`
`RESSION
`
`Second Edition
`
`Edited by
`
`IAN H. GOTLIB
`CONSTANCE L. HAMMEN
`
`ep
`
`THE GUILFORD PRESS
`New York
`London
`
`Merck 2002
`Merck 2002
`Argentum v. Merck
`Thismaterial wascopied
`Argentum Vv. Merck
`IPR2018-00423
`
`SubjectUSConyristL IPR2018-00423
`Subject US Copyright Laws
`
`
`
`000tne Guilford Press
`A Divison 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
`
`I
`
`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, ete.
`[DNLM: 1. Depressive Disorder.
`2. Depression.
`RCS537.H3376 2009
`616.85'27—dc22
`
`TI. Hammen, Constance L.
`I. Gotlib, lan H.
`3. Risk Factors. WM 171 H2367 2008]
`
`2008010575
`
`This material was copied
`atthe MLM and may be
`Subject US Cooyright Laws
`
`
`
`CHAPTER 9
`
`Neurobiological Aspects of Depression
`
`Michael E. Thase
`
`sion. T. Sine antiquity, there have been speculations about the biological basis of depres-
`of the o take but one example, the term melancholia, which is currently used to describe one
`dere most severe forms of depression, reflects the ancient Greek theory that mooddisor-
`€fS Were caused by an imbalance of black bile (Jackson, 1986). Only during the past 50
`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 beenaniterative 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 biopsychosocial disorders
`that—much more often than not—have multifactorial causality.
`Mycolleagues and I reviewed evidence pertaining to neurobiological disturbances asso-
`ciated with depression in the previous volumeofthis Handbook,including a wide range of
`neurochemical, neuroendocrine, neurophysiological, and neuroanatomical parameters (Thase,
`Jindal, 8& 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 aboutthe function of the central nervous system (CNS)in health, in disease, and
`in response to various states of duress has grown. A number of new hypotheses also have
`been advanced. Some research tools, such as measurementof catecholamine metabolites in
`urine, blood, and cerebrospinal fluid (CSF) or electrophysiological recordings of neuronal
`activity, that were de rigueur in the 1970s and 1980s are now seldom used; others that were
`not technologically feasible, such as functional magnetic resonance imaging (fMRI), posi-
`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 geneactivity. Elsewhere in this volume, the currentstatus of research on
`the genetics (Levinson, Chapter 8, this volume) and studies using brain imaging techniques
`187
`
`This material was copied
`atthe MLM and may be
`Subject US Copyright Laws
`
`
`
`188
`
`VULNERABILITY, RISK, AND MODELS OF DEPRESSION
`
`to examine normal and pathological processes that accompany emotional expression
`(Davidson, Pizzagalli, & Nitschke, Chapter 10, this volume) is reviewed in detail. In this
`chapter, research using other neurobiological paradigmsis emphasized, with a particular fo-
`cus on developments that have taken place since our last comprehensive review ofthis 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 1950s, when con-
`verging lines of evidence pointedto 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 oxidaseinhibitors (MAOIs),
`directly affect NE and/or 5-HT neurons.
`Because the TCAs and MAOIs immediately increased the amount of monoamineactiv-
`ity at neuronal synapses, researchersinitially thought that depression was caused by a deficit
`of S-HT or NEactivity, and presumed that mania was caused by increased NEactivity, per-
`hapsin the context of a deficit of counterbalancing 5-HT activity (Bunney & 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 symptomssuch as fatigue, anhedonia, and
`psychomotorretardation (Korf 8 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 monoaminergicactivity), but it confirmed that the therapeutic effects
`of antidepressants were initiated by actions on S-HT and/or NE neurons, and investigators
`documented disturbed monoaminergic function in subgroupsof individuals with mooddis-
`orders (e.g., see Duman, Heninger, & Nestler, 1997; Maes & Meltzer, 1995; Nemeroff,
`1998; Schatzberg & Schildkraut, 1995; Willner, 1995).
