Altered metabolic and neurochemical responses to chronic unpredictable stressors in ghrelin receptor-deficient mice.

Ghrelin, a hormone produced by the stomach, is generally associated with feeding responses and the regulation of food intake. Recent evidence, however, suggests that ghrelin is also a stress hormone, given that it is released following acute and chronic stressors. The present study examined the role of ghrelin in producing normal metabolic and neurochemical responses to chronic stress. This was achieved by examining these responses in mice with targeted deletions of the ghrelin receptor gene (GHSR KO mice), and comparing them with the same responses in their wild-type (WT) littermates. As expected, WT stressed mice decreased their caloric intake, body weight gain and caloric efficiency while maintaining adiposity. GHSR KO mice, however, did not show these alterations despite having normal glucocorticoid responses to stress. In parallel with these changes, chronic unpredictable stress caused changes in norepinephrine, dopamine and serotonin in a number of brain regions. Of these, norepinephrine neurotransmission in the arcuate nucleus and prefrontal cortex was differentially altered in GHSR KO mice. Within the nucleus acumbens, dopamine utilization was increased in WT mice but not in GHSR KO mice. Finally, there were strain differences in serotonin neurotransmission that may explain interstrain body weight and adiposity differences. These results suggest that the metabolic changes necessary to deal with the energetic challenge presented by repeated exposure to stressors do not occur in GHSR KO mice, and they are discussed within the context of the potential vulnerability to stress-induced pathology.

Stress and coping factors influence tumor growth.

Growth of syngeneic P815 mastocytoma in DBA/2J male mice was evaluated as a result of various stress regimens. A single session of inescapable shock resulted in earlier tumor appearance, exaggeration of tumor size, and decreased survival time in recipient animals. Escapable shock had no such effects. The effects of the inescapable shock were mitigated if mice received long-term shock treatment.

Suicidal thoughts and attempts in First Nations communities: links to parental Indian residential school attendance across development.

The Indian residential school (IRS) system in Canada ran for over a century until the last school closed in 1996. Conditions in the IRSs resulted in generations of Indigenous children being exposed to chronic childhood adversity. The current investigation used data from the 2008-2010 First Nations Regional Health Survey to explore whether parental IRS attendance was associated with suicidal thoughts and attempts in childhood, adolescence and in adulthood among a representative sample of First Nations peoples living on-reserve across Canada. Analyses of the adult sample in Study 1 (unweighted n=7716; weighted n=186,830) revealed that having a parent who attended IRS was linked with increased risk for suicidal thoughts and attempts in adolescence and adulthood. Although females were negatively affected by having a parent who attended IRS, the link with suicidal ideation in adulthood was greater for males. Analyses of the youth sample in Study 2 (unweighted n=2883; weighted n=30,190) confirmed that parental IRS attendance was associated with an increased risk for suicidal ideation and attempts. In contrast to the adult sample, parental IRS attendance had a significantly greater relation with suicidal ideation among female youth. A significant interaction also emerged between parental IRS attendance and age in the youth sample, with the influence of parental attendance being particularly strong among youth ages 12-14, compared with those 15-17 years. These results underscore the need for culturally relevant early interventions for the large proportions of Indigenous children and youth intergenerationally affected by IRSs and other collective traumas.

Nesfatin-1 increases anxiety- and fear-related behaviors in the rat.

RATIONALE: Nesfatin-1, derived from the protein NEFA/nucleobindin2 (NUCB2), is a newly identified peptide that acts as a potent satiety agent. It has been reported that peptides involved in the regulation of ingestive behavior are also involved in the regulation of the stress response. However, the relation between nesfatin-1 and stressor-related behaviors like anxiety and/or fear has not yet been investigated. OBJECTIVE: The effects of intracerebroventricular (ICV) injection of nesfatin-1 (0, 5, and 25 pmol/3 microl) were assessed in several paradigms that are thought to reflect anxiety and/or fear in rats. RESULTS: Consistent with an anxiogenic effect, nesfatin-1 dose-dependently decreased the percentage of time spent on the open arms of the elevated plus maze, increased latency to approach, and decreased consumption of a palatable snack in an anxiogenic (unfamiliar) environment. Moreover, ICV nesfatin-1 increased the fear-potentiated startle response and the time spent freezing to both context and conditioned cues in a conditioned emotional response test. CONCLUSIONS: These findings suggest that in addition to its role as a satiety peptide, nesfatin-1 may also be involved in the mediation of anxiety- and/or fear-related responses.

