HANDBOOK OF
`
`
`ESSEON
`
`Second Edition
`
`Edited by
`
`IAN H. GOTLIB
`
`CONSTANCE L. HAMMEN
`
`{332
`
`THE GUILFORD PRESS
`
`New York
`
`London
`
`TiitshmiiargalgasEégied
`Subj:«:s1U5::py:?::t Saws
`
`Merck 2002
`Merck 2002
`Argentum v. Merck
`Argentum V. Meer
`IPR2018-00423
`IPR2018'00423
`
`

`

`50AM Guilford Press
`A \Di-viém of Guilford Publications, Inc.
`72 Spring Street, New York, NY 10012
`www.guilford.com
`
`All rights reserved
`
`No part of this book may be reproduced, translated, stored in
`a retrieval system, or transmitted, in any form or by any means,
`electronic, mechanical, photocopying, microfilming, recording,
`or otherwise, without written permission from the Publisher.
`
`Printed in the United States of America
`
`This book is printed on acid-free paper.
`
`Last digit is print number:
`
`9
`
`8
`
`7
`
`6
`
`5
`
`4
`
`3
`
`2
`
`1
`
`The authors have checked with sources believed to be reliable in their efforts to provide information
`that is complete and generally in accord with the standards of practice that are accepted at the time
`of publication. However, in View of the possibility of human error or changes in medical sciences,
`neither the authors, nor the editor and publisher, nor any other party who has been involved in the
`preparation or publication of this work warrants that the information contained herein is in every
`respect accurate or complete, and they are not responsible for any errors or omissions or the results
`obtained from the use of such information. Readers are encouraged to confirm the information
`contained in this book with other sources.
`
`Library of Congress Cataloging-in-Publication Data
`
`Handbook of depression / edited by Ian H. Gotlib, Constance L. Hammen. —- 2nd ed.
`p.
`; cm.
`Includes bibliographical references and index.
`ISBN 978-1-59385—450-8 (hardcover : alk. paper)
`1. Depression, Mental—Handbooks, manuals, etc.
`[DNI,M: 1. Depressive Disorder.
`2. Depression.
`RC537.H3376 2009
`616.85'27—dc22
`
`II. Hammen, Constance 1..
`I. Gotlib, Ian H.
`3. Risk Factors. WM 171 HZ367 2008]
`
`2008010575
`
`This material was copied
`atthe NLM and maybe
`Subject US {lapyright Laws
`
`

`

`CHAPTER 9
`
`Neurobiological Aspects of Depression
`
`Michael E. Thase
`
`Since antiquity, there have been speculations about the biological basis of depres—
`,
`Slon- To take but one example, the term melancholia, which is currently used to describe. one
`Of the most severe forms of depression, reflects the ancient Greek theory that mood disor—
`ders Were caused by an imbalance of black bile (Jackson, 1986). Only during thepast 5.0
`years, however, has the methodology been available to study directly alterations in brain
`funCtiOn associated with depression. What has emerged from this half~century of research
`has been an iterative and evolving process, answering some questions and opening new .and
`more SOphisticated lines of inquiry. One certainty is that the heterogeneous conditions
`grouped together under the construct of clinical depression are biopsychosoc1al disorders
`that\Inuch more often than not—have multifactorial causality.
`My colleagues and I reviewed evidence pertaining to neurobiological disturbances asso-
`ciated With depression in the previous volume of this Handbook, including a Wide range of
`neurOChemical, neuroendocrine, neurophysiological, and neuroanatomical parameters (Thase,
`Jindal, 8C Howland, 2002). Over the past two decades, various hypotheses have been ad-
`vanced’ tested, and either rejected or modified as research paradigms have evolved and
`knowledge about the function of the central nervous system (CNS) in health, in disease, and
`in 1response to various states of duress has grown. A number of new hypotheses also have
`been advanced. Some research tools, such as measurement of catecholamine metabolites in
`urine, blood, and cerebrospinal fluid (CSF) or electrophysiological recordings of neuronal
`activity, that were de rigueur in the 1970s and 19803 are now seldom used, others that were
`not technologically feasible, such as functional magnetic resonance imaging (fIVIRI), p051-
`tron emission tomography (PET) imaging of receptor binding, and fast through-put geno-
`typing, are now commonplace.
`.
`Perhaps the most notable advances have come from research on the intracellular pro—
`cesses that link receptors, second messengers, and various transcription factors to the up- or
`down—regulation of gene activity. Elsewhere in this volume, the current status of research on
`the genetics (Levinson, Chapter 8, this volume) and studies using brain imaging techniques
`[87
`
`This material was {a pied
`attha NLM and may be
`S-Lilzject US “Copyright Laws
`
`

