Correspondence : Yong-Ku Kim, MD, Department of Psychiatry, College of Medicine, Korea University, Ansan Hospital, 516 Gojan-dong, Ansan 425-020, Korea
Tel : +82-31-412-5140, Fax : +82-31-412-5144, E-mail : yongku@korea.ac.kr

Introduction

Most suicides (about 60%) occur in the context of depressive disorders.¹ The prediction of suicide risk in patients with major depression is very important in preventing suicide. However, the current approaches to predicting suicidality are based on clinical history and have low specificity.² Biological markers may provide a more specific means of identifying individuals at high risk of suicide.
Clinical, postmortem, genetic, and animal studies have suggested that at least 3 neurobiological systems are involved in the pathogenesis of suicidal behavior in major depression. First, a large number of studies using blood platelets, CSF, postmortem brains, functional neuroimaging, and genetics have convincingly shown that a deficient serotonergic system is associated with suicidal behavior in major depression.^3,4,5,6 In addition, evidence of the excessive release of norepinephrine and a decrease in the activity of the dopamine system has been found.^7,8,9 Second, urinary cortisol production, CSF studies, dexamethasore suppression test (DST), and postmortem brain studies suggest that the hyperactivity of the hypothalamicpituitary adronal (HPA) axis is associated with suicidal behavior in major depression.^7,10,11 Third, several studies have shown the involvement of cholesterol in suicidal behavior during major depression.^12,13,14,15 Studies of the biology of suicidal behavior have provided considerable insight into the mechanisms of risk for suicidal behavior, but there are still a variety of candidates that could be biological markers for suicidal behavior in major depression.
It has been postulated that major depression may be accompanied by significant changes in cell-mediated and humoral immunity, and these changes may be related to the pathophysiology or pathogenesis of the illness.^16,17,18 The levels of pro-inflammatory cytokines including IL- 1β, IL-6, IL-12, and TNF-α in the blood are increased in major depression.^19,20,21,22 These findings suggest that innate immunity is activated by the secretion of these cytokines from monocytes and macrophages during major depression. A previous study measured the cytokine secretion of T-cells of suicidal and non-suicidal depressed patients and healthy controls and showed that the T-cells of suicidal depressed patients have Th1 characteristics, while those of non-suicidal depressed patients have Th2 characteristics.²³ A new hypothesis concerning the associations between serum lipids, depression, suicide and atherosclerosis has been formulated that IL-2 plays an important role in lipid metabolism, depression, suicide and atherosclerosis.^24,25 Taken together, these findings suggest that cytokines might constitute a biological marker of suicidal behavior in major depression, but this hypothesis has not been investigated intensively. 
NO is an endogenous gas that is known to influence cerebral monoaminergic activity, including the activity of serotonin.^26,27 In patients with major depression, the total amount and the density of neurons with immunoreactivity to nitric acid synthase (NOS) was found to be reduced in paraventricular neurons,²⁸ and NOS activity was found to be decreased in the prefrontal cortex.²⁹ A recent study revealed that the plasma NO levels were dramatically lower in patients with major depressive disorder compared with healthy controls.³⁰ However, another study found elevated NO levels in patients with major depression compared with both anxiety disorder patients and normal control subjects.³¹ We previously examined the plasma NO metabolites in depressed patients immediately after attempted suicide, non-suicidal depressed patients, and normal controls. Our data showed that increased NO production in the plasma is associated with suicide attempt in depressive patients.³²
Neurotrophins play an important physiological role in the maintenance and growth of neurons and synaptic plasticity in the adult brain.³³ They are known to be involved in the pathogenesis of depression and suicide. In particular, the existence of an association between low serum BDNF and major depressive disorder (MDD) has been suggested.^34,35 In addition, in animals subjected to forced swimming and chronic immobilization stress, the BDNF mRNA expression levels were significantly decreased.^36,37 Moreover, chronic antidepressant treatment increases the expression of BDNF and neurogenesis in the adult rat hippocampus.^34,38 Recently, it was reported that BDNF and neurotrophin-3 (NT-3) were decreased in the post-mortem brains of suicide victims.³⁹ In addition, the mRNA levels of nerve growth factor (NGF), NT-3, NT-4/5, cyclophilin, and neuron-specific enolase (NSE) were decreased in the hippocampus of suicide victims.⁴⁰
In the present study, we examined the plasma levels of various cytokines (IL-2, IL-4, IL-6, TNF-α, IFN-γ, and TGF-β1), neurotrophins (BDNF, β-NGF, and IGF-1) and the NO levels in suicidal depressed patients and normal controls. We then determined the correlation between these biological factors and the severity of the depression or the lethality of the suicide attempt. The ultimate goal of this study is to understand the biological factors that contribute to the risk of suicidal behavior in major depression. 

