Neurobiological Aetiology of Mood Disorders

Technical article

Topics covered

  • Introduction
  • Neurobiology of normal emotion
  • The neurobiology of mood disorder
    • Vulnerability to mood disorder
    • Early adverse experience
    • The neurobiology of life events
  • Biological studies of the depressed state
    • The depressed state: functional anatomy
    • Neuroendocrine challenge tests
    • Hypercortisolaemia
    • Thyroid abnormalities
    • Sleep disturbance
    • Monoamine metabolite turnover
    • Tryptophan depletion
  • Does mood disorder have a functional neuropathology?
  • The neurobiology of treatments
  • Conclusions
  • References


Neurobiology provides an explanation of behaviour or experience at the level either of systems of neurones or individual cells. There is no particular difficulty now in producing accounts of normal emotion and disorders of emotion, based on the currently available techniques for the investigation of brain function. These accounts remain preliminary, but they are no longer purely speculative. However, there has been a curious divorce between the extensive literature on normal emotion and the predominantly clinical or biochemical accounts of mood disorder. It was probably inevitable that the psychology of normal emotion would take a ‘top-down' direction, while clinical approaches would be rather more ‘bottom-up', dominated by symptoms and the effects of treatments. However, it now seems certain that one field has the potential to inform the other, and that such unifying activity might well take its origins from contemporary cognitive neuroscience. As well as witnessing conceptual advances, we are at a stage of rapid evolution in the platform technologies of imaging and genetics. These will allow us to improve our accounts of the functional anatomy of the component elements of mood and its disorder, their functional neurochemistry, and, in all probability, give meaning to what a cellular account of depressive illness may eventually mean.

Neurobiology of normal emotion

Any account of function in the human brain must start with its anatomy. Indeed, the most influential modern theory of how emotion is represented within the brain was an essentially anatomical speculation. (1) Building on the observation that decorticate animals could express ‘sham rage', Papez argued that areas projecting to the hypothalamus would be essential for the experience of emotion. Movement, thought, and emotion were identified with sensory projections to the striatum, neocortex and limbic system respectively. The limbic cortex had been observed by neuropathologists to be a particular locus for infection with the rabies virus, and the association of early stages of disease expression with psychiatric symptoms was well established. The particular symptoms seen most floridly in rabies are intermittent outbursts of fury or terror.

The term ‘limbic cortex or limbic system' is dignified by use, but not by coherent definition. Broca referred to le grand lobe limbique as a ring of grey matter bordering the hemispheres lying against the central structures of the brain and arranged in a circular fashion around the interventricular foramen. It is usually taken to include the cingulate gyrus, the hippocampus including dentate gyrus, the subiculum and related areas, the entorhinal cortex, the septum, the olfactory tubercle, and the amygdala. The grouping together of such anatomically different structures of uncertain function has long seemed convenient, but only now are we beginning to accumulate a functional understanding.

The central place of limbic structures in the experience of emotion also emerged from accounts of temporal lobe or psychomotor epilepsy, where a range of phenomena are described that may be relevant to normal emotion and its disorders. The corollary of these observations is the effect induced by stimulation of limbic brain areas in patients with seizures. As well as the simple subjective experiences described in the aura of a seizure, electrical stimulation of cortex exposed prior to surgery produced automatisms of the same type as those involved in spontaneous seizures. This usually required direct activation close to the amygdala for a full range of emotional effects. However, lesions in or stimulation of the inferior frontal and cingulate cortex were also implicated in ‘psychomotor' seizures.

These early observations provided the focus for subsequent efforts aimed at understanding the underlying neural mechanisms. Lesions of the amygdala effectively produced the full range of abnormalities described in the Klüver–Bucy syndrome after more extensive temporal lobe excisions, most notably the loss of spontaneous aggression. This and more formal investigation of learning and behaviour after amygdalectomy led to the view that the amygdala was the key structure assigning motivational significance to stimuli. A critical feature of the more recent elaborations of this theory is that some sensory impressions are conveyed to the amygdala independent of the cortex. Such input may deliver unconditioned stimuli (e.g. pain), which allows the amygdala to be critical for conditioning and emotional processing generally. The set of the amygdala is then postulated to be critical in determining the action-repertoire of an individual, from eager readiness for action on the one hand to fearful avoidance or inhibition on the other. Notice how pivotal such a function is and how it may underlie what goes wrong in mood disorder.

In rare clinical cases, bilateral lesions of the amygdala are associated with defective social decision-making and poor recognition of emotional expression. (2) The amygdala appears to be peculiarly adapted to the representation of emotionally significant stimuli. Thus if fearful faces are presented briefly before a neutral face, the conscious perception of the fearful face is prevented–the neutral face masks the fearful face. However, functional magnetic resonance imaging (MRI) studies show that the amygdala can still signal the difference between a masked fearful face and a masked happy face. (3) Such subliminal processing could provide the neurological basis for emotional associations that are developed and experienced in particular social circumstances, with specific people, or relating to specific events. The amygdala has extensive projections that can account for effects on autonomic, neuroendocrine, and motor activation.

To study ‘emotion' is not necessarily to study mood. Negative and positive mood induction in normal volunteers has been associated with activation or deactivation in a variety of brain areas in different studies. Inferior frontal and temporal cortex, rather than subcortical structures such as the amygdala were the most commonly implicated regions. (4) Differences in experimental design will have contributed to the variability of these findings, which remain difficult to interpret.

