To understand the biological underpinnings of multi-determined behaviours such as suicide and attempted suicide it is necessary to situate them within an explanatory model that can elaborate the causal pathways and interrelations between biological, clinical, genetic, and environmental factors that all play a role in suicidal behaviour. Where possible, such a model should be clinically explanatory, incorporate biological correlates, be testable in both clinical and biological studies, and have some utility in identifying high-risk individuals.

We have proposed a stress-diathesis model of suicidal behaviour wherein exposure to a stressor precipitates a suicidal act in those with the diathesis, or propensity, for suicidal behaviour.⁽¹⁾ Stressors are generally state-dependent factors such as an episode of major depression or adverse life event. The diathesis, we have hypothesized, comprises trait characteristics such as impulsive aggression, and pessimism.⁽¹⁾ Uncovering the biological mechanisms relevant to the stress and the diathesis dimensions of suicidal behaviour will facilitate the identification of both enduring and proximal markers of risk, as well as potential targets for treatment.

One biological correlate of the diathesis for suicidal behaviour appears to be low serotonergic activity. Abnormal serotonergic function may be the result of numerous factors including genetics, early life experience, chronic medical illness, alcoholism or substance use disorder, many of which have been correlated with increased risk for suicidal behaviour. Moreover, serotonergic dysfunction may underlie recurrent mood disorders or behavioural traits that characterize the diathesis, such as aggression and impulsivity. In terms of stress response, the noradrenergic and HPA axis have been the focus of biological studies in suicidal behaviour. This chapter gives an overview of the major neurobiological findings in suicide and attempted suicide, as well as emerging findings from studies of genes related to those systems.

Serotonin is involved in brain development, behavioural regulation, modulation of sleep, mood, anxiety, cognition, and memory and is shown to be disturbed in various psychiatric disorders. Serotonergic function is under genetic control and, moreover, deficits in functioning have been shown to be enduring, marking it as a biological trait. The serotonergic system became a target for investigation in relation to suicide when, more than 30 years ago, Asberg and colleagues observed that depressed individuals who had either attempted suicide by violent means or subsequently died by suicide in the study follow-up period were more likely to have lower CSF 5-HIAA levels.⁽²⁾ Since that time the function of the serotonergic system in suicide and attempted has been examined in many paradigms, and while not all studies agree, there is substantial consensus that individuals who die by suicide, or make serious non-fatal suicide attempts, exhibit a deficiency in CNS serotonin neurotransmission.

Evidence of hypofunction comes from cerebrospinal fluid and postmortem studies. 5-hydroxyindoleacetic acid (5-HIAA) is the major metabolite of serotonin and level of CSF 5-HIAA is a guide to serotonin activity in parts of the brain including the prefrontal cortex. There have been over 20 studies of CSF 5-HIAA and suicidal behaviour in mood disorders, and a meta-analysis of prospective studies of 5-HIAA found that in mood disorders lower CSF 5-HIAA increased the chance of death by suicide over fourfold over follow-up periods of 1-14 years.⁽³⁾

Multiple postmortem studies of suicide, report lower brainstem levels of 5-HIAA and serotonin (5-hydroxytryptamine, 5-HT) (see Mann et al. for a review⁽⁴⁾). These deficits in 5-HT or 5-HIAA are observable across diagnostic groups⁽⁵⁾ and, despite early reports to the contrary, independent of suicide method. This abnormality appears to be largely specific to the brainstem, and multiple studies have reported no differences between suicides and controls in 5-HT level in other brain regions including the hippocampus, the occipital cortex, the frontal cortex, the temporal cortex, the caudate, the striatum, or the hypothalamus.⁽⁴⁾ Serotonin neurone cell bodies are in the brainstem raphe nuclei, while their axons innervate most of the brain including the ventral prefrontal cortex. Morphological analysis of stained serotonin neurones in the brainstem of depressed suicides and non-suicides observed greater cell density in the dorsal raphe nucleus in the suicides⁽⁶⁾ suggesting that reduction in serotonin activity is associated with dysfunctional neurones and not with fewer neurones.