Methods

Subjects 
   The suicidal subjects were depressed patients admitted to the emergency rooms of Korea University Ansan Hospital, Soonchunhyang University Cheonan Hospital and Seoul Hospital following suicide attempts between August 2004 and October 2006. Initial psychiatric interviews and blood sampling were conducted as soon as possible after admission. The patients met the DSM-IV criteria for major depressive disorder⁴¹ as determined by the Structured Clinical Interview for DSM-IV.⁴² We defined attempted suicide as self-harm behaviors with at least some intent to end one's life. Subjects were excluded if their self-injurious behaviors were determined to have no suicidal intention or ideation. Also, those subjects with general medical problems or a psychiatric comorbidity with Axis I or Axis II disorders such as substance abuse or personality disorders according to the DSM-IV criteria were excluded. Finally, 48 suicidal depressed patients (18 males and 30 females) were included. Out of these 48 suicidal patients, 33 had not taken psychotropic medications within the last 2 weeks. The other 15 were taking various antidepressants and benzodiazepine medications on admission. 
The corresponding normal controls were randomly selected healthy individuals who visited the above University hospitals for regular health screenings during the same period. Only healthy persons who had neither a self-reported personal or familial psychiatric history nor medication history and had scores below 10 on the Beck Depression Inventory (BDI) and below 40 on the State-Trait Anxiety Inventory (STAI) were included in this study. Fifty normal controls (26 male, 24 female) were selected. No statistically significant differences were noted between the groups in terms of their age (F=-0.707, p=0.481) or sex (χ²=081, p=0.149). The de-mographic data on both groups are summarized in Table 1. The study protocol was approved by the Ethics Committee of Korea University, and written informed consent was obtained from all of the patients and control subjects.

Clinical evaluation
A trained psychologist (HJW) evaluated the lethality of each individual suicide attempt using Weisman and Worden's RRR system⁴³ and LSARS-II.^44,45 The Weisman and Worden RRR system⁴³ is a descriptive and quantitative method of assessing the lethality of suicidal acts. According to this system, lethality can be expressed as the ratio of 5 risk and 5 rescue factors that are operationally defined, weighted, and scored. In this system, each of the 5 risk factors is rated on a scale of 1 to 3 points, and the total risk points are then converted to an overall risk score ranging from 1 to 5. The highest risk score is 5 and the lowest is 1. Similarly, each of the 5 rescue factors is rated on a 1 to 3 scale, and the total rescue points are converted into a rescue score ranging from 1 to 5. The RRR score is determined by the formula [A/ (A+B)]×100, in which A is the risk score and B is the rescue score. Lethality ratings range from a low of 17 (representing a low risk score of 1 and a high rescue score of 5) to a high of 83 (representing a high risk score of 5 and a low rescue score of 1). 
The LSARS⁴⁴ is an 11-point scale (0="death is an impossible result" to 10="death is almost certain"). Each point on the scale has comprehensive descriptive anchors that incorporate both the lethality of the means and the context or circumstances of the event. This scale also offers an appendix listing drugs and chemicals and lethal ranges of ingestion by body weight. The updated version of the LSARS (LSARS-II) has a revised table of drugs and chemicals based on more current medications and more recent data on the minimum lethal doses in humans.⁴⁵
The severity of depressive symptoms was also evaluated using Hamilton's 17-item depression rating scale (HDRS).⁴⁶ In our suicidal patients, the means and standard deviations of the RRR, LSARS, and HDRS scores were 32±13, 3.95±2.06, and 27.0±6.3, respectively.