Relatively restricted lesions in the inferior frontal cortex have catastrophic consequences for general behaviour without impairing performance on a whole range of knowledge-based tests of higher cognitive function. Damasio (5) has speculatively distinguished ‘background feeling' from extremes of emotion in determining this mechanism. Animal experiments suggest that primary reinforcing stimuli, such as taste and smell, are represented in a secondary sensory area directly within the inferior frontal cortex, together with elaborated visual representations. Efferent projections may regulate autonomic and monoamine projections. Although the objective failures in social or executive activity after such lesions are most obvious, the subjective experience of emotion may also be impaired. (6) Understanding the functions of the amygdala and the inferior prefrontal cortex is likely to prove central to the neurobiology of anxiety, depression, and mania.

As well as a functional anatomy, mood also has a neurochemistry. The original finding that lysergic acid diethylamide (LSD) profoundly modified mood and perception formed a critical background to efforts to understand such mechanisms in biochemical terms. The neurotransmitters involved are dopamine and noradrenaline (norepinephrine), together with serotonin (5-hydroxytryptamine, 5-HT). 3,4-Methylenedioxymethamphetamine (MDMA, ecstasy), which more specifically releases 5-HT, (7) has profound effects on mood, activity, and sleep. Ecstasy's particular selectivity for 5-HT neurones may be directly linked to mood modulation; its effects are valued precisely for the sense of contact, happiness, and pleasure that is evoked. Such effects will ultimately be framed in terms of chemical addressing in particular brain regions. 

The neurobiology of mood disorder

Vulnerability to mood disorder


Positional mapping of the genetic loci associated with mood disorder has not yet yielded important clues in searching for fundamental neurobiological causes. However, neurobiology informs the genetic search for candidate genes such as the human serotonin transporter ( SERT) gene. (8,9) Cerebrospinal fluid monoamine levels of the noradrenaline metabolite 3-methoxy-4-hydroxyphenylglycol (MHPG) appear to vary in relation to the two SERT polymorphisms. (10) This is the sort of finding that, if replicated, will start to illuminate abnormal function in the affected phenotype.

An alternative approach to looking for the heritability of mood disorder itself is to examine the heritability of traits that predispose to its development. Such quantitative trait loci can be sought relatively simply in a normal population. High-trait neuroticism accounts for approximately half of the genetic liability to major depression in women. (11) Most importantly, high-trait neuroticism is itself highly heritable and stable in normal populations. Accordingly, individual differences in learning, stress hormone secretion, and a variety of other behaviours lend themselves to genetic studies in animals and ultimately in humans also. The first such successful study showed that ‘emotionality' in mice was associated with at least three loci on the mouse genome. (12) These loci may be resolvable to the level of individual genes whose function could represent conserved mechanisms also contributing to human neuroticism. By analogy with other genes linked to common diseases like diabetes, this may lead to the development of novel models of vulnerability to mood disorder which have genuine neurobiological validity.

When animals are selected for differences in emotional behaviour they also show different hypothalamic–pituitary–adrenal (HPA) axis function. Specifically, Roman high- and low-avoidance rats differentially acquire a two-way active avoidance response in a shuttle box. High-avoidance animals show greater prolactin and HPA axis responsivity to stress compared with low-avoidance animals. However, young Roman strain rats show identical HPA axis reactivity, although prolactin responses and behaviour are different. (13) In other words, reactivity to the environment may share a measure of common genetic control across physiological and behavioural domains, but HPA abnormality per se develops secondary to emotional experience, or at least is magnified by it. However, the key point is that important differences in stress reactivity may reflect a genetic predisposition and have a biological basis.

Early adverse experience

Adverse childhood experience was identified in genetically uncontrolled studies as a risk factor predisposing women to subsequent depression and has been confirmed in genetically informative designs. (11,14) In a clinical context, such developmental or social effects are usually viewed as separable from biology. Indeed, their very existence is usually taken to validate a ‘social' approach to psychiatry. From a more unified point of view,however, one would predict measurable neurobiological consequences. In fact, the effects have proved to be more profound than most biologists anticipated.

Variations in maternal care produce individual differences in neuroendocrine responses to stress in rats. The offspring of mothers that exhibited more licking and grooming of pups during the first 10 days after birth showed, in adult life, reduced plasma ACTH and corticosterone responses to acute stress. (15) In addition, there was increased hippocampal glucocorticoid-receptor messenger RNA (mRNA) expression, enhanced glucocorticoid feedback sensitivity, and decreased levels of hypothalamic corticotrophin-releasing hormone (CRH) mRNA. Greater early maternal attention also substantially reduced subsequent behavioural fearfulness in response to novelty, increased benzodiazepine receptor density in the amygdala and locus coeruleus, increased a (2) -adrenoreceptor density in the locus coeruleus, and decreased CRH receptor density in the locus coeruleus. Thus, maternal care serves to programme behavioural responses to stress in the offspring by altering the development of the neural systems that mediate fearfulness.

When BALB/cByJ mice were raised by an attentive C57BL/6ByJ dam, their excessive stress-elicited HPA activity was reduced, as were their behavioural impairments. However, cross-fostering the more resilient C57BL/6ByJ mice to an inattentive BALB/cByJ dam failed to elicit behavioural disturbances. In other words, vulnerable offspring may have their problems exacerbated by maternal behaviour, while early-life manipulations may have less obvious effects in relatively hardy animals. (16) Whether separation or stress paradigms in rodents can be taken as precise models of the mechanisms underlying the risk of mood disorder or other psychiatric problems cannot yet be decided, but their general relevance to the human case seems obvious.

Gene–environment interaction is the likely basis of the neurobiology of mood disorder. In general terms this must be correct. Whether the genetic mechanisms can be brought into sufficient focus to allow specific pathways to be identified remains the major current challenge. There is some reason for caution. The genetic and developmental routes into distal common pathways regulating stress responses may be very numerous. Disorders that are both common and very variable in expression, such as depression, may turn out to have little specificity that is worth talking about. Every illness may be an ensemble of many specific factors, none of which is individually going to lead to a more focused treatment or a better prediction of treatment response. We shall see.