Neuroendocrine challenge studies using fenfluramine provide further evidence of anomalous serotonergic function associated with suicidal behaviour. Fenfluramine is a serotonin-releasing drug and a reuptake inhibitor that may also directly stimulate postsynaptic 5-HT receptors. The release of serotonin by fenfluramine causes a measurable increase in serum prolactin levels that is an indirect index of central serotonergic responsiveness. In depressed patients, those with a history of suicide attempts have a more blunted prolactin response to fenfluramine challenge than non-attempters with some evidence that the effect is more strongly related to seriousness of past attempt.⁽⁷⁾

Studies of receptors suggest lower serotonergic transmission in the central nervous system may be accompanied by a compensatory upregulation of some serotonergic postsynaptic receptors such as the 5-HT_1A and 5-HT_2A, and a decrease in the number of serotonin reuptake sites.⁽⁴⁾ There is a reported increase in the concentration of the postsynaptic 5-HT_2A receptors in the prefrontal cortex of suicides compared with non-suicides.⁽⁸⁾ This increased binding is reflected in more protein and may be due to elevated gene expression in youth suicide.⁽⁹⁾ Elevated 5-HT2A binding has also been reported in the amygdala in depressed suicides. In depressed and non-depressed suicides there is evidence that 5-HT_2A receptors are upregulated in the dorsal prefrontal cortex but not the rostral prefrontal cortex.⁽⁸⁾

Platelet studies examine 5-HT_2A in living subjects with respect to non-fatal suicide attempt. 5-HT_2A receptors, serotonin reuptake sites, and serotonin second messenger systems are present in blood platelets, and changes in these platelet measurements may reflect similar changes in the CNS. Multiple studies have reported higher platelet 5-HT_2A receptor numbers in suicide attempters compared with non-attempters and healthy controls.⁽¹¹⁾

Studies of second messengers indicate impaired 5-HT_2A receptor mediated signal transduction in the prefrontal cortex of suicides, (12) and in platelets 5-HT_2A receptor responsivity is significantly blunted in patients with major depression who have made high-lethality suicide attempts compared to depressed patients who have made low-lethality suicide attempts.⁽¹³⁾ The implications of such a defect in signal transduction, if present in the brain, would be that although there may be greater density of 5-HT_2A receptors, the signal transduced by 5-HT_2A receptor activation may be blunted, which would compound deficient serotonergic input as seen in the lower levels of brainstem serotonin and/or 5-HIAA in suicide victims.

Some postmortem studies of the postsynaptic 5-HT_1A receptor report higher binding in prefrontal cortex and more rostral segments of raphe nuclei, and lower binding in more caudal raphe nuclei, hippocampus, prefrontal cortex, and temporal cortex.⁽⁷⁾ Less 5-HT_1A autoreceptor gene expression is also reported in the dorsal raphe⁽¹⁴⁾ and would favour higher serotonin neurone firing rates.

Postmortem studies of depressed suicides report fewer 5-HT transporters in prefrontal cortex, hypothalamus, occipital cortex, and brainstem.⁽¹⁵⁾ Moreover, in suicides this deficit appears localized to the ventromedial prefrontal cortex, whereas depressed individuals who died of other causes had lower binding throughout the prefrontal cortex.⁽¹⁶⁾

The emerging picture from postmortem studies of greater 5-HT_2A receptor binding in the frontal cortex of depressed individuals who die by suicide, fewer brainstem 5-HT_1A autoreceptors, and fewer serotonin transporters in the cortex, as well as findings of greater tryptophan hydroxylase (the rate-limiting step in serotonin synthesis) immunoreactivity in serotonin nuclei in the brainstem⁽¹⁷⁾ all point to homeostatic changes designed to increase deficient serotonergic transmission evidenced by low 5-HIAA in CSF and brain, low 5-HT and 5-HIAA in brainstem, and blunted prolactin response to fenfluramine challenge.

Serotonergic dysfunction and suicide endophenotypes. Increased aggression has been associated with suicide and more highly lethal suicide attempts and impulsivity has shown a stronger relationship to non-fatal suicide attempts.⁽¹⁸⁾ Impulsive aggressive traits are potentially part of the diathesis for suicidal behaviour.⁽¹⁾ Reduced activity of the serotonin system has been implicated in impulsive violence and aggression in studies in a variety of paradigms including: Low CSF 5-HIAA in individuals with a lifetime history of aggressive behaviour with personality and other psychiatric disorders;^{(19, 20)} a blunted prolactin response to serotonin-releasing agent fenfluramine in personality disorder patients,^{(21, 22)} and; greater platelet 5-HT_2A binding correlated with aggressive behaviour in personality and other psychiatric disorder patients.^{(23, 24)} In a postmortem study of aggression, suicidal behaviour, and serotonergic function a positive relationship between lifetime history of aggression scores and 5-HT_2A binding in several regions of prefrontal cortex of individuals who had died by suicide was found.⁽²⁵⁾