`Three findings from thefirst generation of research on the neurobiology of depression
`have ongoingrelevance.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 &
`Nemeroff, 1999; Schatzberg & Schildkraut, 1995) or low CSFlevels of the serotonin metab-
`olite 5-hydroxyindoleacetic 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 & Schildkraut, 1995), whereas low CSF 5-HIAA has
`been associated with increasedrisk of suicide, potentially lethal suicide attempts, and other
`
`This meterial was copied
`atthe NLM and maybe
`Subject US Copyright Laws
`
`
`
`Neurological Aspects of Depression
`
`189
`
`alehnife-threatening behaviors (Maes & Meltzer, 1995; Mann,Brent, & Arango, 2001),
`(Maesee xt with preferential response to medications that powerfully affect 5-HT neurons
`least part
`e€ tren 1995). Low CSF 5-HIAA levels subsequently have been shown to be at
`nomenon y under heritable control and, across primate species, appearto be a trait-like phe-
`associated with various types of aggressivity and impulsivity (Mannetal., 2001).
`elucsonecond enduring and well-replicated finding concerns the hypersecretion of the
`oid hormone cortisol,
`the primary effector of stress responses of humans.
`Spon. is synthesized in the adrenalcortex and released into the systemic circulation in re-
`neuroe to acascade of neuropeptides (i.e., small chains of amino acids that act as
`ransmitters). The stress response cascadeis initiated by corticotropin-releasing factor
`he also known as corticotropin-releasing hormone, or CRH), whichis released in the ce-
`lished cortex and hypothalamus inresponse to perceived stress. Recent research has estab-
`a link between a polymorphism of the gene coding for the CRF receptor andrisk of
`“Pression (Liu et al., 2006). CRFin turn triggers the release of adrenocorticotropic hormone
`(ACTH), whichis secreted by specialized neuroendocrinecells in the anterior pituitary gland
`and travels via systemic circulation to stimulate cortisol release from the adrenal glands.
`Plasmacortisol levels (i.e., the end product of the hypothalamic-pituitary—adrenocortical
`[HPA] 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 undergenetic control (van
`Rossum et al., 2006) and can be up- or down-regulated by a numberoffactorsthat are rele-
`vant to depression (Neigh & Nemeroff, 2006).
`levels
`A significant minority of individuals with depression show elevated cortisol
`throughoutthe 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
`with measurably increased concentrationsin virtually all body fluids, including urine, saliva,
`and CSF (Holsboer, 1995; Swaab, Bao, & 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/CRHtest
`(Holsboer, 2001). In studies of depression, variousindicators of hypercortisolism are linked
`to 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, &
`Nemeroff, 2008; Newport, Heim, Bonsall, Miller, & Nemeroff, 2004). In someindividuals
`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 (Carpenteretal.,
`2007). Blunted HPA response tostress 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-
`sponseto stress and a state-dependentincrease in plasmacortisol (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 meterial was copied
`atthe NLM and maybe
`Subject US Copyright Laws
`
`
`
`190
`
`VULNERABILITY, RISK, AND MODELS OF DEPRESSION
`
`et al., 2002). Before it was possible to visualize subtle changesin regional cerebralactivity 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 revealeda de-
`crease of “deeper” slow-wave sleep (SWS) and anintensification 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 remainingasleep. Although
`neither of these alterations is pathognomonic to depression, the combination was shown to
`be relatively specific andof 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 underthe control of 5-
`HTand NE(Thase, Frank, & Kupfer, 1985). The PSG abnormalities associated with depres-
`sion were somewhat more prevalent than washypercortisolemia, but were nevertheless also
`age-dependent and more commonly observed among people with more severe, recurrent de-
`pressions (Thase et al., 2002).
`More recently, studies of alterations of neuronal circuitry in depression have utilized
`neuroimaging strategies, including PET and fMRIscans, to measure both the structural in-
`tegrity andfunctional activity (i.e. metabolic activity and regional blood flowat 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, decreasedactivity 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, & Jessell, 1991; Kingsley, 2000). Such diffuse ascending pro-
`jectionsreflect the role of NE in initiating and maintaining arousal in the brainstem, limbic
`system, andcerebral 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, & McGaugh, 1999), andstimulation of noradrenergic fibers in
`the medial forebrain bundle enhances attention and increases levels of goal-directed or re-
`ward-seeking behavior (Aston-Jones, Rajkowski, & Cohen, 1999).
`Noradrenergic neurotransmissionplays anessential role in the experienceof 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 NEcellular 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, & Aghajanian, 1999). The peripheral com-
`ponent of stress responseto stress is transmitted from the LC via the sympathoadrenal path-
`way to the endochromafincells in the medulla of the adrenal glands, which in turn release
`
`This material was copied
`at the NLM and may be
`Subject US Copyright Laws
`
`
`
`Thalamus—Cingulum
`
`Cingulate gyrus
`
`Neurological Aspects of Depression
`
`191
`
`“~~ Cerebellar cortex Z
`
`Olfactory and Hippocampus
`entorhinal
`cortices
`
`Lateral
`tegmental
`NA cell
`system
`
`.
