The neurobiological underpinnings of suicidal behavior

When you consider that so much of our energy and such a large portion of our behavioral repertoire is devoted to ways of ensuring our survival, suicide appears to be perhaps the most inexplicable human behavior. What would make this human machine--which most of the time seems to be resolutely programmed to scratch, claw, and fight to endure through even the most dire situations--so easily decide to give it all up, even when the circumstances may not objectively seem all that desperate? Suicide is a difficult behavior to justify rationally, and yet it is shockingly common. More people throughout the world end their lives by suicide each year than are killed by homicide and wars combined.

The multitudinous influences that are thought to contribute to suicidal behavior are also very convoluted and difficult to untangle. Clearly, among different individuals the factors that lead to an act of suicide will vary considerably; nevertheless, there are some variables that are thought to generally increase the risk of suicidal behavior. A number of studies have, for example, demonstrated that genetic factors are associated with a predisposition to suicidal behavior. Also, early-life adversity--like sexual abuse, physical abuse, or severe neglect--has been strongly linked to suicide. However, even among groups with higher suicide risk there is a great deal of variability, which adds to the complexity of the issue. For example, personality traits like impulsiveness and aggression have been associated with an increased risk of suicide--but this relationship is seen primarily in younger people. It is not as apparent in older individuals who display suicidal behavior; they are often characterized by higher levels of harm avoidance instead of risk-taking.

While there are a number of predisposing factors involving personal characteristics or previous life events that make suicidal ideation and behavior more likely, there are also factors that immediately precede a suicide attempt which are thought to be directly linked to the transition from thinking about suicide to acting on those thoughts. Of course, some of those factors are likely to involve changes in neurobiology and neurochemistry that cause suicide--which may have previously just been an occasional thought--to become the focus of a present-moment plan that is sometimes borne out with great urgency and determination. And, while it is important to be able to identify influences that predispose individuals to suicidal thinking in general, an understanding of the neurobiological factors that precipitate a suicide attempt might open the door for treatments designed to protect an individual from acting on (or experiencing) sudden impulses to complete a suicide.

While the distally predisposing factors for suicidal behavior are difficult to identify due to the myriad influences involved, however, the proximal neurobiological influences are hard to pinpoint due both to their complexity and the fact that a suicidal crisis is often short-lived and difficult to study. The most direct way to investigate changes in the suicidal brain would be to look at brains of individuals who are suicide completers (i.e. those who are now deceased due to suicide). One reason for focusing on suicide completers is that we can expect some neurochemical--and possibly psychological--differences between suicide completers and those who attempted suicide but are still alive. However, working with postmortem brains has its own limitations: obtaining accurate background information may be challenging without the ability to interview the patient, there may be effects on the brain (e.g. from the process of death and its associated trauma or from drugs/medications taken before death) that may make it hard to isolate factors involved in provoking one towards suicide, and the limitation of only being able to examine the brain at one time makes causal interpretations difficult.

Regardless, investigations into irregularities in the brains of those who exhibit suicidal behavior (both attempters and completers) have identified several possible contributing factors that may influence the decision to act on suicidal thoughts. Many of these factors are also implicated in depressed states, as most suicidal individuals display some characteristics of a depressed mood even if they don't meet the criteria for a diagnosis of major depressive disorder. (This, of course, adds another layer of complexity to interpretation as it is difficult to determine if suicide-related factors are simply characteristics of a depressed mood and not solely related to suicidal actions.) The role of each of these factors in suicidal behavior is still hypothetical, and the relative contribution of each is unknown. However, it is thought that some--or all--of them may be implicated in bringing about the brain state associated with suicidal actions.

Alterations in neurotransmitter systems

Abnormalities in the serotonin system have long been linked to depressive behavior, despite more recent doubts about the central role of serotonin in the etiology of depression. Similarly, there appear to be some anomalies in the serotonin system in the brains of suicidal individuals. In an early study on alterations in the serotonin system in depressed patients, Asberg et al. found that patients with lows levels of 5-hydroxyindoleacetic acid, the primary metabolite of serotonin (and thus often used as a proxy measure of serotonin levels), were significantly more likely to attempt suicide. Additionally, those who survive a suicide attempt display a diminished response to the administration of fenfluramine, which is a serotonin agonist that in a typical brain prompts increased serotonin release. A number of neuroimaging studies have also detected reduced serotonin receptor availability in the brains of suicidal patients. This evidence all suggests that abnormalities in the serotonin system play some role in suicidal behavior, although the specifics of that role remain unknown.

As we have learned from investigations of depression, however, it is important to avoid focusing too much on one-neurotransmitter explanations of behavior. Accordingly, a number of other neurotransmitter abnormalities have been detected in suicidal patients as well. For example, gene expression analyses in postmortem brains of individuals who died by suicide have identified altered expression of genes encoding for GABA and glutamate receptors in various areas of the brain. Although the consequences of these variations in gene expression is unknown, abnormalities in GABA and glutamate signaling have both also been hypothesized to play a role in depression

Abnormalities in the stress response

Irregularities in the stress response have long been implicated in depression, and thus it may not be surprising that stress system anomalies have been observed in patients exhibiting suicidal behavior as well. The hypothalamic-pituitary-adrenal (HPA) axis is a network that connects the hypothalamus, pituitary gland, and adrenal glands; it is activated during stressful experiences. When the HPA axis is stimulated, corticotropin-releasing hormone is secreted from the hypothalamus, which causes the pituitary gland to secrete adrenocorticotropic hormone, which then prompts the adrenal glands to release the stress hormone cortisol. In depressed patients, cortisol levels are generally higher than normal, suggesting the HPA axis is hyperactive; this may be indicative of the patient being in a state of chronic stress.