Anxiety responses, plasma corticosterone and central monoamine variations elicited by stressors in reactive and nonreactive mice and their reciprocal F1 hybrids.

Stressor-provoked anxiety, plasma corticosterone, and variations of brain monoamine turnover are influenced by genetic factors, but may also be moderated by early life experiences. To evaluate the contribution of maternal influences, behavioral and neurochemical stress responses were assessed in strains of mice that were either stressor-reactive or -resilient (BALB/cByJ and C57BL/6ByJ, respectively) as well as in their reciprocal F(1) hybrids. BALB/cByJ mice demonstrated poorer maternal behaviors than did C57BL/6ByJ dams, irrespective of the pups being raised (inbred or F(1) hybrids). The BALB/cByJ mice appeared more anxious than C57BL/6ByJ mice, exhibiting greater reluctance to step-down from a platform and a greater startle response. Although the F(1) behavior generally resembled that of the C57BL/6ByJ parent strain, in the step-down test the influence of maternal factors were initially evident among the F(1) mice (particularly males) with a BALB/cByJ dam. However, over trials the C57BL/6ByJ-like behavior came to predominate. BALB/cByJ mice also exhibited greater plasma corticosterone elevations, 5-HT utilization in the central amygdala (CeA), and greater NE turnover in the paraventricular nucleus of the hypothalamus (PVN). Interestingly, among the F(1)'s corticosterone and 5-HIAA in the CeA resembled that of the BALB/cByJ parent strain, whereas MHPG accumulation in the PVN was more like that of C57BL/6ByJ mice. It seems that, to some extent, maternal factors influenced anxiety responses in the hybrids, but did not influence the corticosterone or the monoamine variations. The inheritance profiles suggest that anxiety was unrelated to either the corticosterone or monoamine changes.

Multiple neurochemical and behavioral consequences of stressors: implications for depression.

Animal models of clinical depression have frequently focused on the contribution of stressors to the induction of behavioral impairments and pharmacological intervention in the amelioration of these disturbances. Stressors provoke various behavioral disturbances and influence the activity of central neurotransmitters implicated in depression. It is our contention that those variables which favor the provocation of amine depletions or prevent the development of a neurochemical adaptation will increase vulnerability to behavioral disturbances. It is essential to consider, however, that marked interindividual and interstrain differences exist in the behavioral and neurochemical response to stressors, and in the effectiveness of antidepressant treatments.

The pathogenesis of clinical depression: stressor- and cytokine-induced alterations of neuroplasticity.

Stressful events promote neurochemical changes that may be involved in the provocation of depressive disorder. In addition to neuroendocrine substrates (e.g. corticotropin releasing hormone, and corticoids) and central neurotransmitters (serotonin and GABA), alterations of neuronal plasticity or even neuronal survival may play a role in depression. Indeed, depression and chronic stressor exposure typically reduce levels of growth factors, including brain-derived neurotrophic factor and anti-apoptotic factors (e.g. bcl-2), as well as impair processes of neuronal branching and neurogenesis. Although such effects may result from elevated corticoids, they may also stem from activation of the inflammatory immune system, particularly the immune signaling cytokines. In fact, several proinflammatory cytokines, such as interleukin-1, tumor necrosis factor-alpha and interferon-gamma, influence neuronal functioning through processes involving apoptosis, excitotoxicity, oxidative stress and metabolic derangement. Support for the involvement of cytokines in depression comes from studies showing their elevation in severe depressive illness and following stressor exposure, and that cytokine immunotherapy (e.g. interferon-alpha) elicited depressive symptoms that were amenable to antidepressant treatment. It is suggested that stressors and cytokines share a common ability to impair neuronal plasticity and at the same time altering neurotransmission, ultimately contributing to depression. Thus, depressive illness may be considered a disorder of neuroplasticity as well as one of neurochemical imbalances, and cytokines may act as mediators of both aspects of this illness.

Amphetamine psychosis and schizophrenia: a dual model.

Several of the behavioral consequences of acute and chronic amphetamine treatment were evaluated and related to the underlying neurochemical correlates of drug treatment. It was suggested that decreased noradrenergic activity after long-term amphetamine treatment influences stimulus sampling, whereas enhanced dopaminergic activity was responsible for the progressive augmentation of stereotypy and self-stimulation behavior observed after long-term exposure to amphetamine. It was hypothesized that amphetamine-induced psychosis and the symptomatology associated with schizophrenia are related to alterations in both norepinephrine and dopamine activity.