`
`|
`Penis
`ceruleus
`
`To spinal cord
`
`FIGURE 9.1. A lateral view of the brain demonstrates the course of the major noradrenergic pathways ema-
`Nating from the locus ceruleus and from the lateral brainstem tegmentum. From Kandel, Schwartz, and
`Jessel] (1991). Copyright 1991 by Appleton & Lange. Reprinted by permission.
`
`NEinto systemiccirculation. Thus, the principal effectors of peripheral stress response, NE
`andcortisol, are released from glands that are located only a few centimeters apart, deep in
`the abdomen. The peripherally arousing effects of the sympathoadrenalresponseare largely
`mediated by cells expressing the o, and {%-type of NE receptors.
`The activity of NE neuronsis regulated in part by the autoinhibitory effects of 0 recep-
`tors, Neuronal release of NE almost immediately begins to decrease the sensitivity of LC
`neurons to repeatedfiring. 0, receptors also are located on serotoninergic cell bodies, and
`stimulation of these heteroceptors activates nearby (colocalized) inhibitory S-HT neurons. A
`sustained increase in LC firing (i.e., a normal response to persistent stress) also causes the
`numberof 0, and &-receptors to decrease, a process knownas down-regulation or desensiti-
`zation. Together, these fouractions(i.e., 0 autoinhibition, a, and {¢-receptor down-regulation,
`andactivation of adjacent inhibitory 5-HT neurons) constitute a homeostatic counterregu-
`latory force that dampens anexcessive responseto a transientthreat. If, however, the stress
`is sustained or unresolvable, intracellular stores of NE may become depleted when demand
`begins to exceed synthetic capacity. Whenthis occurs, there is diminished inhibitory 0, and
`5-HT input to the LC. Thus, homeostasis of NE neurotransmission may become dysregu-
`lated, resulting in increasedfiring of the LC butinefficient signal transduction. Over time,
`the net effect is that ascending central NE neurotransmission decreases (which probably
`causes reducedurinary excretion of MHPGin depressedpatients with psychomotorretarda-
`tion), although the outputof the adrenal medulla may remain high (which may explainthe
`observation of high levels of NE andits metabolites in some severely depressedpatients).
`The consequencesofsustainedstress on NE systems in animalstudies include decreased
`exploratory and consummatory behavior,asillustrated in studies using the learned helpless-
`
`This material was copied
`at the NLM and may be
`Subject US Copyright Laws
`
`
`
`192
`VULNERABILITY, RISK, AND MODELS OF DEPRESSION
`ness paradigm (Maier & Seligman, 1976; Maier & Watkins, 2005). Learned Helpiessttess
`should not be thought ofas strictly analogous to human depression: Cognitive CONStructs
`such as entrapment, powerlessness, hopelessness, and guilt distinguish depressionin humans
`from the behavioral states experienced by rodents and dogs in learned helplessness experi-
`ments (Gilbert, 1992). Nevertheless, the changes in NE activity observedin learned helpless-
`ness experiments do parallel those associated with other animal models of depression 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 al., 2007; Maier & Watkins, 2005; Weiss & 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; Krishnanetal.,
`2007).
`Despite the continued relevance of NE neurotransmission as a reliable target for medi-
`cationsthat exert antidepressanteffects, studies in the 1990s indicated that it is unlikely that
`dysfunction of NE systems has a primary role in the etiology of depression (Anandetal.,
`2000; Ressler & Nemeroff, 1999). Nevertheless, several polymorphismsassociated withei-
`ther synthesis of NEorits signal transduction maybe 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 responseto stress may likewise play a role as a modu-
`lator of other implicated factors in depression, including both pathological processes, such
`as the proinflammatory cytokines (Szelényi & Vizi, 2007), and processes that promote
`neuronalresilience, such as those mediated by brain-derived neurotropic factor (Chen,
`Nguyen, Pike, & 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, & Moller, 2007). Second, the spe-
`cific additive therapeutic effect of enhancing NEis also suggested by the modest yet repro-
`ducible advantage of the so-called dual-reuptake inhibitors (i.e., medications that inhibit
`reuptake of both 5-HTand NE) versus SSRIs in meta-analyses of controlledclinical trials
`(Nemeroff et al., 2008; Papakostas, Thase, Fava, Nelson, & 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 andbloodpressure responses to changes in
`posture (Golden, Markey, Risby, Cowdry, & Potter, 1988; Ressler & Nemeroff, 1999).