The neurobiology of life events

Like early adversity, the role of life events in depression has been affirmed in genetically controlled studies. (11,14) Life events are relevant to almost all first episodes of depression, but are less significant in its recurrence. The biology of life events will be subsumed in the biology of stress, at best a clumsy term. In human studies it will always be difficult to isolate the critical ingredients of a particular psychological stress from the individual differences that stressed individuals bring to their experience. There have been several small neurobiological studies of well-defined events such as bereavement. Parents who had experienced the sudden death of a formerly healthy child showed immunological changes (decreased T-suppressor cells, significantly increased T-helper cells) and depression compared with controls, but no difference in cortisol levels.

Disturbances of immune function also occur in depressed patients. The evidence so far is essentially correlational and establishes no direction of effect. Indeed, the functional changes implied by different proportions of active lymphocytes or such similar ‘measures' are uncertain. Peripheral and central injections of interleukin 1 and lipopolysaccharide induce the expression of proinflammatory cytokines in animal brain and depress spontaneous and learned behaviours. This may reflect the psychological effects more specifically associated with physical illness, although it is possible that immune mechanisms may sometimes produce psychiatric symptoms, perhaps as part of a more generalized stress response. Drugs that potently modify the immune response should be examined for psychotropic properties. 

Biological studies of the depressed state

In the majority of biological studies of affective disorder patients have been studied when ill and compared with normal controls. Over the years, this kind of design has produced a range of positive findings, usually of modest effect. It remains true to say that no biological changes have ever been found that distinguish between depressed patients and controls better than does the clinical assessment of the patients. What is curious, and not a little tantalizing, is the impression that symptoms may, in part, represent biological adaptations directed to put things right. Thus, on the one hand, there may be consistent effects upon hormone secretion or sleep that represent phenomena of illness. On the other, deliberate changes in hormone status or sleep deprivation may modify the state of depression.

The depressed state: functional anatomy

Perfusion or metabolic imaging can indirectly detect changes in neuronal activity. Signals can be well localized, but they may reflect either reversible changes in function or a permanent loss of neuronal activity. Reductions in function in anterior brain structures have been typical in major depression. Hypoperfusion tends to be greatest in frontal, temporal, and parietal areas and most extensive in older patients; high Hamilton scores tend to be associated with reduced perfusion. (17) Reductions in frontal areas may be more likely in patients with impoverished mental states. Thus, neuropsychological testing in major depression shows evidence of slowing in motor and cognitive domains, with additional prominent effects on mnemonic function that are most marked in the elderly. These effects are correlated with reduced frontal perfusion in the elderly. In younger patients, there may actually be increased perfusion in the frontal and cingulate cortex. Metabolic increases in the cingulate gyrus have been associated with a good treatment response. (18) Highly localizing findings have been unusual, however. The only exceptions have been within-subject changes on recovery in the mesial frontal cortex and perhaps the basal ganglia. (19)

Isotope-based imaging of receptor occupation in depression has so far been limited. Single-photon emission tomography ( SPET) with the dopamine D 2/3 ligand [ 123 I] IBZM showed increased binding in the striatum. (20) There were significant correlations between IBZM binding in the left and right striatum and measures of reaction time and verbal fluency, but not of mood as such. Increased D 2/3 binding in the striatum probably reflects a reduced dopamine function, whether due to a reduced release or secondary upregulation of receptors. Positron emission tomography (PET) with the selective radioligand [ 18 F]altanserin maps 5-HT (2) -receptor distribution; in a small sample of depressed patients, tracer uptake was significantly reduced in the posterolateral orbitofrontal cortex and the anterior insular cortex, and was most marked in the right hemisphere. (21)

In summary, much of the functional imaging work so far completed has been inconclusive. It has served to implicate frontal and limbic rather than posterior brain areas, but this has done little more than broadly confirm anatomical conclusions derived from observing the effects of lesions or brain stimulation. (19) It has failed to illuminate mechanisms underlying the depressed state either at the neuropsychological or neurochemical level. Progress requires that relevant neuropsychological and drug challenges are incorporated into imaging protocols. The likely neuropsychological candidates include the representation of facial emotion; the choice of relevant drug challenges will be guided, as outlined below, by experience in neuroendocine challenge studies and the pharmacology of effective treatments. Finally, ‘functional' abnormalities may importantly predict structural abnormality in depression.

Neuroendocrine challenge tests

Secretion of hormones in the anterior pituitary is under the control, both direct and indirect, of central neuronal cell bodies that may project relatively widely within the brain. The secretion of a given hormone in response to specific precursors or agonists for individual neurotransmitter receptors has been proposed as a way of testing the security of such connections. Hormone secretion provides a bioassay of the system of interest. There is a measure of consensus about the findings in major depression.

Precursor loading of the serotonergic system with intravenous tryptophan produces prolactin and growth hormone secretion. In depressed patients, prolactin and growth hormone responses are blunted but recover with treatment. (22) Responses to intravenous clomipramine, which similarly releases 5-HT by a presynaptic mechanism, are also blunted in depression. (23) The receptors mediating these responses are uncertain, although in any case the requirement for an intact presynaptic system precludes interpretation in terms of receptor subtype.

