 Andersen, S.L. et al. (1999) Enduring neurochemical effects of early maternal separation on limbic structures. Ann. N. Y. Acad. Sci. 877, 756759 43 Caldji, C. et al. (1998) Maternal care during infancy regulates the development of neural systems mediating the expression of fearfulness in the rat. Proc. Natl. Acad. Sci. U. S. A. 95, 53355340 44 Caldji, C. et al. (2003) Variations in maternal care alter GABA(A) receptor subunit expression in brain regions associated with fear. Neuropsychopharmacology 28, 19501959 45 Teicher, M.H. et al. (2006) Neurobiological consequences of early stress and childhood maltreatment: are results from human and animal studies comparable? Ann. N. Y. Acad. Sci. 1071, 313323 46 Andersen, S.L. et al. Preliminary evidence for sensitive periods in the effect of childhood sexual abuse on regional brain development. J. Neuropsychiatry Clin. Neurosci. (in press) 47 Leussis, M.P. et al. (2008) The enduring effects of an adolescent social stressor on synaptic density. Part II: Post-stress reversal of synaptic loss in the cortex by adinazolam and MK-801. Synapse 62, 185192 48 Widom, C.S. et al. (2007) A prospective investigation of major depressive disorder and comorbidity in abused and neglected children grown up. Arch. Gen. Psychiatry 64, 4956 49 Nemeroff, C.B. et al. (2003) Differential responses to psychotherapy versus pharmacotherapy in patients with chronic forms of major depression and childhood trauma. Proc. Natl. Acad. Sci. U. S. A. 100, 1429314296 50 McEwen, B.S. (2000) Effects of adverse experiences for brain structure and function. Biol. Psychiatry 48, 721731 51 Perera, T.D. et al. (2007) Antidepressant-induced neurogenesis in the hippocampus of adult nonhuman primates. J. Neurosci. 27, 48944901 52 Pine, D.S. et al. (2004) Face-memory and emotion: associations with major depression in children and adolescents. J. Child Psychol. Psychiatry 45, 11991208 53 MacMaster, F.P. and Kusumakar, V. (2004) Hippocampal volume in early onset depression. BMC Med. 2, 2 54 MacMillan, S. et al. (2003) Increased amygdala:hippocampal volume ratios associated with severity of anxiety in pediatric major depression. J. Child Adolesc. Psychopharmacol. 13, 6573

