The current issue of Acta Neuropsychiatrica presents a series of papers which together provide a broad overview relating stress, immunity, cytokine activity and depressive illness, as well as the influence of cytokines on other neurological disorders. This introduction to the issue presents a broad perspective of the impact of stressors on immune functioning in animal studies and in humans, considering the potential effects of acute, subchronic and chronic stressors, as well as the contribution of previous stressor experience in promoting neurochemical and immunological alterations. Given the supposition that cytokines may act as immunotransmitters, and immune challenge may be viewed as a stressor, a brief review is provided concerning the impact of stressors and cytokine challenges on central neurochemical functioning, with particular emphasis on the commonalties between the effects of these treatments. It is suggested that by virtue of the neurochemical changes imparted by cytokines, a depressive affect may be instigated, just as it is in response to psychogenic stressors. To this end, an overview is presented concerning the relationship between cytokines and depression, as well as the influence of cytokine treatments on behavioral changes in animal studies and among patients receiving immunotherapy. Provisionally, the data support the view that activation of the inflammatory response system may contribute to affective illness, and that cytokines may act as signaling molecules to activate central nervous system processes regulating mood states.

A fundamental question in stress research is when the glucocorticoid stress hormone (cortisol in man) stops being neuroprotective and becomes harmful to the brain with negative consequences for cognition and mood. To address this question Section 1 focuses on the action mechanism of glucocorticoids. These hormones act via high and low affinity nuclear receptors, which regulate gene transcription in a coordinate manner. The receptors are expressed abundantly in hippocampus, amygdala and frontal cortex involved in cognitive processes. In Section 2 hypercortisolism is considered a potential disease factor for about 50% of the patients suffering from major depression. Recent data show that these patients recover within a few days when excess cortisol action is blocked with high doses of an antiglucocorticoid. Section 3 concerns animal models with ‘depression-like’ features of hypercorticism generated by manipulation of gene X environment inputs. Using gene expression profiling technology in the hippocampal transcriptome of these animals we identified about 700 potential targets for antidepressants out of 30 000 detectable gene products. One of our models is based on early life programming of the stress system. Rats exposed as pups to maternal deprivation display at senescence an enhanced individual difference in cognitive performance. The maternally deprived senescent animals age either successfully or become senile, at the expense of the average performance of non-deprived controls. The essay is concluded with the notion that the new generation of antidepressants ameliorates specific psychic dysfunctions (e.g. cognitive performance) linked to aberrant stress hormone action in discrete brain regions.

This paper summarizes the possible interrelation between peripheral and/or cerebral inflammation and depression. Often, depression is regarded as a consequence of life events, including disabling diseases. The question addressed here is whether activation of the inflammatory response system (IRS) can cause depression. Epidemiological studies suggest that depression can be precipitated by bacterial or viral infections. In depressed patients, peripheral markers of the IRS are often increased. There is some evidence that some forms of depression are caused by a viral infection of the limbic system. More consistent are the observations that depression in diseases with active cerebral inflammatory processes (e.g. multiple sclerosis, Alzheimer's disease) may concur. Direct evidence of a relation between depression and inflammation was found in post-mortem brain material of patients with a vascular depression. In both inflammatory brain diseases and in depression, a state-dependent increased hypothalamus–pituitary–adrenal axis activity is seen. Animals studies have shown that intact cerebral serotonin systems are required for the activation of the IRS following an endotoxin challenge and that long-term treatment with antidepressants may change such a response. Gender differences between the prevalence of depression and inflammatory diseases are similar, as more females are affected. We hypothesize that cerebral or peripheral activation of the IRS may contribute to the course of some antidepressant treatment-resistant depressions. Clinical trials combining antidepressants and drugs that reduce the activation of the IRS may provide evidence for such proposed depression subtypes.

This paper examines the relationship of recent life events to specific kinds of depression using published studies and the author's own work. An overall effect of life events on depression has been found consistently and is moderate in degree. In suicide attempts there are stronger and more immediate effects than in depression. Life events precede both non-melancholic and melancholic depressions. It is only in recurrent depressions that life events are less common with melancholic pictures. Life events influence bipolar disorder as well as unipolar. Mania may be preceded by life events, particularly those involving social rhythm disruption, but it is harder to rule out events which are consequences of insidious development of illness. There are strong effects of life events and social support in postpartum depressions but in postpartum psychoses these effects are absent. Events precede depression comorbid with other disorders as well as pure depression. The course of depression is also influenced by life stress with less remission where negative events occur after onset and better outcome where earlier adverse events are neutralized. Relapse is related to immediately preceding life events. However, where depressions are both severe and recurrent life stress effects weaken and as the number of episodes increases preceding life events lessen. These findings suggest that some kinds of depression are more related to psychosocial causation and some are more biological in origin.