Appropriately selective agonists with full efficacy are not freely available for testing postsynaptic or presynaptic responses. However, buspirone, gepirone, and ipsapirone are all partial agonists at the 5-HT 1A receptor and have been looked at in a preliminary way in depressed patients. There is disagreement about whether postsynaptic (endocrine) responses are or are not blunted, but there is more consensus that presynaptic (hypothermic) responsesare blunted. (24)

Noradrenergic and dopaminergic function has been examined using the a (2)-adrenoceptor agonist clonidine and the mixed D 1/2 receptor agonist apomorphine. Growth hormone responses to clonidine are usually reported as being blunted in major depression. (25) Apomorphine-stimulated growth hormone responses are also blunted in depressed patients, (26) while, in symmetry, patients at risk of postpartum psychosis (usually manic in form) have enhanced apomorphine responses. (27)

Cholinergic challenge with an acetylcholinesterase inhibitor such as pyridostigmine or physostigmine produces a secretion of growth hormone; responses in major depression were increased compared with controls. (28)

Neuroendocrine drug challenge suggests attenuated serotonergic function and increased cholinergic function in depression. Reduced responses to adrenergic and dopaminergic challenge also suggest impaired neurotransmission. Interpretation of tests with agonists is always difficult, because blunting may occur in an overactive system that has been downregulated. In addition, if the secretion of the assay hormone itself is actually directly affected by the state of depression, interpretation in terms of specific neurotransmitter abnormalities may be misleading. This is a particular problem for ACTH/cortisol responses (see below). In fact, enthusiasm for neuroendocrine surrogate markers of monoamine transmission within the brain has probably diminished in recent years, but the paradigm of drug challenge nevertheless remains interesting. We must assay brain responses of the monoamine projections more centrally involved in mood regulation.


About half of all patients with major depression have a raised cortisol output, which tends to return to normal on recovery. It is most consistently associated with an ‘endogenous' pattern of illness. While cortisol is always regarded as a ‘stress' hormone, and is secreted in response to various types of acute stress, the stresses that commonly result in long-term hypercortisolaemia are poorly understood. The idea that there is a relatively specific link between chronic high cortisol levels and mood disorder is notably persistent. In major depression there is peripheral hypertrophy of the adrenal glands, measurable in MRI body scans, and an enhanced response to corticotrophin. The MRI change, like the hypercortisolaemia itself, reverses on recovery. (29)

Suppression of cortisol secretion occurs normally via glucocorticoid receptor-mediated inhibitory feedback to the hypothalamus; it is readily produced by dexamethasone, which is a potent exogenous glucocorticoid (the dexamethasone suppression test (DST)). Non-suppression of endogenous cortisol after dexamethasone occurs in Cushing's disease for example. It implies either reduced feedback and/or enhanced central drive to release cortisol. It was initially observed that the 1-mg DST showed high specificity (96 per cent) and sensitivity (67 per cent) as a putative diagnostic test for melancholia. (30) At the time this result attracted intense interest, but has since proved difficult to generalize. The high specificity established against normal controls was less against other patient groups. Thus, DST non-suppression has not been accepted as a diagnostic test. This failed effort to give medical respectability to psychiatric diagnosis came to devalue what remains an important observation. Non-suppression usually reflects hypercortisolaemia, which is itself a robust phenomenon of mood disorder that requires explanation like any other core biological symptom. Other symptoms that we identify as part of the depressive syndrome lend themselves less easily to investigation. The DST also has potential clinical uses beyond diagnosis. DST non-suppression predicts a low placebo response rate to drug treatment, (31) and hypercortisolaemia predicts a low rate of clinical response to psychological intervention. (32)

It remains unclear whether cortisol contributes to the clinical state of depression by a direct action on the brain. Exogenous cortisol administration is associated with affective symptoms, and chronic excessive cortisol secretion commonly appears to produce depressive symptoms in Cushing's disease. An HPA axis programmed to hypersecrete cortisol under stress could be a pathogenic mechanism explaining why depression or mania develops. This view has provoked efforts to treat mood disorder by inhibition of cortisol synthesis with metyrapone. Preliminary results suggest this may be effective.

However, when depressed patients are given large doses of cortisol they tend to show acute mood enhancement, (33) and oral dexamethasone has been reported to elevate mood in major depression, especially in hypersecretors. (34) This leads to the converse hypothesis that an HPA axis appropriately adapted to chronic stress early in development might be unable to mount a normal effective response to acute stress later in life. Cortisol may then be seen as a euphoriant (or antidepressant), and hypercortisolaemia as an antidepressant response of the stress-regulating mechanisms of the brain. Based on this view, all cortisol levels seen in depression may be set inappropriately low for the ongoing stress, however high or low they are compared with the normal range.

Whether one supposes cortisol levels to be set too high or too low in depression, it remains inconvenient that either a suppression or an augmentation of steroid effect seems, initially at least, to elevate mood. A way out of this complication may lie in cortisol's action on two receptors in the brain (the glucocorticoid and mineralocorticoid receptors) that may have opposite actions. However, we need better-controlled replicated data on the effects of steroid manipulations.

An increased cortisol production is associated with an increased release of hypothalamic b-endorphin (35) and probably a pulsatile increase in ACTH. The paraventricular nucleus of the hypothalamus represents the highest level of dedicated neurones in the HPA axis. The neurosecretory cells of the paraventricular nucleus release the peptides CRH and AVP into the portal hypophyseal blood. These hormones in turn stimulate the release of ACTH from the anterior pituitary. Major depression is characterized by a blunted ACTH response to CRH, (36) an elevated level of CRH in the cerebrospinal fluid, (37) and increased numbers of neurones expressing CRH mRNA in the paraventricular nucleus of the hypothalamus postmortem. (38) CRH is not confined to the paraventricular nucleus, but is expressed in a variety of other central nuclei whence it can produce anxiogenic behavioural effects. CRH receptors, which exist in two forms, are widely distributed in the hypothalamus and cortex. A related peptide, urocortin, has a similar pharmacology. Knocking out the CRH-1 receptor gene in mice impaired the HPA stress response and reduced anxiety-like behaviour. (39) Non-peptide CRH antagonists must be taken seriously as putative anxiolytics or antidepressants and are now in clinical trials. If effective, they will be among the first of a new generation of truly novel treatments based on peptide neurotransmission.