Blood sample collection and testing
Blood samples were drawn from the patients' antecubital veins within 2 hours of their admission to the emergency room after the failed attempt. For the normal controls, blood samples were drawn from the antecubital veins at the time of their visit to the hospital. Approximately 20 ml of blood was collected and placed in a lithium heparin vacuum tube. 

Cytokine assays 
   The human cytokines, IL-2, IL-4, IL-6, TNF-α and IFN-γ, were examined by stimulating the whole blood with phytohemagglutin (PHA) and lipopolysaccharide (LPS) in the culture supernatant.⁴⁷ A mixture of 750μl of RPMI-1640 medium with l-glutamine (Biowhittaker, Walkersville, Maryland, USA) supplemented with 10% fetal bovine serum (FBS, Gibco BRL, USA, Invitrogen, Carlsbad, CA, USA) and 100 IU/ml penicillin (Sigma, Saint Louis, Missouri, USA), 100μg/ml streptomycin (Sigma), and 250μl of whole blood were placed into 24 well cell culture plates. PHA (4μg/ml; Sigma) and LPS (20μg/ml; Sigma) were added and the samples were incubated for 48 h in a humidified atmosphere at 37℃ under 5% CO₂. After incubation, the supernatants were drawn off carefully under sterile conditions, divided into eppendorf tubes, and frozen immediately at -70℃ until thawed for assay. TGF-β1 ELISA was performed on the plasma samples activated with acidic solution, since TGF-β1 is secreted mainly in latent forms.⁴⁸ The human cytokines IL-2, IL-4, IL-6, TNF-α, IFN-γ, and TGF-β1 were assayed using a DuoSet ELISA Development System (R&D Systems, Minneapolis, MN, USA). All of the assays were carried out by the same operator using the recommended buffers, diluents, and substrates. The concentrations of the samples in each plate were calculated according to the appropriate standard curve and dilution factor. The intra- and inter assay coefficients of variation for all of the analyses were less than 8%. 

Nitric Oxide (NO) assays 
   Plasma NO was measured by the Griess reaction as the nitrite concentration after nitrate reduction to nitrite.⁴⁹ For this assay, 50μl of the standards and samples were put into the wells of a 96-well plate, 50μl of the sulfanilamide solution (Promega, Madison, WI, USA) was added and then the plate was incubated for 10 min at room temperature, in an environment protected from light. After adding 50μl of NED solution (Promega) to all of the wells, the plate was incubated for 10 min at room temperature, in an environment protected from light. The absorbance was measured at 540 nm. The nitrite concentration was quantified using various NaNO2 concentrations in plasma as standards and the data were expressed as μM. The inter- and intra-assay coefficient of variation of the plasma NO analysis was <10%. 

BDNF, IGF-1 and β-NGF assays
Human BDNF, IGF-1, and β-NGF were assayed using a DuoSet ELISA Development System (R&D Systems). All assays were performed in duplicate using the manufacturer's recommended buffers, diluents, and substrates. In brief, we diluted the capture antibody to a final concentration of 4μg/ml in phosphate buffered saline (PBS), and 100μl of this mixture was immediately put into the wells of a 96-well plate. We sealed the plate and incubated it overnight at room temperature. The wells were washed 3 times with 400μl/well of PBS containing 0.05% Tween-20 (PBST). Following the addition of 300μl/well of 1% BSA, the plate was incubated for 2 hrs at room temperature, and then washed again 3 times with PBST. 100μl per well of the standards and samples were added and then the plate was incubated for 2 hrs at room temperature and washed 3 times with PBST. After removing all the buffer, we added 100μl of the detection antibody to each well: this was incubated for 2 hrs at room temperature and then washed 3 times with PBST. Diluted streptavidin conjugated to horseradish-peroxidase (100μl/well) was added and this was allowed to incubate for 20 min at room temperature, and then the wells were washed 3 times with PBST. Following a final washing procedure, we added 100μl of substrate solution to each well and this was allowed to incubate for 20 min at room temperature. The substrate was converted to a colored product by the enzyme and the reaction was terminated by the addition of 50μl of 2 N sulfuric acid. The optical density of the color reaction in the wells was read using a microtiter plate reader set to 450 nm. The inter- and intra-assay coefficients of variation were below 10%. The concentrations of the samples in each plate were calculated on the standard curve and the dilution factor. 