Stressor-induced anhedonia in the mesocorticolimbic system.

It has been suggested that uncontrollable stressors induce motivational changes in animals which are reminiscent of reward alteration in human depression. Although there is considerable support for this position, most animal models of depression do not adequately address this issue. The present review suggests that stressor-induced reductions in the rewarding value of electrical brain stimulation (ICSS) from the mesocorticolimbic system may simulate the anhedonia of human depression. The magnitude, severity and the site of these stressor-induced reward alterations within the mesocorticolimbic system vary with the strain of animal employed. The anhedonic effects of stressors are attenuated by treatments which influence mesocorticolimbic DA turnover, including systemic antidepressant and intraventricular neuropeptide administration. Although the diverse symptom profile of depression should be addressed by consideration of the constellation of behavioral disturbances induced by stressors, considerable emphasis should be devoted to an assessment of reward loss in depression. The implications of these data to the stressor depression topography and the potential role of mesocorticolimbic DA in depression and anhedonia are discussed.

Neural plasticity, neuropeptides and anxiety in animals--implications for understanding and treating affective disorder following traumatic stress in humans.

Exposure of rats to cats (predator stress) lastingly increases rodent anxiety-like behavior (ALB) in the elevated plus-maze. Previous work shows that lasting changes in ALB following predator stress depend on NMDA and CCKB receptors. In this paper we describe the effects of differing degrees of predator exposure on behavior. Effects depend on the behavioral measure. In general, exposure to predator odor is less provocative of lasting change in ALB than is unprotected exposure to a cat. In addition, we examine the development of effects of unprotected predator exposure over time. Lasting effects on ALB begin at 30 min to 1 h after predator stress and persist for at least 3 weeks. We also report a complex pattern of effects of predator stress on neuroendocrine and stress peptide (bombesin, CRF and AVP) levels in a variety of brain areas. Not surprisingly, predator exposure increases plasma levels of corticosterone and ACTH. Central changes in peptide content in the hypothalamo-pituitary axis, related hypothalamic nuclei, limbic and brain stem areas are also noted. Finally, path analysis demonstrates a replicable relationship between cat behavior, rat defensive behavior and degree of increase in ALB one week later. It is proposed that behavioral changes following predator stress may model anxiety associated with PTSD.

Interleukin-1 beta production in dysthymia before and after pharmacotherapy.

BACKGROUND: Like major depression, dysthymia has been associated with elevated production of interleukin-1 (IL-1 beta) in mitogen-stimulated lymphocytes. In the present investigation, we assessed whether the elevated IL-1 beta production in dysthymic patients would normalize following treatment with sertraline. METHODS: The production of IL-1 beta was determined in dysthymic patients and in nondepressed control subjects. Patients then received 12 weeks of doses of either sertraline or placebo in a double-blind trial, after which cytokine production was again determined. RESULTS: Basal IL-1 beta was elevated in dysthymic patients relative to control subjects. Cytokine production was modestly correlated with the severity of symptoms and with the age of illness onset. Relative to placebo treatment, sertraline attenuated the symptoms of depression; however, this was not accompanied by normalization of IL-1 beta production. CONCLUSIONS: While dysthymia is associated with elevated IL-1 beta production, the failure for the cytokine to normalize following symptom alleviation suggests that either the IL-1 beta may be a trait marker of the illness, or that more sustained treatment is necessary to reduce cytokine production. Given the neuroendocrine and central neurochemical consequences of exogenously administered IL-1 beta, the possibility ought to be explored that increased IL-1 beta production may play a role in the pathophysiology of dysthymia.

Posttraumatic stress symptoms and salivary cortisol levels.

OBJECTIVE: This study assessed the relationship between posttraumatic stress symptoms and salivary cortisol levels after a severe ice storm. METHOD: Posttraumatic stress symptoms (Impact of Event Scale scores) and salivary cortisol levels were determined in 115 victims of an ice storm and in 27 healthy comparison subjects 1 month and approximately 1 year after the ice storm. RESULTS: One month after the storm, Impact of Event Scale scores for the victims (mean=20.31, SD=15.23) exceeded those of the comparison subjects (mean=5.30, SD=9.78) but were reduced approximately 1 year later (mean=14.01, SD=13.68). A quadratic relation was found to exist between Impact of Event Scale scores and cortisol levels. CONCLUSIONS: One month after the storm, cortisol levels were found to be elevated among the victims but were diminished among those with the highest Impact of Event Scale scores. This relationship was found not to exist approximately 1 year later.