`

`[88
`
`VULNERABILITY, RISK, AND MODELS OF DEPRESSION
`
`to examine normal and pathological processes that accompany emotional expression
`(Davidson, Pizzagalli, 86 Nitschke, Chapter 10, this volume) is reviewed in detail. In this
`chapter, research using other neurobiological paradigms is emphasized, with a particular f0-
`cus on developments that have taken place since our last comprehensive review of this litera-
`ture (Thase et al., 2002). The overarching conceptual framework of this review centers on
`two basic tenets:
`(1) Clinical forms of depression comprise a related yet heterogeneous
`group of syndromes associated with disturbances of the brain systems that regulate the nor-
`mal processes of mood, cognition, and appetitive behavior; and (2) most—if not all—forms
`of depression involve dysfunctional adaptations of the brain systems that regulate adapta-
`tions to stress.
`
`BACKGROUND
`
`Research on the neurobiology of depression began in earnest in the late 19505, when con-
`verging lines of evidence pointed to the possibility of dysfunction of CNS systems subserved
`by the monoamine neurotransmitters, particularly the catecholamine norepinephrine (NE)
`and the indoleamine serotonin (also known as 5—hydroxytryptamine, or 5-HT). Early stud-
`ies indicated that these neurotransmitters are important regulators of bodily functions that
`are commonly disturbed in depression, including sleep, appetite, libido, and psychomotor
`tone; by the mid-1960s, there was strong evidence that both types of medication used to
`treat depression, tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors (MAOIS),
`directly affect NE and/or 5-HT neurons.
`Because the TCAs and MAOIs immediately increased the amount of monoamine activ-
`ity at neuronal synapses, researchers initially thought that depression was caused by a deficit
`of 5 -HT or NE activity, and presumed that mania was caused by increased NE activity, per—
`haps in the context of a deficit of counterbalancing 5—HT activity (Bunney 86 Davis, 1965;
`Glassman, 1969; Schildkraut, 1965). Although the role of a third monoamine neurotrans-
`mitter, dopamine (DA), was generally thought to be more relevant to psychosis and to the
`activity of the phenothiazine-type medications used to treat schizophrenia, some theorists
`also emphasized the putative role of DA in symptoms such as fatigue, anhedonia, and
`psychomotor retardation (Korf 86 van Praag, 1971). Research over the next two decades
`failed to support the most simplistic models (e.g., deficit states corrected by medications that
`“restored” neuronal monoaminergic activity), but it confirmed that the therapeutic effects
`of antidepressants were initiated by actions on 5-HT and/or NE neurons, and investigators
`documented disturbed monoaminergic function in subgroups of individuals with mood dis-
`orders (e.g., see Duman, Heninger, 86 Nestler, 1997; Maes 8C Meltzer, 1995; Nemeroff,
`1998; Schatzberg & Schildkraut, 1995; Willner, 1995).
`Three findings from the first generation of research on the neurobiology of depression
`have ongoing relevance. First, although depression is no longer thought to be caused by defi-
`cits of NE or 5-HT, it is true that subgroups of patients with depression have either low uri—
`nary levels of the NE metabolite 3—methoxy-4-hydroxyphenylglycol (MHPG) (Ressler 8C
`Nemeroff, 1999; Schatzberg 8C Schildkraut, 1995) or low CSF levels of the serotonin metab-
`olite 5-hydroxyind0leacetic acid (5-HIAA) (Maes & Meltzer, 1995). These findings have on-
`going import, because the former abnormality is associated with psychomotor retardation
`(and, possibly, with preferential response to antidepressants that strongly affect noradren—
`ergic neurotransmission; Schatzberg 8c Schildkraut, 1995 ), whereas low CSF 5-HIAA has
`been associated With increased risk of suicide, potentially lethal suicide attempts, and other
`
`Thia material was {opied
`atthe NLM and maybe
`Subject US {lopyright Laws
`
`