Statistical analysis
All statistical analyses were performed with SPSS version 12.0 for Windows. The chi-square test was used to test the categorical variables such as the sex or diagnosis. To evaluate the group differences between the suicidal depressed patients and normal controls for the demographic data such as the age or body mass index (BMI), an independent-samples t-test was conducted. To analyze the biological predictors between the two groups, ANCOVA (Analysis of Covariance) was used if a covariate factor existed. The correlations between the biological factors and individual lethality of suicide attempt or severity of depression were calculated using Pearson's correlation coefficients. A two-tailed probability of less than 0.05 was considered statistically significant. 

Results

General characteristics of the subjects
The demographic data on the suicidal depressed patients and normal controls are summarized in Table 1. There were no significant differences in age or sex between the 2 groups. Among the suicide patients, 36 of the suicidal depressed patients were melancholic type (75%) and the remaining 12 were atypical types (25%). Among the 48 suicide patients, 28 had never made a suicidal attempt previously (58.3%), 12 had attempted once (25.0%), 6 had attempted twice (12.5%) and 2 had attempted over 5 times (4.2%). The average BMI of the suicidal depressed patients was significantly lower than that of the normal controls. The methods of suicide attempts were divided into violent and nonviolent methods, based on Traskman's criteria.⁵⁰ In this study, 26 patients used nonviolent methods: overdose of psychiatric drugs (n=23, 47.9%) and superficial cutting (n=3, 6.3%). The other 22 used violent methods: ingestion of pesticide or agricultural chemicals (n=10, 20.9%), deep laceration (n=9, 18.8%), and hanging (n=3, 6.1%). 

Comparison of biological factors between suicidal depressed patients and normal controls
ANCOVA was performed with BMI as a covariate factor. The suicidal depressed patients had significantly higher IL-6, TGF-β1, and NO levels compared with the normal controls after controlling for BMI (Table 2, Figure 1). On the other hand, the BDNF, IL-2, IL-4 and IFN-γ levels were significantly lower in the suicidal depressed patients than in the normal controls. However, there were no significant differences in the IGF-1, NGF, or TNF-α levels between the two groups. In addition, there were no significant differences in the biological factors between the males and females (data not shown).

Correlations among biological factors, suicidal severity, and depression symptoms in suicidal depressed patients 
   The NO levels in the suicidal depressed patients were negatively significantly correlated with their LSARS and RRR scores (γ=-0.444, γ=-0.368 respectively) (Table 3). The IFN-γ levels in the suicidal depressed patients were positively correlated with their LSARS score (γ=0.356), but the TGF-β1 levels were negatively correlated with their LSARS score (γ=-0.344). The BDNF, IL-2, IL-4, IL-6, TNF-α, IGF-1, and β-NGF levels were not significantly correlated with the LSARS or RRR scores. The BDNF and IL-6 levels in the suicidal depressed patients were positively correlated with their HDRS scores (γ=0.370, γ=0.391, respectively), while the IL-4 levels were negatively correlated with their HDRS scores (γ=-0.337). However, the NO, IL-2, β-NGF, TGF-β1, IFN-γ, and TNF-α levels were not significantly correlated with the HDRS scores.