`Fourth, inhibition of the synthetic enzyme tyrosine hydroxylase via administration of o-
`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 neuromodulatoryrole in the activity of antidepressant medications.
`
`Serotoninergic Systems
`
`Mostof the serotonin (5-HT) in the brain is synthesizedin 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., 1991; Kingsley, 2000). Serotonin pathways are
`largely colocalized with NE pathways andgenerally have tonic andinhibitory effects that
`counterbalance NE activity. For example, muchevidence indicates that 5-HT input to the
`
`This material was copied
`at the NLM and may be
`Subject US Copyright Laws
`
`
`
`Neurological Aspects of Depression
`
`193
`
` Thalamus
`Cingulum
`
`Striatum
`LD
`Neocortex
`\ee™ :
`Cingulate gyrus
`
`NK
`fo ‘
`
`oO~
`
`
`Nucleus
`accumbens
`(ventral
`striatum) —
`
`\
`
`~
`~~~. Cerebellar
`
`Olfactory and-— SZ
`Connex
`
`entorhinal
`‘Deep
`Amygdala
`cortices
`cerebellar
`
`cortex
`
`Hippocampus
`
`Rostral
`raphe
`
`nuclei
`Caudal
`raphe
`nuclei
`
`To spinal cord
`
`FIGURE 9.2. A lateral view of the brain demonstrates the course of the major serotoninergic pathways. Al-
`thoughthe raphé nuclei forma fairly continuous collectionof cell groups throughoutthe brainstem, they are
`8raphically 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 importantfacilitator 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 & Bunney, 2000; Duncan, 1996). An intact 5-HT system also is needed to modu-
`late the 90-minute infraradian cycle of alternating periods of REM and non-REM sleep
`(Duncan, 1996).
`There are at least 15 types of serotonin receptors in the mammalian brain, each of
`whichis under genetic control. Two of these receptors, 5-HT,, and S-HT),, have been of
`greatest relevance to the pathophysiology of depression and/or the mechanismofantidepres-
`sant action (Mann etal., 2001), although research on some of the more recently identified
`receptors, such as the 5S-HT, and 5-HT,,is in its infancy. All 5-HT neurons express mem-
`brane-bound transporters (S-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 contemporaryclinical practice, the
`SSRIs. As I discuss later in this chapter, identification of a functional polymorphism in the
`promoter region ofthe gene that codes for the 5-HTThas opened multiple new lines of re-
`search and helps to explain individual differences in response to stress and antidepressant
`medications.
`Anintact basal ortonic level of 5S-HTneurotransmissionis necessary for bothaffiliative
`social behaviors (Insel & 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 wes copied
`atthe NLM and maybe
`Subject US Copyright Laws
`
`
`
`194
`
`VULNERABILITY, RISK, AND MODELS OF DEPRESSION
`
`wild, primates with lower levels of tonic 5-HT neurotransmission (as measured by CSF 5-
`HIAAlevels) are more impulsive, aggressive, and generally have lower rankings on social
`dominance hierarchies than do animals with higher basal levels of S-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 (Mehiman et al., 1995),
`and treatment with SSRIs decreases impulsive aggression (Fairbanks, Melega, Jorgensen,
`Kaplan, & McGuire, 2001). There is ample documentation of parallel associations in hu-
`mans and low 5-HIAAis associated with suicide and other violent behaviors (Mann et al.,
`2001).
`The tonic level of 5-HT neurotransmission in primatesis relatively stable, with a slight
`seasonal variation (i.e., higher levels in the summer than in the fall) (Zajicek et al., 2000).