Thyroid abnormalities

In unselected major depression, thyroid hormone levels are usually normal, but there may be abnormalities of the thyrotropin (thyroid-stimulating hormone) response to thyrotropin-releasing hormone. The thyrotropin response is blunted in a significant number of patients, but this effect is poorly understood and has few accepted clinical associations. In contrast, a subgroup of patients may show an enhanced thyrotropin response with normal thyroid hormone levels (referred to as grade II hypothyroidism). These associations and the use of thyroid hormones in treatment suggest that there is more to be learned in this area.

Sleep disturbance

Sleep is often disturbed in depression but in a variety of ways. Early-morning waking is the most typical in endogenous or melancholic depression, with the sleep patterns in such patients being similar to those seen in patients with mania. Trouble getting to sleep, frequent wakings, and unsatisfactorily prolonged sleep are also commonly seen in depression. Like other biological manifestations of the disorder, the extent to which sleep is simply a consequence of the state of depression or a contribution to its biology is uncertain. Patients with severe depression or mania may respond to sleep deprivation with a transient elevation in mood. It implies that the sleep–wake cycle is directly involved with mood regulation and its disorder.

In severe depression (melancholia) the typical effects are a reduction in the total length of slow-wave sleep and a shortened latency in the appearance of rapid eye movement (REM) or dreaming sleep. (40) The cholinergic projections from the hindbrain may be REM-ON cells, while serotonergic and noradrenergic cells may be REM-OFF cells. The disturbed sleep of depression could be due to an increased cholinergic and/or a decreased serotonergic/noradrenergic drive; simplistic though it sounds, the experimental evidence is supportive. Depressed patients challenged with a cholinergic agonist in the second non-REM period enter REM significantly faster than psychiatric and normal control subjects. (41) The reduced sensitivity of the noradrenergic system is suggested because clonidine fails to suppress REM in depressed patients compared with controls. (42) Tryptophan depletion (to attenuate 5-HT function) partially mimics the changes seen in depression in recovered patients. (43)

Sleep tends to recover on recovery from depression, and the tricyclic antidepressants in particular suppress REM sleep. However, sleep disturbance may be an early predictor of relapse, and disturbed sleep parameters predict a poor response to cognitive–behaviour therapy. (44) Indeed, depressed patients may have inherently weak slow-wave sleep processes because unaffected subjects with a family history of depression show reduced slow-wave sleep and increased REM density in the first sleep cycle. (45)

Monoamine metabolite turnover

The earliest studies to investigate the actions of tricyclic antidepressants highlighted their actions on the turnover of the monoamine metabolites in animal brain. The ‘monoamine theory of depression' proposed the reduced functioning of monoamine transmission in depression. Therefore it was natural to seek relevant measures of monoamine chemistry in the cerebrospinal fluid of patients and controls. The study of what became irreverently known as ‘neuralurine' and indeed of urine itself, since peripheral measures of monoamine turnover are also potentially relevant, virtually defined a decade of biological psychiatry in the 1970s and 1980s. Drugs had similar effects on neurotransmitter turnover as seen in animal studies, demonstrating that the human techniques were sufficiently sensitive. Indeed the monoamine theory is, at its best, a theory about drug action because the monoamine and metabolite changes produced by illness in patients have proved remarkably unconvincing. (46) The findings for the noradrenaline metabolite MHPG and the 5-HT metabolite 5-hydroxyindoleacetic acid were negative. The dopamine metabolite homovanillic acid did show the predicted decrease,but only significantly in women. There were trends to modest increases in all the major metabolites in mania. Although disappointing, cerebrospinal fluid studies could never reflect the activities of smaller groups of neurones localized in areas critical for the modulation of mood. Such a focus is only possible in isotope imaging (PET or SPET) or better postmortem studies of the brain. 

Tryptophan depletion

The most convincing evidence that 5-HT is intimately involved in mood disorder has come from depletion of tryptophan, the amino acid precursor of 5-HT. The level of tryptophan in both peripheral blood and the brain can be driven to very low levels by a short-term low-protein diet and subsequent loading with large neutral amino acids. These compete with tryptophan for access to the brain amino acid transporter and also increase its peripheral metabolism, which results in the reduced synthesis and release of 5-HT. Initial observations appeared to bear primarily on the mechanism of drug action. Thus, patients who had recovered from major depression while taking a serotonin-selective reuptake inhibitor, experienced a clear-cut return of severe symptoms lasting for several hours after tryptophan depletion. This finding has now been critically extended to patients with a history of recurrent major depression who were euthymic but not taking any medication. (47) Prominent objective symptoms of retardation and cognitive distortion returned in a stereotyped and severe way, reflecting previous symptoms. The effects on mood in patients who have had a previous episode of depression are qualitatively different from the more minor changes seen in normal female controls or even subjects with a strong family history. This may imply the formation of a form of neurobiological template, which increases the vulnerability to subsequent relapse or recurrence. The immediacy of the link between neurotransmitter function and symptoms may be the reason why patients with recurrent major depression need long-term treatment with antidepressant drugs to remain well. 

Does mood disorder have a functional neuropathology?

Severe mood disorder is virtually defined by its frequent recurrence or its chronicity. The first episodes of severe depression occur more frequently with increasing age and tend to be more refractory to treatment. Severe mood disorder is associated with ventricular enlargement and sulcal prominence (48) and persistent cognitive impairment, again most strikingly in older patients. (49) Therefore something may be permanently wrong in the brains of patients with particularly intractable mood disorder and it may be linked to poor outcome. (50) Indeed, there is an increased rate of white matter lesions, perhaps related to vascular disease, in older patients. (51) However, the key hypothesis must be that it is the particular pattern of functional disruption resulting from any cellular pathology that increases the risk of depression. It may be reasonable to describe such a change as a functional neuropathology.