`

`Neurological Aspects of Depression
`
`l89
`
`:iffiztt:fiife-threatening behaviors (Maes 8C Meltzer, 1995; Mann, Brent, 85 Arango, 2001),
`(Macs £385 1:3; W1th preferential response to medications that powerfully affect 5—HT neurons
`16213: part1
`eltzer, 19.95). Low CSF 5—HIAA levels subsequently have been shown to be at
`“OmenOn Y under heritable control and, across primate species, appear to be a trait-like phe—
`assoc1ated With various types of aggresswlty and 1mpuls1v1ty (Mann et al., 2001).
`gluC:OS:'COr-1d enduring and well-replicated finding concerns the hypersecretion of the
`Corti3011 11C01d hormone cortisol,
`the primary effector . of stress responses of .humans.
`SPOnse
`IS SynthCSlZCd 1n the adrenalcortex and released into the systemic circulation in re-
`t0 a cascade of neuropeptzdes (i.e., small chains of amino acrds that act as
`neumtfansmitters). The stress response cascade is initiated by corticotropin—releasing factor
`(CRF; also known as corticotropin-releasing hormone, or CRI—I), which is released in the ce-
`lrgiizl Cottex and hypothalamus inresponse to perceived stress. Recent research has estab-
`a link between a polymorphism of the gene coding for the CRF receptor and risk of
`depression (Liu et al., 2006). CRF in turn triggers the release of adrenocorticotropic hormone
`(ACTH), which is secreted by specialized neuroendocrine cells in the anterior pituitary gland
`and travels via systemic circulation to stimulate cortisol release from the adrenal glands.
`PlaSma cortisol levels (i.e., the end product of the hypothalamic—pituitary—adrenocortical
`lHPAl axis in humans) normally follow a well-regulated diurnal rhythm: highest in the
`morning and lowest in the late evening. Intracellular actions of cortisol are mediated by
`Intracellular glucocorticoid receptors, the expression of which are under genetic control (van
`ROSSUm et al., 2006) and can be up— or down-regulated by a number of factors that are rele-
`Vant to depression (Neigh & Nemeroff, 2006).
`levels
`A significant minority of individuals with depression show elevated cortisol
`throughout the day and blunting of the normal circadian secretory rhythm. Given the im—
`portance of glucocorticoids in systemic responses to a variety of acute stresses, including in—
`fection, hypothermia, and traumatic injury, elevated plasma cortisol levels are associated
`Wlth measurably increased concentrations in virtually all body fluids, including urine, saliva,
`and CSF (Holsboer, 1995; Swaab, Bao, 8C Lucassen, 2005). In addition to elevated cortisol
`Concentrations, increased HPA activity can be detected by several challenge paradigms, such
`as the dexamethasone (DEX) suppression test (DST) and the combined DST/CRI-I test
`(Holsboer, 2001). In studies of depression, various indicators of hypercortisolism are linked
`F0 older age and increased syndromal symptom severity, including psychosis and suicidal
`Ideation, as well as a lower response to placebo and nonspecific therapeutic interventions
`(Thase et al., 2002).
`A history of severe maltreatment or trauma during critical developmental periods can
`have lasting effects on regulation of the HPA axis (Heim, Mletzko, Purselle, Musselman, 86
`Nemeroff, 2008; Newport, Heim, Bonsall, Miller, 85 Nemeroff, 2004). In some individuals
`with a history of neglect or maltreatment during childhood, including those who have never
`developed depression, there is blunting of the axis, with reduced cortisol secretion in re-
`sponse to experimentally contrived stresses, such as a public speaking task (Carpenter et al.,
`2007). Blunted HPA response to stress is also seen in individuals with posttraumatic stress
`disorder (PTSD) and chronic fatigue syndrome (Bremner, 2006). Those with a history of
`early trauma and depression, by contrast, are more likely to show an exaggerated HPA re-
`sponse to stress and a state-dependent increase in plasma cortisol (Bremner, 2006; Holsboer,
`2001)
`The third set of pivotal findings emanate from various experimental paradigms that
`measure the activity of localized neuronal circuits within the brain, including several subre-
`gions of the prefrontal cortex and the core structures that comprise the limbic system (Thase
`
`This material was {opied
`atthe NLM and maybe
`Subject US {lopyright Laws
`
`