Treatment of primary dysthymia with group cognitive therapy and pharmacotherapy: clinical symptoms and functional impairments.

OBJECTIVE: This study assessed the efficacy of antidepressant treatment (sertraline) and group cognitive behavior therapy, alone or in combination, in primary dysthymia. The clinical features of dysthymia, as well as the functional impairments associated with the illness (e.g., quality of life, stress perception, coping styles), were evaluated. METHOD: Patients (N = 97) diagnosed with primary dysthymia, but no other current comorbid disorder, received either sertraline or placebo in a double-blind design over 12 weeks. In addition, a subgroup of the patients (N = 49) received a structured, weekly group cognitive behavior therapy intervention. RESULTS: Treatment with sertraline, with or without group cognitive behavior therapy, reduced the functional impairment of depression. The reductions were similar in the drug-cognitive therapy group and in subjects who received the drug alone. Furthermore, while group cognitive behavior therapy alone reduced the depression scores, this effect was not significantly greater than the effect of the placebo. The drug treatment also induced pronounced improvement in the functional measures, and in some respects these effects were augmented by group cognitive behavior therapy. Among patients who responded favorably to cognitive behavior therapy, the improvements in the functional measures were similar to those who responded to drug treatment, whereas such functional changes were not seen among patients who responded to placebo. CONCLUSIONS: Sertraline treatment effectively reduces the clinical symptoms and functional impairments associated with dysthymia. Although the group cognitive behavior therapy intervention was less effective in alleviating clinical symptoms, it augmented the effects of sertraline with respect to some functional changes, and in a subgroup of patients it attenuated the functional impairments characteristic of dysthymia.

The impact of stressors on immune and central neurotransmitter activity: bidirectional communication.

Antigenic challenge may have broad ranging effects which include not only immunological changes, but also endocrine and central neurotransmitter repercussions, and may thus elicit profound behavioral sequelae. Commensurate with the notion that bidirectional communication exists between the immune and central nervous systems it has been demonstrated that manipulations which influence central neurotransmitter or endocrine activity provoke alterations of immune functioning, and conversely immunological alterations will affect central neurotransmitter and endocrine activity. It seems, as well, that environmental stressors may provoke marked alterations of the activity of each of these systems. Indeed, in several respects the variables that influence vulnerability to stressor-provoked neurotransmitter changes, likewise affect the immunological alterations engendered by stressors. Moreover, immunological challenges will affect central neurotransmitter functioning in much the same way as stressors provoke such effects. It is thought that immune derived products (including cytokines as well as peptide hormones) may act directly or indirectly to moderate neurotransmitter functioning, and centrally derived neurotransmitters and hormones may affect receptors present on lymphocytes. In accordance with earlier suggestions, it is maintained that the immune system may be acting as a sensory organ informing the brain of the presence of antigenic challenges, and the brain may interpret such challenge as a stressor, hence leading to behavioral alterations.

Synergistic effects of interleukin-1beta, interleukin-6, and tumor necrosis factor-alpha: central monoamine, corticosterone, and behavioral variations.

The proinflammatory cytokines interleukin-1beta (IL-1beta), IL-6, and tumor necrosis factor-alpha (TNF-alpha) influence neuroendocrine activity, promote central neurotransmitter alterations, and induce a constellation of symptoms collectively referred to as sickness behaviors. These cytokines may also elicit anxiety and anhedonia, and have been associated with psychological disturbances in humans. In the present investigation, systemic IL-1beta and TNF-alpha dose-dependently and synergistically disrupted consumption of a highly palatable food source (chocolate milk), possibly reflecting anorexia or anhedonia engendered by the treatments. As well, these cytokines synergistically increased plasma corticosterone levels. Although IL-1beta and TNF-alpha provoked variations of amine turnover in the hypothalamus, locus coeruleus, and central amygdala, synergistic effects were not evident in this respect. Nevertheless, in view of the central amine variations induced by the cytokines, it is suggested that immune activation may come to influence complex behavioral processes, as well as affective state.