Borderline personality disorder can be characterized in terms of a profile of abnormal deviations on multiple personality dimensions using the temperament and character inventory (TCI). Borderline patients show poor character development, including low TCI self-directedness (irresponsible, blaming) and low TCI cooperativeness (hostile, intolerant). Their temperament is explosive or unstable due to a combination of high TCI harm avoidance (anxious, shy), high TCI novelty seeking (impulsive, quick-tempered), and low reward dependence (cold, aloof). Consequently they are usually dysthymic with an admixture of anxiety and anger, and regulate their social problems and intense emotions in immature ways. Genetic and psychobiological studies have led to identification of biological correlates of each of the TCI dimensions of personality, including individual differences in regional brain activity, psychophysiological variables, neuroendocrine abnormalities and specific gene polymorphisms. Each dimension of personality involves complex non-linear interaction of multiple genetic and environmental factors and, in turn, each personality dimension interacts with the others in influencing the way an individual directs and adapts to his or her life experiences. Systematic clinical trials have shown that these personality variables predict the response to pharmacological and psychotherapeutic treatments. For example, high harm avoidance and low self-directedness predict slower response and more rapid relapse with both antidepressants and cognitive-behavioral therapy. Treatment with drugs and/or psychotherapy can be individually matched to the patient's profile of temperament and character traits, rather than treating a heterogeneous group of patients as if they had a discrete, homogeneous illness. Fundamental change in cognitive schemas depends on attention to all aspects of character, especially self-transcendence, which has previously been neglected in cognitive-behavioral therapy. Personality integration requires non-resistance to our natural intuitive awareness, rather than intensified intellectual and emotional defenses.

Cytokines (e.g. various interleukins and subfamily members, tumor necrosis factors, interferons, chemokines and growth factors) act in the brain as immunoregulators and neuromodulators. Over a decade ago, the integrative article ‘Immunoregulators in the Nervous System’ (Neurosci Biobehav Rev 1991; 15: 185–215) provided a comprehensive framework of pivotal issues on cytokines and the nervous system that recently have been extensively studied. Cytokine profiles in the brain, including cytokine generation and action, have been studied in multiple models associated with neuropathophysiological conditions. These include: (1) acute conditions and disorders such as stroke (cerebral ischemia or infarction and intracranial hemorrhage), traumatic brain injury, spinal cord injury and acute neuropathies; (2) chronic neurodegenerative disorders and chronic conditions, including Alzheimer's disease, Parkinson's disease, neuropathic pain, epilepsy and chronic neuropathies; (3) brain infections, including bacterial meningitis and encephalitis; (4) brain tumors; (5) neuroimmunological disorders per se, such as multiple sclerosis; (5) psychiatric disorders, including schizophrenia and depression; (6) neurological and neuropsychiatric manifestations associated with non- central nervous system (CNS) disorders such as peripheral cancer, liver, kidney and metabolic compromise, and peripheral infectious and inflammatory conditions; and (7) cytokine immunotherapy, which can be accompanied by neuropsychiatric manifestations when administered either via peripheral or brain routes. Cytokine profiles have also been studied in multiple animal models challenged with inflammatory, infectious, chemical, malignant and stressor insults. Essentially data show that cytokines play a pivotal role in multiple neuropathophysiological processes associated with different types of disorders and insults. Cytokine expression and action in the brain shows a different profile across conditions, but some similarities exist. Under a defined temporal sequence, cytokine involvement in neuroprotection or the induction of a deleterious pathophysiological cascade and in resolution/healing is proposed depending on the type of cytokine. In the brain, functional interactions among cytokines, balance between pro-inflammatory and anti-inflammatory cytokines and functional interactions with neurotransmitters and neuropeptides play a pivotal role in the overall cytokine profile, pattern of neuropathophysiological cascades, and quality and magnitude of neuropsychiatric manifestations. In this brief review various selected cytokine-related issues with relevance to the brain are discussed.

The present paper reviews recent studies on the effects of stress on immune function in laboratory animals. The emphasis is on those studies where a simultaneous comparison of acute and chronic stress regimens was determined, although additional relevant studies are also reviewed. The effects of stress on basic measurements of cellular and humoral immune measures are discussed, including the growing number of studies that have reported alterations in macrophage functions. The latter are key elements in the innate immune response, and like measurements of T cell function and antibody production, are inhibited and enhanced by stressor exposure. This review does not focus on the mechanisms by which stress alters immune function, there being little to add conceptually in terms of what was reported previously (see Kusnecov AW, Rabin BS, Int Arch Allergy Immunol 1994;105:107–121.). However, a question is raised in the conclusion as to how stressor effects on immune function should be interpreted, for it is clear that immunological processes in and of themselves elicit central nervous system responses that neurochemically and endocrinologically do not differ from those produced in response to psychological stressors. Therefore, at least in the short term stressor-induced immune changes may not necessarily reflect maladaptive adjustments, although, as demonstrated by some studies reviewed in this paper, they may pose a serious risk to health should stressor exposure be persistent and uncontrolled.