Comparison of biological factors between violent and nonviolent suicide patients 
   Overdoses of drugs or superficial wrist cutting were considered to be nonviolent suicide attempts, while other methods such as pesticide ingestion or hanging were considered to be violent suicide attempts. When the suicidal behaviors were divided into these categories, the IL-6 and IFN-γ levels of the violent suicide patients were significantly higher than those of the nonviolent suicide patients (Table 4). However, there were no significant differences in the NO, BDNF, IGF-1, NGF, TNF-α, IL-2, IL-4, or TGF-β1 levels between the violent and nonviolent groups. 

Discussion 

The first major finding of this study is that the MDD patients with suicide attempt had significantly lower plasma levels of IL-6, TGF-β, and NO than the normal controls after controlling for BMI. On the other hand, the BDNF, IL-2, IL-4, and IFN-γ levels of the suicidal depressed patients were significantly lower than those of the normal controls. The IGF-1, NGF, and TNF-α levels were not different between the groups.
In major depression, several physiological changes have been reported, viz. activation of the inflammatory response system, increased concentrations of proinflammatory cytokines, and prostaglandin E2, and negative immuno-regulatory cytokines in peripheral blood.^22,51,52 The imbalance between Th1 cytokines and Th2 cytokines or pro-inflammatory and anti-inflammatory cytokines plays a role in the appropriate modulation of cellular responses in the brain during psychological stress and major depression.^16,53 In the present study, it was found that the suicidal depressed patients had decreased Th1-specific cellular immunity (lower IL-2 and IFN-γ), increased Th2-specific cellular immunity (higher IL-6 and TGF-β) and a Th1-Th2 imbalance with a shift toward the Th2 system. In the CNS, IL-6 has both developmental neurotrophic and neurodegenerative activities,^54,55 and both pro-inflammatory and anti-inflammatory activities⁵⁶ under various conditions. In several studies, IL-6 was considered to be a Th2 cytokine, thus supporting the Th2 hypothesis.^57,58 Therefore, our result suggests that suicidal depressed patients may have immune profiles which are different from those of non-suicidal depressed patients. We thus speculate that the T-cells of suicidal depressed patients have Th2 characteristics, while those of non-suicidal depression may have Th1 characteristics. However, our results are not consistent with the findings of Mendlovic et al.⁴³ who reported Th1 activation in suicidal depression. However, since their sample size was very small (n=6), it is difficult to generalize their results. In our study, we found higher TGF-β levels in suicidal depression, and the action of TGF-β has been shown to suppress the production of Th1 cytokines, such as IFN-γ, TNF-α, IL-2, and IL-2R.^59,60 Hence, our finding of a lower level of Th1 cytokines in suicidal depression may come about through the major role of TGF-β1 in maintaining the Th1 and Th2 balance in this disorder. In this regard, immunological switches from Th1 to Th2 dominance may be important in the pathogenesis and pathophysiology of suicidal depression. 
We also found increased NO in suicidal depression. We previously examined NO metabolites in depressed patients who recently attempted suicide, non-suicidal depressed patients, and normal controls and demonstrated that increased NO production in plasma is associated with suicide attempt in depressive patients.³² Our present result is consistent with our previous finding. One explanation for the high NO level in suicidal depression is the interaction of NO and serotonin. Many studies have reported that 5-HT dysfunction is correlated with suicidal behavior or violence.^61,62 Mice lacking endothelial NOS showed enhanced 5-HT turnover in the frontal cortex and ventral striatum and increased 5-HT metabolites in the cerebellum.⁶³ Thus, NO appears to play an important role in normal brain 5-HT function. Another possible explanation is related to the response of the NO system to stress. Some studies have found that the expression of NOS increases in the limbic brain regions of rats subjected to restraint stress.^64,65 A recent study showed that stress evoked an immediate increase in NOS activation that was sustained for 3 weeks poststress in the rat hippocampus.⁶⁶ Thus, it is possible that stress induces higher plasma NO levels in suicidal depression. 