`Central serotoninergic toneis partly under genetic control (Higley, Mehlman, Poland,etal.,
`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 commonlong (L) form
`of the allele (Barr et al., 2003, 2004; Shannonetal., 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 & Markus, 2007; Levinson,
`2006). Studies of the association of these polymorphisms and vulnerability to depression
`have yielded relatively consistent evidence of gene x environmentinteractions. A relation
`amongtheSallele of the serotonin transporter, stress, and increased risk of depression and
`suicidal ideation wasfirst reported by Caspi and colleagues (2003) and subsequently widely
`(albeit not universally) replicated (e.g., see Jacobs et al., 2006; Kendler, Kuhn, Vittum,
`Prescott, & Riley, 2005). Importantly, individuals with one or two copiesof theSallele are
`notat increasedrisk of depression per se, but are at increased risk of depression when ex-
`posedtolife stress (Firk & Markus, 2007; Levinson, 2006). Such heightened vulnerability to
`stress is apparentat several levels, including increased limbic blood flow (Hariri et al., 2002)
`or cortisol secretion (Gotlib, Joorman, Minor, & Hallmayer, 2007) in response to experi-
`mentally inducedthreat, elevated levels of trait-like neuroticism (Jacobs et al., 2006) or dys-
`functional attitudes (Haydenet al., 2008), and use ofless 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—whichtypically has a later age of onset—is disproportionately associ-
`ated with L alleles for the serotonin transporter. Although results of individual studies are
`notfully 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, & Kinoshita, 2007; see also Levinson, Chapter 8, this volume).
`Reduced numbers of 5-HT uptake transporters also have been demonstrated in blood
`platelets (Maes & Meltzer, 1995) in the brains of depressed individuals who committed sui-
`cide (Lin & Tsai, 2004; Mannet al., 2001), and by iz vivo receptor imaging in depressed pa-
`tients (Parsey et al., 2006). This reduction in S-HTT capacity appearsto belinked directly to
`inheritance of the S allele of 5-HTT (Li & He, 2007; Wasserman et al., 2007).
`Available evidence from studies using receptor imaging techniques suggests that dys-
`function of 5-HT,, receptorsis clearly implicated in depression (Drevets et al., 2007). Al-
`though this abnormality could be an artifact“of exposure to antidepressant medication, it
`has recently been demonstrated in a study of treatment-naive individuals (Hirvonenet al.,
`
`This material was copied
`atthe MLM and may be
`Subject US Copyright Laws
`
`
`
`Neurological Aspects of Depression
`
`195
`
`how? Down-regulation of 5-HT,, receptors is a consequence of exposure to chronic stress,
`(Le ever, which—in the absence of a heritable risk factor—is the most likely explanation
`bez et al., 1999; Maier & Watkins, 2005). Nevertheless, an allelic variation of the 5-
`ter la receptor has recently been reported to be associated with risk of depression during in-
`ity €ron therapy (Krauset al., 2007), so the potential contribution of a heritable vulnerabil-
`Sannot be discounted.
`.
`The integrity of 5-HT neurotransmission also can be transiently compromised by di-
`al manipulation, specifically, by eliminating the precursor tryptophan (one of the essen-
`*' amino acids) from the food source. Complete disruption of 5-HT synthesis haslittle im-
`Mediate impact on moodin studies of healthy individuals, but it does impact more subtle
`“SPeets of cognitive-affective processing, such as enhanced anticipation of punishment
`Cools, Robinson, & Sahakian, in press) and reduction of the normalattentional bias to
`PSitive emotionally valenced stimuli (Roiseretal., in press). In studies of depressed people,
`a brieg period of tryptophan depletion does not worsen untreated depression, but it doessig-
`nificantly increases depressive symptoms in some unmedicated people with remitted depres-
`SIV€
`episodes (e.g., see Neumeister et al., 2006). Neumeister and colleagues (2006) also
`‘ound that response to tryptophan depletiondiffers significantly between remitted depressed
`individuals and controls as a function of genetic vulnerability. Within the group of remitted
`Pressed people, tryptophan depletion had strongereffects in individuals with at least one
`COPY ofthe L allele of the 5-HTT, whereas within normal controls, only those who had two
`OPies of the S polymorphism showed an increase in depressive symptoms.
`Amongpatients treated for depression, tryptophan depletion can reverse acute response
`Overnight in about 50 to 60% of people treated with SSRI antidepressants (Delgado, 2004;
`Delgado et al., 1991; Mooreet al., 2000). Tryptophan depletion does not reverse response
`to Placebo (Delgado, 2004). The lack of effect of tryptophan depletion on the improvement
`of patients treated with NRIs (Delgado, 2004), repetitive transcranial magnetic stimulation
`(O’Reardonet al., 2006), and cognitive therapy (O’Reardon etal., 2004) also points to the
`SP€Cificity of this mechanism.
`
`Dopaminergic Systems
`There are four principal DA pathways in the brain (see Figure 9.3) (Kandel et al., 1991;
`Kin