`

`I90
`
`VULNERABlLlTY, RISK, AND MODELS OF DEPRESSION
`
`et al., 2002). Before it was possible to visualize subtle changes in regional cerebral activity in
`the living human, researchers obtained evidence of depression—related alterations in neuronal
`activity using all—night electroencephalographic (EEG) recordings during sleep (see Thase,
`2006, for a comprehensive review). Such polysomnographic (PSG) recordings revealed a de—
`crease of “deeper” slow-wave sleep (SWS) and an intensification in the amount and intensity
`of rapid eye movement (REM) sleep, and provided objective documentation of the difficul-
`ties that people with depression experience falling asleep and remaining asleep. Although
`neither of these alterations is pathognomonic to depression, the combination was shown to
`be relatively specific and of direct pathophysiological relevance. Because waking EEGS gen-
`erally did not reveal characteristic alterations in depression, sleep appeared to unmask a
`characteristic alteration in the electrical activity of nuclei in the brain under the control of 5—
`HT and NE (Thase, Frank, 86 Kupfer, 1985). The PSG abnormalities associated with depres-
`sion were somewhat more prevalent than was hypercortisolemia, but were nevertheless also
`age-dependent and more commonly observed among people with more severe, recurrent d6”
`pressions (Thase et al., 2002).
`More recently, studies of alterations of neuronal circuitry in depression have utilized
`neuroimaging strategies, including PET and fMRI scans, to measure both the structural in—
`tegrity and functional activity (i.e., metabolic activity and regional blood flow at rest and in
`response to experimental challenges) (Drevets, 2000; Mayberg, 2003). Results of these stud—
`ies, reviewed later in this chapter, have underscored the heterogeneity of depression and
`yielded evidence of several prototypical abnormalities, including increased activity of the
`amygdala, decreased activity of the dorsolateral prefrontal cortex (DLPFC), and reduced
`hippocampal volume (see also Davidson et al., Chapter 10, this volume).
`
`ABNORMALITIES OF MONOAMINERGIC SYSTEMS
`
`Noradrenergic Systems
`
`Almost all of the NE cell bodies in the brain are located in a single nucleus, the locus
`ceruleus (LC), which is located in the rostral brainstem. Noradrenergic neurons project from
`the LC to the thalamus, hypothalamus, limbic system, basal ganglia, and cerebral cortex (see
`Figure 9.1) (Kandel, Schwartz, 8c Jessell, 199']; Kingsley, 2000). Such diffuse ascending pro—
`jections reflect the role of NE in initiating and maintaining arousal in the brainstem, limbic
`system, and cerebral cortex, and as a modulator of other neural systems. Noradrenergic pro—
`jections to the amygdala and hippocampus have been implicated in behavioral sensitization
`to stress (Ferry, Roozendaal, 8c McGaugh, 1999), and stimulation of noradrenergic fibers in
`the medial forebrain bundle enhances attention and increases levels of goal—directed or re—
`ward-seeking behavior (Aston-Jones, Rajkowski, 86 Cohen, 1999).
`Noradrenergic neurotransmission plays an essential role in the experience of stress. Per—
`ception of novel or threatening stimuli is relayed from the cerebral cortex to the LC via the
`thalamus and hypothalamus, and from the periphery via the nucleus prepositus hypoglossi.
`These inputs can provoke an almost immediate increase in NE activity. Thus, cognitive pro—
`cesses affecting perception can amplify or dampen NE cellular responses to internal or exter—
`nal stimuli. In addition, activation from fibers projecting from the nucleus paragiganto—
`cellularis (probably using a small, excitatory neurotransmitter, e.g., glutamate), and release
`of CRH can “turn on” the LC (Nestler, Alreja, 86 Aghajanian, 1999). The peripheral com—
`ponent of stress response to stress is transmitted from the LC via the sympathoadrenal path—
`way to the endochromafin cells in the medulla of the adrenal glands, which in turn release
`
`This. mate rial was :u-piad
`atthe NLM and may the
`Eubja :t U5 Cupvright Laws
`
`