In suicidal individuals, the HPA axis seems to be dysregulated as well. For example, in one study the HPA activity of a group of psychiatric inpatients was tested using what is known as the dexamethasone suppression test (DST). In this procedure, patients are injected with dexamethasone, a synthetic hormone that should act to suppress cortisol secretion if HPA axis activity is normal; if it does not do so, however, it suggests the HPA axis is hyperactive. Out of 78 patients, 32 displayed abnormal HPA activity on the DST. Over the next 15 years, 26.8% of the individuals with abnormal HPA activity committed suicide, while only 2.9% of the individuals with normal DST results killed themselves.

Another system involved in stress responses that may display irregularities in suicidal individuals is the polyamine stress response (PSR). Polyamines are molecules that are involved in a number of essential cellular functions; their potential role in psychiatric conditions has only been recognized in the past few decades. It is believed that the activation of the PSR and the associated increases in levels of polyamines in the brain may be beneficial, serving a protective role in reducing the impact of a stressor on the brain. And, there appear to be abnormalities in the PSR in the brains of those who have committed suicide. Because the PSR and its role in psychiatric conditions is still just beginning to be understood, however, it is unclear what these alterations in the PSR might mean; future investigations will attempt to elucidate the connection between PSR abnormalities and suicidal behavior.

One of the consequences of stress is the initiation of an inflammatory response. This is thought to be an adaptive reaction to stress, as the stress system likely evolved to deal primarily with physical trauma, and the body would have benefited from reflexive stimulation of the immune system in cases where physical damage had been sustained. This immune system activation would prepare the body to fight off infection that could occur due to potential tissue damage (the inflammatory response is the first step in preventing infection). Thus, it may not be surprising that suicidal patients often display markers of inflammation in the brain. This inflammatory reaction may on its own promote brain changes that increase suicide risk, or it may just be a corollary of the activation of the stress system.

Glial cell abnormalities

While we have a tendency to focus on irregularities in neurons and neuronal communication when investigating the causes of behavior, it is becoming more widely recognized that glial cells also play an essential role in healthy brain function. Accordingly, anomalies in glial cells have been noted in the brains of suicidal patients. Several studies, for example, have identified deficits in the structure or function of astrocytes in the suicidal brain. One study found that cortical astrocytes in post-mortem brains of suicide patients displayed altered morphology. Their enlarged cell bodies and other morphological abnormalities were consistent with the hypothesis that they had been affected by local inflammation. Analyses of gene expression in the postmortem brains of suicide victims also found that genes associated almost exclusively with astrocytes were differentially expressed. While the implications of these studies are not yet fully clear, abnormalities in glial cells represents another area of investigation in our attempts to understand what is happening in the suicidal brain.

Future directions

Irregularities in neurotransmitter systems, a hyperactive stress response, and anomalous glial cell morphology and density all may be factors that contribute to the suicidal phenotype. But it is unclear at this point if any one of these variables is the factor that determines the transition from suicidal ideation to suicidal behavior. It is more likely that they all may contribute to large-scale changes throughout the brain that lead to suicidal activity. Of course, all of the factors mentioned above may simply be associated with symptoms (like depressed mood) commonly seen in suicidal individuals, and the true culprit for provoking suicidal actions could be a different mechanism altogether, of which we are still unaware.

As mentioned above, this area of research is fraught with difficulties as the brains of suicide completers can only be studied postmortem. One research approach that attempts to circumvent this obstacle while still providing relevant information on the suicidal brain involves the study of pharmacological agents that reduce the risk of suicide. For, if a drug reduces the risk of suicide then perhaps it is reversing or diminishing the impact of neurobiological processes that trigger the event. One example of such a drug is lithium. Lithium is commonly used to treat bipolar disorder but is also recognized to reduce the risk of suicide in individuals who have a mood disorder. Gaining a better understanding of the mechanism of action that underlies this effect might allow for a better understanding of the neurobiology of suicidal behavior as well. Additionally, ketamine is a substance that appears to have fast-acting (within two hours after administration) antidepressant action and also may cause a rapid reduction (as soon as 40 minutes after administration) in suicidal thinking. Understanding how a drug can so quickly cause a shift away from suicidal thoughts may also be able to shed some light on processes that underlie suicidal actions.

Whatever the neurobiological underpinnings of suicidal behavior may be, the search for it should have some urgency about it. Suicide was the 10th leading cause of death in 2013, and yet it seems like a treatment for suicidal behavior specifically is not approached with the same fervor as treatments for other leading causes of death, like Parkinson's disease, that actually don't lead to as many deaths per year as suicide. Perhaps many consider suicide a fact of life, as something that will always afflict a subset of the population, or perhaps the focus is primarily directed toward treating depression with the assumption that better management of depression will lead to a reduction in suicide attempts. However, if we can come to understand what really happens in the brain of someone immediately before he makes the fatal decision to kill himself, treatment to specifically reduce the risk of suicide--regardless of the underlying disorder--is not out of the realm of possibility. Thus, it seems like a goal worth striving for.

Turecki, G. (2014). The molecular bases of the suicidal brain Nature Reviews Neuroscience, 15 (12), 802-816 DOI: 10.1038/nrn3839