In this study, we found a lower level of BDNF in suicidal depression. Some postmortem studies found a significant decrease of BDNF and neurotrophin-3 levels in the prefrontal cortex and hippocampus of suicide victims.³⁹ One study also reported that the mRNA levels of the BDNF were significantly reduced in both the prefrontal cortex and hippocampus in suicidal subjects compared with those in the control subjects, regardless of the psychiatric diagnosis.⁶⁷ Furthermore, in the present study, the finding of decreased BDNF may be specific to suicidal depression, since the levels of other neurotrophic factors such as β-NGF and IGF-1 were not different between the suicidal depressed patients and normal controls. One explanation for the low BDNF levels in suicidal depressed patients is that decreased serotonin function in suicidal depression could down-regulate BDNF expression. BDNF and serotonin are known to regulate synaptic plasticity, neurogenesis, and neuronal survival, and these two signals have been reported to co-regulate one another.⁶⁸ Therefore, impaired serotonin signaling could decrease the expression of BDNF in suicidal depression. Another possible explanation for the low BDNF levels observed in suicidal depression is that the severe stress associated with attempted suicide could abruptly decrease the level of BDNF after the attempt. Stress can affect the expression of BDNF, and BDNF itself functions as a defense mechanism against stress.⁶⁹ Moreover, stressful events such as a suicide attempt may alter the responsiveness of the HPA system, and the stress-induced elevation of glucocorticoids is known to reduce the expression of BDNF levels in the blood and brain. 
Taken together, these findings suggest that suicidal depressed patients may represent a distinct subgroup of depressed patients, possibly with a different biological and genetic background. A recent study revealed that the hormone response to CRH in the combined dexamethasone-suppression/CRH stimulation test is decreased in depressed patients with suicidal behavior compared with those without suicidal behavior.⁷⁰ Family, twin, and adoption studies have suggested that the overlap of genes predisposing to suicidal behavior and depression may not be complete.⁷¹ This concept is related to the "stress and diathesis" model of suicide, in which the risk for suicidal acts is determined not merely by a psychiatric illness (the stressor) but also by a diathesis.⁷²
In this study, the lethality of suicide measured by two lethality scales was negatively correlated with NO and TGF-β1 levels, but positively correlated with IFN-γ levels. The fact that these factors were not correlated with the severity of depression measured by HDRS suggests that NO, TGF-β1 and IFN-γ may be related to the suicide risk in depression, but not with depression itself. 
Those depressed patients with violent suicidal attempts had significantly higher levels of IL-6 and IFN-γ than those with non-violent suicidal attempts. Our findings are in agreement with the results of other studies which indicated that patients who have had a violent attempt are biologically different from patients with non-violent attempts and healthy controls. Violent patients tend to have low serum cholesterol,¹² increased urinary cortisol,⁷³ low CSF 5-hydroxyindoleactic acid,⁵⁰ high CSF insulin,⁷⁴ and the non-suppression of dexamethasone.⁷⁵ The results of the present study indicate that IL-6 and IFN-γ may be additional biological markers of the tendency toward violent suicide among depressed patients.
In conclusion, the levels of NO, BDNF, and cytokines might be able to be used as biological factors of suicidal behavior in major depression. Future studies should reveal the longitudinal course of the biological markers in depressed patients with and without suicide attempts.