Trends in Neurosciences Vol.31 No.4

55 Vythilingam, M. et al. (2002) Childhood trauma associated with smaller hippocampal volume in women with major depression. Am. J. Psychiatry 159, 20722080 56 Carrion, V.G. et al. (2001) Attenuation of frontal asymmetry in pediatric posttraumatic stress disorder. Biol. Psychiatry 50, 943 951 57 De Bellis, M.D. et al. (2002) Brain structures in pediatric maltreatment-related posttraumatic stress disorder: a sociodemographically matched study. Biol. Psychiatry 52, 10661078 58 Giedd, J.N. (2004) Structural magnetic resonance imaging of the adolescent brain. Ann. N. Y. Acad. Sci. 1021, 7785 59 Davey, C.G. et al. (2008) The emergence of depression in adolescence: development of the prefrontal cortex and the representation of reward. Neurosci. Biobehav. Rev. 32, 119 60 Ernst, M. et al. (2006) Triadic model of the neurobiology of motivated behavior in adolescence. Psychol. Med. 36, 299312 61 Bremner, J.D. (2002) Neuroimaging of childhood trauma. Semin. Clin. Neuropsychiatry 7, 104112 62 Jakovcevic, D. and Harder, D.R. (2007) Role of astrocytes in matching blood ow to neuronal activity. Curr. Top. Dev. Biol. 79, 7597 63 Botteron, K.N. et al. (2002) Volumetric reduction in left subgenual prefrontal cortex in early onset depression. Biol. Psychiatry 51, 342 344 64 Mayberg, H.S. et al. (2005) Deep brain stimulation for treatmentresistant depression. Neuron 45, 651660 65 Williamson, D.E. et al. (1998) Nature of life events and difculties in depressed adolescents. J. Am. Acad. Child Adolesc. Psychiatry 37, 10491057 66 Roberson-Nay, R. et al. (2006) Increased amygdala activity during successful memory encoding in adolescent major depressive disorder: an FMRI study. Biol. Psychiatry 60, 966973 67 Fales, C.L. et al. (2008) Altered emotional interference processing in affective and cognitive-control brain circuitry in major depression. Biol. Psychiatry 63, 377384 68 Dannlowski, U. et al. (2008) 5-HTTLPR biases amygdala activity in response to masked facial expressions in major depression. Neuropsychopharmacology 33, 418424 69 Rosso, I.M. et al. (2005) Amygdala and hippocampus volumes in pediatric major depression. Biol. Psychiatry 57, 2126 70 Phillips, A.G. et al. (2003) Amygdalar control of the mesocorticolimbic dopamine system: parallel pathways to motivated behavior. Neurosci. Biobehav. Rev. 27, 543554 71 Yurgelun-Todd, D.A. and Killgore, W.D. (2006) Fear-related activity in the prefrontal cortex increases with age during adolescence: a preliminary fMRI study. Neurosci. Lett. 406, 194199 72 Cunningham, M.G. et al. (2002) Amygdalo-cortical sprouting continues into early adulthood: implications for the development of normal and abnormal function during adolescence. J. Comp. Neurol. 453, 116130 73 Lyss, P.J. et al. (1999) Degree of neuronal activation following FG-7142 changes across regions during development.. Brain Res. Dev. Brain Res. 116, 201203 74 Gogtay, N. et al. (2006) Dynamic mapping of normal human hippocampal development. Hippocampus 16, 664672 75 Giedd, J.N. et al. (1996) Quantitative MRI of the temporal lobe, amygdala, and hippocampus in normal human development: ages 418 years. J. Comp. Neurol. 366, 223230 76 Brenhouse, H. et al. Transient D1 dopamine receptor over-expression on prefrontal cortex projection neurons: a mechanism for enhanced motivational salience of drug cues in adolescence. J. Neurosci. (in press) 77 Perlman, W.R. et al. (2007) Age-related differences in glucocorticoid receptor mRNA levels in the human brain. Neurobiol. Aging 28, 447 458 78 Nestler, E.J. and Carlezon, W.A., Jr (2006) The mesolimbic dopamine reward circuit in depression. Biol. Psychiatry 59, 11511159 79 Dunlop, B.W. and Nemeroff, C.B. (2007) The role of dopamine in the pathophysiology of depression. Arch. Gen. Psychiatry 64, 327337 80 Andersen, S.L. and Gazzara, R.A. (1993) The ontogeny of apomorphineinduced alterations in dopamine release: I. Effects on spontaneous release. J. Neurochem. 61, 22472255 81 Spear, L.P. (2000) The adolescent brain and age-related behavioral manifestations. Neurosci. Biobehav. Rev. 24, 417463

190

Review
82 Zahrt, J. et al. (1997) Supranormal stimulation of D1 dopamine receptors in the rodent prefrontal cortex impairs spatial working memory performance. J. Neurosci. 17, 85288535 83 Bridge, J.A. et al. (2007) Clinical response and risk for reported suicidal ideation and suicide attempts in pediatric antidepressant treatment: a meta-analysis of randomized controlled trials. JAMA 297, 16831696

Trends in Neurosciences

Vol.31 No.4

84 Cryan, J.F. et al. (2002) Assessing antidepressant activity in rodents: recent developments and future needs. Trends Pharmacol. Sci. 23, 238245 85 Seligman, M.E. and Maier, S.F. (1967) Failure to escape traumatic shock. J. Exp. Psychol. 74, 19

Have you contributed to an Elsevier publication? Did you know that you are entitled to a 30% discount on books?
A 30% discount is available to all Elsevier book and journal contributors when ordering books or stand-alone CD-ROMs directly from us. To take advantage of your discount: 1. Choose your book(s) from www.elsevier.com or www.books.elsevier.com 2. Place your order Americas: Phone: +1 800 782 4927 for US customers Phone: +1 800 460 3110 for Canada, South and Central America customers Fax: +1 314 453 4898 author.contributor@elsevier.com All other countries: Phone: +44 (0)1865 474 010 Fax: +44 (0)1865 474 011 directorders@elsevier.com Youll need to provide the name of the Elsevier book or journal to which you have contributed. Shipping is free on prepaid orders within the US. If you are faxing your order, please enclose a copy of this page. 3. Make your payment This discount is only available on prepaid orders. Please note that this offer does not apply to multi-volume reference works or Elsevier Health Sciences products.

For more information, visit www.books.elsevier.com

191