`

`Neurological Aspects of Depression
`
`l9l
`
`Thalamus
`
`Cingulum
`
`Cingulate gyrus
`
`‘ T Cerebellar cortex
`
`Olfactory and Hippocampus
`entorhinai
`cortices
`
`,
`
`Lateral
`tegmental
`NA cell
`system
`
`\\
`
`‘
`I
`Locus
`ceruleus
`
`To spinal cord
`
`HGURE 9.1. A lateral view of the brain demonstrates the course of the major noradrenergic pathways ema-
`“atlng from the locus ceruleus and from the lateral brainstem tegmentum. From Kandcl, Schwartz, and
`jessCII (1991). Copyright l99l by Appleton 8c Lange. Reprinted by permission.
`
`NE into systemic circulation. Thus, the principal effecrors of peripheral stress response, NE
`and cortisol, are released from glands that are located only a few centimeters apart, deep in
`the abdomen. The peripherally arousing effects of the syinpathoadrenal response are largely
`mediated by cells expressing the or! and iii—type of NE receptors.
`The activity of NE neurons is regulated in part by the autoinhibitory effects of (X2 recep-
`tors. Neuronal release of NE almost immediately begins to decrease the sensitivity of LC
`neurons to repeated firing. (x2 receptors also are located on serotoninergic cell bodies, and
`stimulation of these beteroceptors activates nearby (colocalized) inhibitory 5-HT neurons. A
`sustained increase in LC firing (i.e., a normal response to persistent stress) also causes the
`number of 0L] and f§—receptors to decrease, a process known as down—regulation or desensiti-
`zation, Together, these four actions (i.e., a2 autoinhibition, or, and B-receptor down—regulation,
`and activation of adjacent inhibitory 5—HT neurons) constitute a homeostatic counterregu—
`latory force that dampens an excessive response to a transient threat. If, however, the stress
`is sustained or unresolvable, intracellular stores of NE may become depleted when demand
`begins to exceed synthetic capacity. When this occurs, there is diminished inhibitory (x1 and
`5-HT input to the LC. Thus, homeostasis of NE neurotransmission may become dysregu—
`lated, resulting in increased firing of the LC but inefficient signal transduction. Over time,
`the net effect is that ascending central NE neurotransmission decreases (which probably
`causes reduced urinary excretion of MHPG in depressed patients with psychomotor retarda-
`tion), although the output of the adrenal medulla may remain high (which may explain the
`observation of high levels of NE and its metabolites in some severely depressed patients).
`The consequences of sustained stress on NE systems in animal studies include decreased
`exploratory and consummatory behavior, as illustrated in studies using the learned helpless-
`
`This material was EDpiEEl
`atthe NLM and may be
`Sub] art US Copyright Laws
`
`