1. Alvarez JC, Cremniter D, Gluck N, Quintin P, Leboyer M, Berlin I, et al. Low serum cholesterol in violent but not in non-violent suicide attempters. Psychiatry Res 2000;95:103-108.

2. APA. Diagnostic and statistical manual of mental disorders (4rd ed). Washington: American Psychiatric Press, 1994.

3. Arango V, Underwood MD, Gubbi AV, Mann JJ. Localized alterations in pre- and postsynaptic serotonin binding sites in the ventrolateral prefrontal cortex of suicide victims. Brain Res 1995;688:121-133.

4. Berman AL, Shepherd G, Silverman MM. The LSARS-II: lethality of suicide attempt rating scale-updated. Suicide Life Threat Behav 2003;33:261-276.

5. Bernstein HG, Stanarius A, Baumann B, Henning H, Krell D, Danos P, et al. Nitric oxide synthase-containing neurons in human hypothalamus: reduced number of immunoreactive cells in the paraventriculat nucleus of depressive patients and schizophrenics. Neuroscience 1998;83:867-875.

6. Brent DA, Bridge J, Johnson BA, Connolly J. Suicidal behavior runs in families. A controlled family study of adolescent suicide victims. Arch Gen Psychiatry 1996;53:1145-1152.

7. Carlson KW, Nawy SS, Wei ET, Sadee W, Filov VA, Rezsova VV, et al. Inhibition of mouse melanoma cell proliferation by corticotropin-releasing hormone and its analogs. Anticancer Res 2001;21:1173-1179.

8. Chrapko WE, Juracs P, Radomski MW, Lara N, Archer S, Le Melledo JM. Decreased plasma nitric oxide metabolites and platelet nitric oxide synthase activity in patients with major depression. Biol Psychiatry 2004;56:129-134.

9. Colin A, Reggers J, Castronovo V, Ansseau M. Lipids, depression and suicide. Encephale 2003;29:49-58.

10. Cooper SJ, Kelly CB, King DJ. 5-Hydroxyindoleacetic acid in cerebrospinal fluid and prediction of suicidal behavior in schizophrenia. Lancet 1992;340:940-941.

11. Coryell W, Schlesser M. The dexamethasone suppression test and suicide prediction. Am J Psychiatry 2001;158:748-753.

12. Cremniter D, Jamain S, Kollenbach K, Alvarez JC, Lecrubier Y, Gilton A, et al. CSF 5-HIAA levels are lower in impulsive as compared to nonimpulsive violent suicide attempters and control subjects. Biol Psychiatry 1999;45:1572-1579.

13. De Groote D, Zangerle PF, Gevaert Y, Fassotte MF, Beguin Y, Noizat-Pirenne F, et al. Direct stimulation of cytokines (IL-1 beta, TNF-alpha, IL-6, IL-2, IFN-gamma and GM-CSF) in whole blood. I. Comparison with isolated PBMC stimulation. Cytokine 1992;4:239-248.

14. Duman RS, Heninger GR, Nestler EJ. A molecular and cellular theory of depression. Arch Gen Psychiatry 1997;54:597-606.

15. Dwivedi Y, Mondal AC, Rizavi HS, Conley RR. Suicide brain is associated with decreased expression of neurotrophins. Biol Psychiatry 2005;58:315-324.

16. Dwivedi Y, Rizavi HS, Conley RR, Roberts RC, Tamminga CA, Pandey GN. Altered gene expression of brain-derived neurotrophic factor and receptor tyrosine kinase B in postmortem brain of suicide subjects. Arch Gen Psychiatry 2003;60:804-815.

17. Fiddler RN. Collaborative study of modified AOAC method of analysis for nitrite in meat and meat products. J Assoc Off Anal Chem 1977;60:594-599.

18. Frisch C, Dere E, Silva MA, Godecke A, Schrader J, Huston JP. Superior water maze performance and increase in fear-related behavior in the endothelial nitric oxide synthase-deficient mouse together with monoamine changes in cerebellum and ventral striatum. J Neurosci 2000;20:6694-6700.

19. Gadient RA, Otten U. Expression of interleukin-6 (IL-6) and inter-leukin-6 receptor (IL-6R) mRNAs in rat brain during postnatal development. Brain Res 1994;637:10-14.

20. Gadient RA, Otten UH. Interleukin-6 (IL-6)-a molecule with both beneficial and destructive potentials. Prog Neurobiol 1997;52:379-390.

21. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960;23:56-62.

22. Harvey BH, Oosthuizen F, Brand L, Wegener G, Stein DJ. Stress-restress evokes sustained iNOS activity and altered GABA levels and NMDA receptors in rat hippocampus. Psychopharmacology (Berl) 2004;174:494-502.

23. Jozuka H, Jozuka E, Suzuki M, Takeuchi S, Takatsu Y. Psycho-neuro-immunological treatment of hepatocellular carcinoma with major depression-a single case report. Curr Med Res Opin 2003;19:59-63.