Murine tumor necrosis factor-alpha sensitizes plasma corticosterone activity and the manifestation of shock: modulation by histamine.

Murine tumor necrosis factor-alpha (mTNF-alpha) results in the sensitization of mechanisms underlying plasma corticosterone activity and sickness behavior, the latter being reminiscent of septic or anaphylactic shock. The mTNF-alpha induced a sensitization of sickness and corticosterone in mice that was attenuated by pretreatment with the combinations of histamine H(1) (diphenhydramine, mepyramine) and H(2) (cimetidine) antagonists. Likewise, coadministration of diphenhydramine and cimetidine prevented the mTNF-alpha-provoked rise of monoamine activity within the posterior hypothalamus. Although dexamethasone ameliorated the mTNF-alpha-induced sensitization of corticosterone, illness behavior was unaffected. It is suggested that mTNF-alpha-induced illness and the neuroendocrine sensitization are mediated by endogenous histamine.

Variations of nucleus accumbens dopamine and serotonin following systemic interleukin-1, interleukin-2 or interleukin-6 treatment.

The effects of systemically administered interleukin-1beta (1.0 microg), interleukin-6 (1.0 microg) and interleukin-2 (1.0 microg) on in vivo variations of monoamines were assessed in the nucleus accumbens. Administration of interleukin-1beta did not affect extracellular accumbal dopamine, provoked a modest rise of homovanillic acid, and prevented the decline of dihydroxyphenylacetic acid ordinarily seen in saline treated rats. Also, interleukin-1 provoked a modest increase of extracellular 5-hydroxyindoleacetic acid from the nucleus accumbens. Following exposure to the stress of a series of air-puffs, a still greater increase of accumbal 5-hydroxyindoleacetic acid was evident. In contrast to interleukin-1, systemic administration of interleukin-6 and interleukin-2 both induced marked reductions of interstitial dopamine levels. The air-puff exposure further enhanced these effects in rats that had received the cytokine treatment. As well, interleukin-6 and interleukin-2 were both found to reduce the homovanillic acid response associated with the stress, and interleukin-2 promoted a decline of homovanillic acid levels. Treatment with interleukin-6, like that of interleukin-1, prevented the decline of dihydroxyphenylacetic acid ordinarily observed over time, while interleukin-2 was without effect in this respect. Finally, interleukin-6 provoked a modest rise of 5-hydroxyindoleacetic acid, which was most apparent following air-puff exposure, while administration of interleukin-2 did not affect accumbal 5-hydroxyindoleacetic acid. It is suggested that the cytokines may influence the release of biogenic amines in the nucleus accumbens, but the profile of changes were cytokine-specific. As well, it appeared that the cytokines, particularly interleukin-1 and interleukin-6, may act synergistically with the stressor in promoting the amine variations. Systemic administration of cytokines clearly influenced monoamine activity at the nucleus accumbens, a region associated with both rewarding and aversive events. Thus, it may be expected that cytokine treatments may affect behavior. Moreover, it seems that the effects of interleukin-1 and interleukin-6 may be influenced by the presence of stressful stimuli. It ought to be underscored that although cytokines share features with the effects of stressors, most notably the variations of hypothalamic-pituitary-adrenal hormones, the pattern of central neurochemical changes elicited by the cytokines could be distinguished from the amine variations ordinarily associated with stressors.

Time-dependent sensitization of corticotropin-releasing hormone, arginine vasopressin and c-fos immunoreactivity within the mouse brain in response to tumor necrosis factor-alpha.