`

`I92
`
`VULNERABILITY, RISK, AND MODELS OF DEPRESSION
`
`riess paradigm (lVIaier SC Seligman, 1976; Maier 8C Watkins, 2005). Learned he‘lPlCISS-HC-SS
`should not be thought of as strictly analogous to human dePF€5519111 COngWC fOHSUUCtS
`such as entrapment, powerlessness, hopelessness, and gurlt distinguish depresSIOn in humans
`from the behavioral states experienced by YOdeS and dogs 1“ learned helplessness experi—
`ments (Gilbert, 1992). Nevertheless, the changes in NE activity observed in learned helpless~
`ness experiments do parallel those associated with other animal models‘of depressmn and
`are associated with other neurobiological correlates of depression,
`including elevated
`glucocorticoid activity, reduced 5~HT activity, and alterations in gene transcription factors
`(Berton et 3]“ 2007; Maier 8C Watkins, 2005; Weiss 86 Kilts, 1998). Moreover, recognition
`of the mediators of individual differences in development of helplessness—both inherited
`and acquired—has opened new avenues for research (Berton et al., 2007; Krishnan et al.,
`2007).
`.
`,
`Despite the continued relevance of NE neurotransmission as a reliable. target for medi—
`cations that exert antidepressant effects, studies in the 1990s indicated that it is unlikely that
`dysfunction of NE systems has a primary role in the etiology of depresSIon (Anand et al.,
`2000; Ressler 86 Nemeroff, 1999). Nevertheless, several polymorphisms associated with ei—
`ther synthesis of NE or its signal transduction may be associated with excessive responses to
`stress, which may in turn increase the risk of depression during vulnerable periods (Jabbi et
`al., 2007; Shelton, 2007). Altered NE response to stress may likewise play a role as a modu-
`lator of other implicated factors in depression, including borh pathological processes, such
`as the proinflammatory cytokines (Szelényi 8c Vizi, 2007), and processes that promote
`neuronal resilience, such as those mediated by brain-derived neurotropic factor (Chen,
`Nguyen, Pike, 86 Russo-Neustadt, 2007).
`The therapeutic relevance of NE is supported by several converging lines of evidence.
`First, antidepressants that selectively block neuronal reuptake of NE have overall clinical ef—
`ficacy that is roughly comparable to that of the selective serotonin reuptake inhibitors
`(SSRIS) (Nutt et al., 2007; Papakostas, Nelson, Kasper, 8c Moller, 2007). Second, the spe—
`cific additive therapeutic effect of enhancing NE is also suggested by the modest yet repro—
`ducible advantage of the so-called dual—reu/Jta/ee inhibitors (i.e., medications that inhibit
`reuptake of both 5-HT and NE) versus SSRls in meta—analyses of controlled clinical trials
`(Nemeroff et al., 2008; Papakostas, Thase, Fava, Nelson, 8c Shelton, 2007; Thase et al.,
`2007). Third, studies of the physiological effects of selective NE reuptake inhibitors (NRIs)
`have documented normalization of a variety of functional disturbances associated with de—
`pression, including pineal secretion of melatonin and blood pressure responses to changes in
`posture (Golden, Markey, Risby, Cowdry, 8c Potter, 1988; Ressler 8c Nemeroff, 1999).
`Fourth, inhibition of the synthetic enzyme tyrosine hydroxylase via administration of (X-
`methylparatyrosine, an analogue of the NE precursor tyrosine, rapidly reverses the effects of
`NRIs but not of SSRIs (Delgado, 2004). Together, these data indicate that NE plays an im—
`portant neuromodulatory role in the activity of antidepressant medications.
`
`Serotoninergic Systems
`
`Most of the serotonin (5~HT) in the brain is synthesized in clusters of cell bodies known as
`the dorsal raphé nuclei, located in the pons. From the dorsal brainstem, these 5—HT neurons
`project to the cerebral cortex, hypothalamus, thalamus, basal ganglia, septum, and hippo—
`campus (see Figure 9.2) (Kandel et al., 199']; Kingsley, 2000). Serotonin pathways are
`largely colocalized With NE pathways and generally have tonic and inhibitory effects that
`counterbalance NE activity. For example, much evidence indicates that 5—HT input to the
`
`This material was :apiad
`attha NLM and may be
`SLI b] art US Cnpyright Laws
`
`