Positron emission tomography (PET) studies have shown a deficient response to serotonergic challenge in the orbitofrontal cortex, medial frontal, and cingulate regions in individuals with impulsive aggression compared to controls^{(26, 27)} and lower serotonin transporter binding in the anterior cingulate cortex in impulsive aggressive individuals compared to healthy controls.⁽²⁸⁾ The prefrontal cortex is important in the inhibitory control of behaviour, including impulsive and aggressive behaviour.⁽²⁹⁾ Thus aggressive/ impulsive traits, related to serotonergic dysfunction, are potentially an aspect of the diathesis for suicidal behaviour, whereby aggressive/suicidal behaviours is manifested in response to stressful circumstances or powerful emotions. This tendency might be conceived of as a diminution in natural inhibitory circuits, or as a volatile cognitive decision style.

Within the stress-diathesis model of suicidal behaviour, it is the confluence of stressful events with the diathesis that is thought to precipitate a suicidal act. Thus, investigating the functioning of stress response systems in suicidal individuals is important for elucidating neurobiological concomitants of suicidal behaviour and identifying targets for preventative intervention. The noradrenergic system and the HPA axis are two key stress response systems.

The majority of norepinephrine neurones in the brain are located in the brainstem locus coeruleus. Postmortem studies of suicides have documented fewer noradrenergic neurones in the locus coeruleus.⁽³⁰⁾ There are also indications of cortical noradrenergic overactivity including lower alpha and high-affinity beta1-adrenergic receptor binding,⁽³¹⁾ and lower β-adrenoceptor density and alpha2-adrenergic binding in the prefrontal cortex in individuals who died by suicide.⁽²³⁾ There is some, but not unanimous, evidence from prospective studies of lower levels of CSF 3-methoxy-4-hydroxphenylglycol (MHPG), a metabolite of noradrenaline, in future suicides,⁽³³⁾ although not in those making non-fatal suicide attempts.⁽³⁴⁾

Fewer noradrenergic neurones observed in depressed suicides may indicate a lower functional reserve of the noradrenergic system, which if accompanied by an exaggerated stress response with greater release of noradrenaline may result in norepinephrine depletion leading to depression and hopelessness, both of which are contributory factors to suicidal behaviour.

Noradrenergic and HPA axis responses to stress in adulthood appears to be greater in those reporting an abusive experience in childhood.⁽³⁵⁾ Such individuals are potentially at greater risk in adulthood for major depression and suicidal behaviour. Childhood abuse may be associated with increased risk for depression and suicidal behaviour because of a dysfunctional stress response both via the noradrenergic system and the HPA axis, and secondary effects of norepinephrine depletion and elevated cortisol levels. There is interaction between the noradrenergic system and the stress response activity of the HPA axis with reciprocal neural connections between corticotropin-releasing hormone neurones in the hypothalamic paraventricular nucleus and noradrenergic neurones in human brainstem and the locus coeruleus.⁽³⁶⁾

The hypothalamic-pituitary-adrenal axis is a major stress response system. Major depression is associated with hyperactivity of the HPA axis,⁽³⁷⁾ and suicidal patients in diagnostically heterogeneous populations exhibit HPA axis abnormalities, most commonly failure to suppress cortisol normally after dexamethasone.⁽³³⁾ We found most future suicides were dexamethasone suppression test (DST) non-suppressors.⁽³³⁾ In mood disorders, DST non-suppressors had a 4.5-fold greater risk of dying by suicide compared with suppressors.⁽³⁾ Moreover, non-suppression may be characteristic of more serious attempts that result in greater medical damage^{(38, 39)} or the use of violent method in the suicide attempt.⁽⁴⁰⁾ In other indices of HPA axis function suicide attempters had attenuated plasma cortisol responses to fenfluramine although that may indicate less serotonin release and not an HPA abnormality,^{(41, 42)} and lower CSF corticotropin-releasing hormone (CRH) compared to non-attempters,⁽⁴³⁾ though not all studies agree.