Stressor or cytokine treatments, such as interleukin-1beta, promote time-dependent alterations of hypothalamic-pituitary-adrenal functioning, including increased arginine vasopressin stores within corticotropin-releasing hormone (CRH) terminals in the external zone of the median eminence. Likewise, we have previously shown that the proinflammatory cytokine, tumor necrosis factor-alpha (TNF-alpha), provoked a time-dependent sensitization of neuroendocrine and brain monoamine activity. To further explore the protracted consequences of TNF-alpha, the present investigation determined whether the cytokine sensitized activity of neuroendocrine regulatory brain regions, as assessed by c-fos expression, and had protracted consequences on amygdaloid CRH, as well as hypothalamic corticotropin secretagogues. Indeed, immunoreactivity for arginine vasopressin and corticotropin-releasing hormone, and their colocalization within cell terminals of the median eminence, varied over time following an initial 4.0-microg tumor necrosis factor-alpha treatment, peaking after 7 days and normalizing within 28 days. Within the central amygdala, a sensitization effect was evident as reflected by increased CRH immunoreactivity, but this effect required re-exposure to the cytokine, unlike the median eminence changes that simply evolved with the passage of time. As well, tumor necrosis factor-alpha provoked a marked sensitization of c-fos staining within the paraventricular nucleus of the hypothalamus, supraoptic nucleus and the central amygdala. From these data we suggest that tumor necrosis factor-alpha influences responsivity of stressor-reactive brain regions and has protracted effects on central neuropeptide expression within the hypothalamus and central amygdala, although the time course for the effects vary across brain regions. Evidently, exposure to tumor necrosis factor-alpha may promote neuroplasticity of brain circuits involved in mediating neuroendocrine, sickness or inflammatory responses. It is suggested that such a sensitization may influence the response to immunological and traumatic insults and may thus be relevant to behavioral pathology.

An assessment of the effects of central interleukin-1beta, -2, -6, and tumor necrosis factor-alpha administration on some behavioural, neurochemical, endocrine and immune parameters in the rat.

Despite a vast amount of research into the actions of cytokines within the central nervous system, the pharmacological role and/or physiological function of the various cytokines within the central nervous system is still not fully understood. The present study evaluated the effects of intracerebroventricular administration of interleukin-1beta, -2, -6 (20 ng) and tumour necrosis factor-alpha (40 ng) on elevated plus maze behaviour, monoamine levels in the hypothalamus, hippocampus and amygdala, plasma corticosterone and catecholamine concentrations and Concanavalin A-induced splenic lymphocyte proliferation in the rat. Both interleukin-1beta and tumour necrosis factor-alpha induced "anxiogenic-like" effects on the elevated plus maze, whereas interleukin-2 and interleukin-6 did not. However only interleukin-1beta led to endocrine variations often associated with stress and anxiety. Cytokine specific alterations in monoamine levels were evident in the hypothalamus and hippocampus, while neurotransmitter concentrations in the amygdala were not significantly altered by cytokine treatment. In addition, interleukin-1beta reduced Concanavalin A-induced lymphocyte proliferation, whereas the other cytokine treatments failed to significantly alter this response. These results demonstrate that in some, but not all, respects interleukin-1beta administration produced "stress like" effects on behaviour, monoamine neurotransmitters, hypothalamic pituitary adrenal axis activity and immune function, while the other cytokines produced less consistent effects on these parameters. It is noteworthy that although interleukin-1beta and tumour necrosis factor-alpha provoked an anxiogenic response in the elevated plus maze test of anxiety, neither cytokine significantly altered amygdaloid noradrenergic or serotonergic activity, as many previous studies have implicated increased amygdaloid noradrenergic and/or serotonergic activity in the pathophysiology of anxiety.

Neuroendocrine measures and lymphocyte subsets in depressive illness: influence of a clinical interview concerning life experiences.

The effects of a clinical interview concerning either positive or negative day-to-day events on lymphocyte subpopulations, and on plasma cortisol, ACTH and norepinephrine, were determined in depressive patients (major depressive and dysthymic) and in normal controls. Irrespective of its content, the interview provoked an elevation of circulating natural killer (NK) cells, suggesting that this effect was related to either a change in mood state (regardless of its valence) or to the stress associated with the interview procedure. Since the interview did not influence plasma cortisol, ACTH or norepinephrine, it is likely that the NK cell variations were independent of these endocrines. Although basal NK cells were elevated in the depressive group relative to controls, the extent of the NK cell increase provoked by the interview was comparable in depressive and control subjects. The failure to detect differences between these populations could not be attributed to ceiling effects precluding more pronounced alterations in the depressed subjects. Indeed, variations of circulating cell subtypes were found to be exquisitely sensitive to differences in stressor intensity. In a subset of control subjects, a more potent stressor (anticipation of an academic examination) increased the plasma endocrine levels, increased circulating NK cell number beyond that associated with the interview stress, and provoked an increase of several T cell subsets (CD3, CD4 and CD8). Evidently, while a clinical interview may be sufficiently stressful to influence circulating NK cells, the stress of such a procedure seems no greater in depressed than in control subjects. It is suggested that although depressed patients may exhibit higher basal NK levels, this effect is likely not related to increased reactivity to stressors.