`

`Neurological Aspects of Depression
`
`I93
`
`Neocortex
`
`
`
`
` Thalamus Cmgu|um Striatum
`
`
`\/”‘
`
`\
`
`Nucleus
`accumbens
`(ventral
`striatum) m
`
`
`
`
`Olfactoryand/- .7-
`entorhinal
`cortices
`Amygdala
`
`
`
`/
`cerebellar
`cortex
`
`Hippocampus
`
`
`Rostral
`To spinal cord
`raphe
`nuclei
`
`Caudal
`raphé
`nuclei
`
`FIGURE 9.2. A lateral view of the brain demonstrates the course of the major serotonincrgic pathways. Al—
`though the raphe nuclei form a fairly continuous collection ofccll groups throughout the brainstem, they are
`graphically illustrated here as two groups, one rostral and one caudal. From Kandel, Schwartz, and Jessell
`(1991). Copyright 1991 by Appleton & Lange. Reprinted by permission.
`
`thalamus is an important facilitator of appetite (Kingsley, 2000). Serotoninergic neurons
`Projecting to the suprachiasmatic nucleus (SCN) of the anterior hypothalamus help to regu-
`late circadian rhythms (e.g., sleep—wake cycles, body temperature, and HPA axis function)
`(Bunney 8c Bunney, 2000; Duncan, I996). An intact 5—HT system also is needed to modu—
`late the 90—minute infraradian cycle of alternating periods of REM and non-RENI sleep
`(Duncan, 1996).
`There are at least 15 types of serotonin receptors in the mammalian brain, each of
`Which is under genetic control. Two of these receptors, 5—HTM and 5—HT”, have been of
`greatest relevance to the pathophysiology of depression and/or the mechanism of antidepres—
`sant action (Mann et al., 2001), although research on some of the more recently identified
`receptors, such as the 5—HT4 and 5—HT7, is in its infancy. All 5—HT neurons express mem—
`braneebound transporters (5—HTT), which permit the uptake of 5—HT from the synaptic
`Cleft. The activity of many antidepressants is initiated by blocking this transporter, includ—
`ing, of course, the most widely used antidepressants in contemporary clinical practice, the
`SSRIs. As I discuss later in this chapter, identification of a functional polymorphism in the
`promoter region of the gene that codes for the 5—HTT has opened multiple new lines of re—
`search and helps to explain individual differences in response to stress and antidepressant
`medications.
`
`An intact basal or tonic level of 5—HT neurotransmission is necessary for both affiliative
`social behaviors (Insel 8c Winslow, 1998) and the expression of goal—directed motor and
`consummatory behaviors primarily mediated by NE and DA. In experimental paradigms,
`defeat reliably lowers basal 5‘HT tone across essentially all vertebrates studied and, in the
`
`This material was copied
`attha NLM and may be
`Su bjatt US Cupyright Laws
`
`

`

`I94
`
`VULNERABILITY, RISK, AND MODELS OF DEPRESSION
`
`wild, primates with lower levels of tonic 5-HT neurotransmission (as measured by CSF 5—
`HIAA levels) are more impulsive, aggressive, and generally have lower rankings on social
`dominance hierarchies than do animals with higher basal levels of 5-HT “tone” (Higley,
`Mehlman, Higley, et al., 1996; Higley, Mehlman, Poland, et al., 1996). Conversely, a rise in
`social dominance is accompanied by an increase in CSF 5-HIAA (Mehlman et al., 1995),
`and treatment with SSRIs decreases impulsive aggression (Fairbanks, Melega, Jorgensen,
`Kaplan, 8c McGuire, 2001). There is ample documentation of parallel associations in hu~
`mans and low 5-HIAA is associated with suicide and other violent behaviors (Mann et al.,
`2001).
`The tonic level of 5-HT neurotransmission in primates is relatively stable, with a slight
`seasonal variation (i.e., higher levels in the summer than in the fall) (Zajicek et al., 2000).
`Central serotoninergic tone is partly under genetic control (Higley, Mehlman, Poland, et al.,
`1996), with heritability at least partly determined by a polymorphism in the promoter re-
`gion of the gene that codes for 5-HT-T. In primates, animals manifesting at least one copy of
`the short (S) allele, which is less functional (i.e., less transporter is synthesized, resulting in
`reduced uptake capability), show greater behavioral dysfunction and more exaggerated re-
`sponses to stress than do animals who have two copies of the more common long (L) form
`of the allele (Barr et al., 2003, 2004; Shannon et al., 2005).
`Humans show a similar polymorphism of 5-HTT, with the recent identification of a
`third variant (a less functional variant of the L form) (Firk 85 Markus, 2007; Levinson,
`2006). Studies of the association of these polymorphisms and vulnerability to depression
`have yielded relatively consistent evidence of gene x environment interactions. A relation
`among the S allele of the serotonin transporter, stress, and increased risk of depression and
`suicidal ideation was first reported by Caspi and colleagues (2003) and subsequently widely
`(albeit not universally) replicated (e.g., see Jacobs et al., 2006; Kendler, Kuhn, Vittum,
`Prescott, 86 Riley, 2005). Importantly, individuals with one or two copies of the S allele are
`not at increased risk of depression per se, but are at increased risk of depression when ex-
`posed to life stress (Firk 8C Markus, 2007; Levinson, 2006). Such heightened vulnerability to
`stress is apparent at several levels, including increased limbic blood flow (Hariri et al., 2002)
`or cortisol secretion (Gotlib, Joorman, Minor, 8c Hallmayer, 2007) in response to experi-
`mentally induced threat, elevated levels of trait—like neuroticism (Jacobs et al., 2006) or dys-
`functional attitudes (Hayden et al., 2008), and use of less active coping strategies (Wilhelm
`et al., 2007). That this inherited vulnerability is typically manifest early in life is indirectly
`reflected by the results of Baune and colleagues (in press), who found that the melancholic
`form of depression—which typically has a later age of onset—is disproportionately associ-
`ated with L alleles for the serotonin transporter. Although results of individual studies are
`not fully consistent, a recent meta-analysis of 15 studies found a significant association be-
`tween the S allele and a lower likelihood of response or remission (Serretti, Kato, De
`Ronchi, 8c Kinoshita, 2007; see also Levinson, Chapter 8, this volume).
`Reduced numbers of 5—HT uptake transporters also have been demonstrated in blood
`platelets (Maes 8c Meltzer, 1995) in the brains of depressed individuals who committed sui-
`cide (Lin 8c Tsai, 2004; Mann et al., 2001), and by in w'uo receptor imaging in depressed pa—
`tients (Parsey et al., 2006). This reduction in 5-HTT capacity appears to be linked directly to
`inheritance of the S allele of 5-HTT (Li 86 He, 2007; Wasserman et al., 2007).
`Available evidence from studies using receptor imaging techniques suggests that dys-
`function of 5-HT1A receptors is clearly implicated in depression (Drevets et al., 2007). Al-
`though this abnormality could be an artifacth exposure to antidepressant medication, it
`has recently been demonstrated in a study of treatment-naive individuals (Hirvonen et al.,
`
`This material was {a pied
`atthe NLM and may be
`Subject US {ea-wright Laws
`
`