In younger patients with unusual refractoriness and chronicity, MRI scanning again suggested reduced grey matter parameters, most significantly in the left hippocampus but also more diffusely in the left parietal and frontal association cortices. Left hippocampal grey matter density was correlated with measures of verbal memory, supporting the functional significance of the imaging changes. In contrast, patients with severe illnesses fully responsive to treatment showed no differences from controls. Any finding in the chronic group could predate the onset of depression, or be the result of the illness process or its treatment. It is fashionable to attribute structural changes in depression to hypercortisolaemia, but in this study that was not the explanation. (20)

It has been claimed recently that a lobule of inferior frontal cortex, thought to be a critical node for the integration of mood, is atrophic in major depressive illness. Preliminary histopathological assessments of postmortem tissue have suggested that the decrement in grey matter volume is associated with a reduction in glia without an equivalent loss of neurones. (52) This is an unusual finding and requires replication. However, it highlights the fact that postmortem studies of the brain in mood disorder have been rare and seldom focused on ‘candidate regions' such as the inferior frontal or cingulate cortex, amygdala, or hippocampus. Such studies in elderly depression have greater potential validity than the much more numerous investigations of schizophrenia; a definitive study is awaited. 

Postmortem studies can also address the neurochemistry, perhaps more directly and completely than other methods. Normal ageing is accompanied by a decline in a variety of indices of monoamine function including presynaptic markers of 5-HT innervation. There is some evidence for a more reduced binding at these sites postmortem in depression, (53) suicide, (54) and cases of Alzheimer's disease with depression. (55) Whether a reduced serotonergic innovation is the critical change that increases the vulnerability to mood disorder of patients with advancing years is not yet established. If so, the potential for MDMA to have long-term effects in heavy users is real and worrying. (56)

In suicide, postmortem findings have broadly paralleled those in depression, with an important emphasis on 5-HT metabolism and neurotransmission. Whether 5-HT neurotransmission, perhaps like that involving the other monoamines, represents a functional domain implicated independently in a variety of psychiatric syndromes and behaviours remains to be well established.

The neurobiology of treatments

It seems quite likely that our understanding of the mechanism of action of drugs or treatments such as electroconvulsive shock, which are antidepressant, antimanic, or mood stabilizing (effective against both poles of bipolar illness), will continue to stimulate ideas about the neurobiology of the illnesses they treat. This approach results from a rather simple linking of the drugs' actions to the possible pathophysiology that they may be correcting. For the first time we are now seeing drugs that have been developed from plausible extrapolations of the neurobiology. The potential of CRH antagonists has already been noticed; neurokinin receptor antagonists that will prevent the action of substance P, another peptide transmitter with a stress profile, also appear to be putative antidepressant drugs. (57)


Mood disorder has an important neurobiological basis. This stretches from a vulnerability, which seems to be attributable to polymorphism in genes critical to stress regulation, through the impact that early experience has on the subsequent programming of the brain for stress responses, to the final responsiveness when encountering particular personal adversity in later life. Biological studies have highlighted the role of key brain areas within the limbic system such as the cingulate cortex and amygdala. We are still a long way from understanding, with any precision, the critical connections and cellular mechanisms, but the function of monoamine neurones generally, and serotonergic projections in particular, is closely associated with mood regulation. Peptide neurotransmitters have long seemed likely to play a central role in stress regulation, and their potential as targets for antidepressant drug action may yet be fulfilled. Finally, observations in the most chronic illnesses and in the elderly with depression have highlighted the possibility of a functional neuropathology underlying severe mood disorder that remains poorly understood.

The approaches of clinicians to the phenomenon of depression still polarize around the biological and the psychosocial. The main purpose is sometimes little more than an assertion of professional territory. This is both regrettable and unnecessary. Any account of depression that claims to be purely social or even psychological misses the point that we are also, in our natures, biological. It is possible to embrace the biology of depressive illness as a fact, while simultaneously acknowledging that it is expressed, and in particular experienced, by patients in psychological terms. Dualism has done psychiatry no favours by teaching us to create a dichotomy between a world of brainless minds and another of mindless brains. The fact that there is an underlying biology which can be unified with a brain-based psychology is, in itself, evidence neither for nor against the likely effect of biological versus psychological treatment. Only with an integrated understanding of the biology and the psychology of mood disorder will we understand the potential and limitations for treatments based on drugs or talking.


1.Papez, J.W. (1937). A proposed mechanism of emotion. Archives of Neurology and Psychiatry, 38, 725–43.

2.Adolphs, R., Tranel, D., Damasio, H., and Damasio, A.R. (1995). Fear and the human amygdala. Journal of Neuroscience, 15, 5879–91.

3.Whalen, P.J., Rauch, S.L., Etcoff, N.L., McInerney, S.C., Lee, M.B., and Jenike, M.A. (1998). Masked presentations of emotionalfacial expressions modulate amygdala activity without explicit knowledge. Journal of Neuroscience, 18, 411–18.

4.George, M.S., Ketter, T.A., Parekh, P.I., Herscovitch, P., and Post, R.M. (1996). Gender differences in regional cerebral bloodflow during transient self-induced sadness or happiness. Biological Psychiatry, 40, 859–71.

5.Damasio, A.R. (1995). Descartes' error. Emotion, reason and the human brain. Picador, London.

6.Hornak, J., Rolls, E.T., and Wade, D. (1996). Face and voice expression identification in patients with emotional and behavioural changes following ventral frontal lobe damage. Neuropsychologia, 34, 247–61.

7.Green, A.R., Cross, A.J., and Goodwin, G.M. (1995). Review of the pharmacology and clinical pharmacology of 3,4-methylenedioxymethamphetamine (MDMA or ‘ecstasy'). Psychopharmacology, 119, 247–60.