Experimental and clinical evidence implicates stress as a major predisposing factor in depression and other severe psychiatric disorders. In this review, evidence is presented to show how the impact of stress on the central sympathetic system leads to changes in the endocrine, immune and neurotransmitter axes which underlie the main clinical symptoms of depression. Thus it can be shown that the noradrenergic system is dysfunctional in depression, a situation which reflects the chronic hypersecretion of glucocorticoids and inflammatory mediators within the brain in addition to an enhanced activity of the locus ceruleus. With regard to the actions of antidepressants in modulating the stress response and alleviating depression it is now evident that, irrespective of the presumed specificity of the antidepressants for the noradrenergic or serotonergic systems, they all normalize noradrenergic function. This action is due partly to the regulation of tyrosine hydroxylase activity in the locus ceruleus but also enhances neuronal sprouting which counteracts the neurodegenerative effects of chronic stress.

Seasonal asymmetry in yearly suicide occurrence is a long-observed phenomenon in psychiatric, suicidological and sociological research, and the effects of seasonal factors on suicidal behaviour have been the focus of a number of earlier studies. Taking into account limitations of data and methods, these studies have in general favoured interpretations based on psychosocial factors. Recent studies have challenged the widely held notion that seasonal effects on suicide are not influenced by age, gender or the circumstances of the act. The suicides committed with violent methods have been shown to follow clearer seasonal patterns than suicides by less violent methods, and differences have been found between male and female cycles of occurrence. The seasonal occurrence of suicides has also been found to differ significantly between the young and the elderly. The use of inappropriate statistics or age- and gender-biased samples may have hidden a seasonal component in some previous studies on attempted suicide. The absence of seasonality in earlier studies on attempted suicide was interpreted as depending upon the minor relevance of psychiatric and biological factors in non-fatal self-harm. However, recent studies have reported clear seasonality in attempted suicide samples, with older people showing greater seasonal effect. Recent literature after 1985 on seasonal variation and weather or climate influence in attempted and completed suicide is reviewed. Suggestions for research and the development of more effective preventative strategies are offered.

Serotonin is one of the neurotransmitters implicated in normal personality. Many psychobiological models of personality include some dimensions related to serotonin. For instance, the harm avoidance dimension of the biosocial model developed by Cloninger is related to serotonergic activity. Higher scores on the harm avoidance dimension should theoretically reflect increased serotonergic activity. However, correlation studies related serotonin activity to harm avoidance dimension have not yielded consistent findings. These controversial results are probably related to the complexity of the neurotransmitter systems, and the different assessment techniques used in these studies. Finally, recent genetic studies have examined the association between personality dimensions and serotonergic receptor polymorphisms with mixed results. Serotonin is not only related to some dimensions of normal personality. Several psychopathological disorders are associated with serotonergic dysfunction. More particularly, borderline personality disorder (BPD) can be defined by many of the symptoms associated with serotonergic dysregulation, including affective lability, suicidal behaviours, impulsivity and loss of impulse control. Indeed, several reports have demonstrated the efficacy of selective serotonin re-uptake drugs in treating the depressive and impulsive symptoms of patients with BPD. Moreover, some challenge studies have reported a lower serotonergic activity in BPD. Because these challenges are not specific, we have assessed the serotonergic activity in BPD with the flesinoxan challenge. Preliminary results showed that the prolactine responses to flesinoxan were significantly lower in BPD patients compared to those observed in controls.

Early life environmental factors have been associated with altered predisposition to a variety of pathologies. A considerable literature examines pre- and postnatal factors associated with increased risk of cardiovascular, metabolic (i.e. insulin resistance, hyperlipidemia) and psychiatric disease, and the importance of hormonal programming. The brain is exquisitely sensitive to environmental inputs during development and the stress responsiveness of the hypothalamic–pituitary–adrenal (HPA) axis has been shown to be both up- and down-regulated by early life exposure to limited nutrition, stress, altered maternal behaviors, synthetic steroids and inflammation. It has been suggested that peri-natal programming of HPA axis regulation might therefore contribute to metabolic and psychiatric disease etiology. In addition, glucocorticoids play modulatory roles regulating many aspects of immune function, notably controlling both acute and chronic inflammatory responses. Neuroendocrine–immune communication is bidirectional, and therefore it is expected that environmental factors altering HPA regulation have implications for stress effects on immune function and predisposition to inflammation. The impact of pre- and postnatal factors altering immune function, stress responsivity and predisposition to inflammatory disease are reviewed. It is also examined whether the early ‘immune environment’ might similarly influence predisposition to disease and alter neuroendocrine function. Evidence indicating a role for early life inflammation and infection as an important factor programming the neuroendocrine–immune axis and altering predisposition to disease is considered.