`

`Neurological Aspects of Depression
`
`I95
`
`2237). Down—regulation of 5—HT1A receptors is a consequence of exposure to chronic stress,
`(L6 ever, which—1n the absence of a herltable risk factor—is the most likely explanation
`Dez et al., 1999; Maier 8C Watkins, 2005). Nevertheless, an allelic variation of the 5-
`terféA receptor has recently been reported to be associated with .risk of depression during in-
`.
`ton therapy (Kraus et al., 2007), so the potential contribution of a heritable vulnerabil-
`lty Cannot be discounted.
`.
`The integrity of 5-HT neurotransmission also can be transiently compromised by di-
`:imlry manipulation, specifically, by eliminating the precursor tryptophan (one of the essen—
`a amino acids) from the food source. Complete disruption of 5-HT synthesis has little im~
`mediate impact on mood in studies of healthy individuals, but it does impact more subtle
`aspeCts of cognitive—affective processing, such as enhanced anticipation of punishment
`C9015, Robinson, 86 Sahakian, in press) and reduction of the normal attentional bias to
`pOSltive emotionally valenced stimuli (Roiser et al., in press). In studies of depressed people,
`a,b_rief period of tryptophan depletion does not worsen untreated depression, but it does sig-
`rI‘lflcantly increases depressive symptoms in some unmedicated people with remitted depres-
`Slve episodes (e.g., see Neumeister et al., 2006). Neumeister and colleagues (2006) also
`.Ound that response to tryptophan depletion differs significantly between remitted depressed
`lndividuals and controls as a function of genetic vulnerability. Within the group of remitted
`depféissed people, tryptophan depletion had stronger effects in individuals with at least one
`COPY of the L allele of the 5—HTT, whereas within normal controls, only those who had two
`C()P’itls of the S polymorphism showed an increase in depressive symptoms.
`Among patients treated for depression, tryptophan depletion can reverse acute response
`OVCFnight in about 50 to 60% of people treated with SSRI antidepressants (Delgado, 2004;
`Delgado et al., 1991; Moore et al., 2000). Tryptophan depletion does not reverse response
`to Placebo (Delgado, 2004). The lack of effect of tryptophan depletion on the improvement
`0f Patients treated with NRIs (Delgad