8.Battersby, S., Ogilvie, A.D., Smith, C.A.D., et al.(1996). Structure of a variable number tandem repeat of the serotonin transporter gene and association with affective disorder. Psychiatric Genetics, 6, 177–81.

9.Heils, A., Teufel, A., Petri, S., et al.(1996). Allelic variation of human serotonin transporter gene expression. Journal of Neurochemistry, 66, 2621–4.

10.Jonsson, E.G., Nothen, M.M., Gustavsson, J.P., et al.(1998). Polymorphisms in the dopamine, serotonin, and norepinephrine transporter genes and their relationships to monoamine metabolite concentrations in CSF of healthy volunteers. Psychiatry Research, 79, 1–9.

11.Kendler, K.S., Kessler, R.C., Neale, M.C., Heath, A.C., and Eaves, L.J. (1993). The prediction of major depression in women: toward an integrated etiologic model. American Journal of Psychiatry, 150, 1139–48.

12.Flint, J., Corley, R., DeFries, J.C., et al.(1995). A simple genetic basis for a complex psychological trait in laboratory mice. Science, 269, 1432–5.

13.Castanon, N., Dulluc, J., Le Moal, M., and Mormede, P. (1994). Maturation of the behavioural and neuroendocrine differences between the Roman rat lines. Physiology and Behavior, 55, 775–82.

14.Kendler, K.S. and Karkowski-Shuman, L. (1997). Stressful life events and genetic liability to major depression: genetic controlof exposure to the environment? Psychological Medicine, 27, 539–47.

15.Liu, D., Diorio, J., Tannenbaum, B., et al. (1997). Maternal care, hippocampal glucocorticoid receptors, and hypothalamic–pituitary–adrenal responses to stress. Science, 277, 1659–62.

16.Caldji, C., Tannenbaum, B., Sharma, S., Francis, D., Plotsky, P.M., and Meaney, M.J. (1998). Maternal care during infancy regulates the development of neural systems mediating the expression of fearfulness in the rat. Proceedings of the National Academy of Sciences of the United States of America, 95, 5335–40.

17.Goodwin, G.M. (1996). Functional imaging, affective disorder and dementia. British Medical Bulletin, 52, 495–512.

18.Mayberg, H.S., Brannan, S.K., Mahurin, R.K., et al.(1997). Cingulate function in depression: a potential predictor of treatment response. Neuroreport, 8, 1057–61.

19.Goodwin, G.M. (1997). Neuropsychological and neuroimaging evidence for the involvement of the frontal lobes in depression. Journal of Psychopharmacology, 11, 115–22.

20.Shah, P.J., Ebmeier, K.P., Glabus, M.F., and Goodwin, G.M. (1998). Cortical grey matter reductions associated with treatment-resistant chronic unipolar depression: controlled magnetic resonance imaging study. British Journal of Psychiatry, 72, 527–32.

21.Biver, F., Wikler, D., Lotstra, F., Damhaut, P., Goldman, S., and Mendlewicz, J. (1997). Serotonin 5-HT-2 receptor imaging in major depression: focal changes in orbito-insular cortex. British Journal of Psychiatry, 171, 444–8.

22.Upadhyaya, A.K., Pennell, I., Cowen, P.J., and Deakin, J.F.W. (1991). Blunted growth hormone and prolactin responses to l-tryptophan in depression; a state-dependent abnormality. Journal of Affective Disorders, 21, 213–18.

23.Anderson, I.M., Ware, C.J., Da Roza, J.M., and Cowen, P.J. (1992). Decreased 5-HT-mediated prolactin release in major depression. British Journal of Psychiatry, 160, 372–8.

24.Cowen, P.J., Power, A.C., Ware, C.J., and Anderson, I.M. (1994). 5-HT (1A) receptor sensitivity in major depression. A neuroendocrine study with buspirone. British Journal of Psychiatry, 164, 372–9.

25.Schittecatte, M., Charles, G., Machowski, R., et al.(1994). Effects of gender and diagnosis on growth hormone response to clonidine for major depression: a large-scale multicenter study. American Journal of Psychiatry, 151, 216–20.

26.Pitchot, W., Hansenne, M., Moreno, A.G., and Ansseau, M. (1995). Effect of previous antidepressant therapy on the growth hormone response to apomorphine. Neuropsychobiology, 32, 19–22.

27.Wieck, A., Kumar, R., Hirst, A.D., Marks, M.N., Campbell, I.C., and Checkley, S.A. (1991). Increased sensitivity of dopamine receptors and recurrence of affective psychosis after childbirth. British Medical Journal, 303, 613–16.

28.O'Keane, V., O'Flynn, K., Lucey, J., and Dinan, T.G. (1992). Pyridostigmine-induced growth hormone responses in healthy and depressed subjects: evidence for cholinergic supersensitivity in depression. Psychological Medicine, 22, 55–60.

29.Rubin, R.T., Phillips, J.J., Sadow, T.F., and McCracken, J.T. (1995). Adrenal gland volume in major depression: increase duringthe depressive episode and decrease with successful treatment. Archives of General Psychiatry, 52, 213–18.

30.Carroll, B.J., Feinberg, M., Greden, J.F., et al.(1981). A specific laboratory test for the diagnosis of melancholia. Standardization, validation, and clinical utility. Archives of General Psychiatry, 38, 15–22.

31.Ribeiro, S.C.M., Tandon, R., Grunhaus, L., and Greden, J.F. (1993). The DST as a predictor of outcome in depression: a meta-analysis. American Journal of Psychiatry, 150, 1618–29.

32.Thase, M.E., Dube, S., Bowler, K., et al.(1996). Hypothalamic–pituitary–adrenocortical activity and response to cognitive behavior therapy in unmedicated, hospitalized depressed patients. American Journal of Psychiatry, 153(Supplement), 886–91.

33.Goodwin, G.M., Muir, W.J., Seckl, J.R., et al.(1992). The effects of cortisol infusion upon hormone secretion from the anterior pituitary and subjective mood in depressive illness and in controls. Journal of Affective Disorders, 26, 73–83.

34.Dinan, T.G., Lavelle, E., Cooney, J., et al.(1997). Dexamethasone augmentation in treatment-resistant depression. Acta Psychiatrica Scandinavica, 95, 58–61.

35.Goodwin, G.M., Austin, M.P., Curran, S.M., et al.(1993). The elevation of plasma beta-endorphin levels in major depression. Journal of Affective Disorders, 29, 281–9.

36.Heuser, I., Yassouridis, A., and Holsboer, F. (1994). The combined dexamethasone/CRH test: a refined laboratory test for psychiatric disorders. Journal of Psychiatric Research, 28, 341–56.

37.Mitchell, A.J. (1998). The role of corticotropin releasing factor in depressive illness: a critical review. Neuroscience and Biobehavioral Reviews, 22, 635–51.

38.Raadsheer, F.C., Van Heerikhuize, J.J., Lucassen, P.J., Hoogendijk, W.J.G., Tilders, F.J.H., and Swaab, D.F. (1995). Corticotropin-releasing hormone mRNA levels in the paraventricular nucleus of patients with Alzheimer's disease and depression. American Journal of Psychiatry, 152, 1372–6.

39.Timpl, P., Spanagel, R., Sillaber, I., et al.(1998). Impaired stress response and reduced anxiety in mice lacking a functional corticotropin-releasing hormone receptor 1. Nature Genetics, 19, 162–6.

40.Berger, M. and Riemann, D. (1993). REM sleep in depression—an overview. Journal of Sleep Research, 2, 211–23.

41.Gillin, J.C., Sutton, L., Ruiz, C., et al.(1991). The cholinergic rapid eye movement induction test with arecoline in depression. Archives of General Psychiatry, 48, 264–70.

42.Schittecatte, M., Garcia Valentin, J., Charles, G., et al.(1995). Efficacy of the ‘clonidine REM suppression test (CREST)' to separate patients with major depression from controls; a comparison with three currently proposed biological markers of depression. Journal of Affective Disorders, 33, 151–7.

43.Moore, P., Gillin, J.C., Bhatti, T., et al.(1998). Rapid tryptophan depletion, sleep electroencephalogram, and mood in men with remitted depression on serotonin reuptakeinhibitors. Archives of General Psychiatry, 55, 534–9.

44.Thase, M.E., Simons, A.D., and Reynolds, C.F., III. (1993). Psychobiological correlates of poor response to cognitive behavior therapy. Potential indications for antidepressant pharmacotherapy. Psychopharmacology Bulletin, 29, 293–301.

45.Lauer, C.J., Schreiber, W., Holsboer, F., and Krieg, J.C. (1995). In quest of identifying vulnerability markers for psychiatricdisorders by all-night polysomnography. Archives of General Psychiatry, 52, 145–53.

46.Schatzberg, A.F., Samson, J.A., Bloomingdale, K.L., et al.(1989). Toward a biochemical classification of depressive disorders. X. Urinary catecholamines, their metabolites, and D-type scores in subgroups of depressive disorders. Archives of General Psychiatry, 46, 260–8.

47.Smith, K.A., Fairburn, C.G., and Cowen, P.J. (1997). Relapse of depression after rapid depletion of tryptophan. Lancet, 349, 915–19.

48.Elkis, H., Friedman, L., Wise, A., and Meltzer, H.Y. (1995). Meta-analyses of studies of ventricular enlargement and cortical sulcal prominence in mood disorders: comparisons with controls or patients with schizophrenia. Archives of General Psychiatry, 52, 735–46.

49.Abas, M.A., Sahakian, B.J., and Levy, R. (1990). Neuropsychological deficits and CT scan changes in elderly depressives. Psychological Medicine, 20, 507–20.

50.Simpson, S., Baldwin, R.C., Jackson, A., and Burns, A.S. (1998). Is subcortical disease associated with a poor response to antidepressants? Neurological, neuropsychological and neuroradiological findings in late-life depression. Psychological Medicine, 28, 1015–26.

51.Coffey, C.E., Figiel, G.S., Djang, W.T., Cress, M., Saunders, W.B., and Weiner, R.D. (1988). Leukoencephalopathy in elderly depressed patients referred for ECT. Biological Psychiatry, 24, 143–61.

52.Drevets, W.C., Ongur, D., and Price, J.L. (1998). Neuroimaging abnormalities in the subgenual prefrontal cortex: implications for the pathophysiology of familial mood disorders. Molecular Psychiatry, 3, 220–6.

53.Lawrence, K.M., Kanagasundaram, M., Lowther, S., Katona, C.L.E., Crompton, M.R., and Horton, R.W. (1998). [ 3 H] imipramine binding in brain samples from depressed suicides and controls: 5-HT uptake sites compared with sites defined by desmethylimipramine. Journal of Affective Disorders, 47, 1–3.

54.Mann, J.J. (1998). The neurobiology of suicide. Nature Medicine, 4, 25–30.

55.Chen, C.H., Alder, J.T., Bowen, D.M., et al.(1996). Presynaptic serotonergic markers in community-acquired cases of Alzheimer's disease: correlations with depression and neuroleptic medication. Journal of Neurochemistry, 66, 1592–8.

56.Green, A.R. and Goodwin, G.M. (1996). Ecstasy and neurodegeneration. British Medical Journal, 312, 1493–4.

57.Kramer, M.S., Cutler, N., Feighner, J., et al.(1998). Distinct mechanism for antidepressant activity by blockade of central substance P receptors. Science, 281, 1640–5