24. Kamali M, Oquendo MA, Mann JJ. Understanding the neurobiology of suicidal behavior. Depress Anxiety 2001;14:164-176.

25. Kapur S, Mann JJ. Role of the dopaminergic system in depression. Biol Psychiatry 1992;32:1-17.

26. Karege F, Vaudan G, Schwald M, Perroud N, La Harpe R. Neurotrophin levels in postmortem brains of suicide victims and the effects of antemortem diagnosis and psychotropic drugs. Brain Res Mol Brain Res 2005;136:29-37.

27. Kim YK, Lee HJ, Kim JY, Yoon DK, Choi SH, Lee MS. Low serum cholesterol is correlated to suicidality in a Korean sample. Acta Psychiatr Scand 2002;105:141-148.

28. Kim YK, Maes M. The role of cytokine network in psychological stress. Acta Neuropsychiatrica 2003;15:148-155.

29. Kim YK, Myint AM. Clinical application of low serum cholesterol as an indicator for suicide risk in major depression. J Affect Disord 2004;81:161-166.

30. Kim YK, Paik JW, Lee SW, Yoon D, Han C, Lee BH. Increased plasma nitric oxide level associated with suicide attempt in depressive patients. Prog Neuropsychopharmacol Biol Psychiatry 2006;30:1091-1096.

31. Kim YK, Suh IB, Kim H, Han CS, Lim CS, Choi SH, et al. The plasma levels of interleukin-12 in schizophrenia, major depression, and bipolar mania: effects of psychotropic drugs. Mol Psychiatry 2002;7:1107-1114.

32. Lee HJ, Kim YK. Serum lipid levels and suicide attempts. Acta Psychiatr Scand 2003;108:215-221.

33. Leonard BE. Changes in the immune system in depression and dementia: causal or co-incidental effects? Int J Dev Neurosci 2001;19:305-312.

34. Lewin GR, Barde YA. Physiology of the neurotrophins. Annu Rev Neurosci 1996;19:289-317.

35. Madrigal JL, Moro MA, Lizasoain I, Lorenzo P, Castrillo A, Bosca L, et al. Inducible nitric oxide synthase expression in brain cortex after acute restraint stress is regulated by nuclear factor kappa B-medicated mechanisms. J Neurochem 2001;76:532-538.

36. Madrigal JL, Moro MA, Lizasoain I, Lorenzo P, Fernandez AP, Rodrigo J, et al. Induction of cyclooxygenase-2 accounts for restrain stress-induced oxidative status in rat brain. Neuropsychopharmacology 2003;28:1579-1588.

37. Maes M. Major depression and activation of the inflammatory response system. Adv Exp Med Biol 1999;461:25-46.

38. Malberg JE, Eisch AJ, Nestler EJ, Duman RS. Chronic antidepressant treatment increases neurogenesis in adult rat hippocampus. J Neurosci 2000;20:9104-9110.

39. Mann JJ. The neurobiology of suicide. Nat Med 1998;4:25-30.

40. Mann JJ, Huang YY, Underwood MD, Kassir SA, Oppenheim S, Kelly TM, et al. A serotonin transporter gene promoter polymorphism (5-HTTLPR) and prefrontal cortical binding in major depression and suicide. Arch Gen Psychiatry 2000;57:729-738.

41. Marazziti D, Presta S, Silvestri S, Battistini A, Mosti L, Balestri C, et al. Platelet markers in suicide attempters. Prog Neuropsychopharmacol Biol Psychiatry 1995;19:375-383.

42. Mattson MP, Maudsley S, Martin B. BDNF and 5-HT: a dynamic duo in age-related neuronal plasticity and neurodegenerative disorders. Trends Neurosci 2004;27:589-594.

43. Mendlovic S, Mozes E, Eilat E, Doron A, Lereya J, Zakuth V, et al. Immune activation in non-treated suicidal major depression. Immunol Lett 1999;67:105-108.

44. Miller DB, O'Callaghan JP. Depression, cytokines, and glial function. Metabolism 2005;54:33-38.

45. Miyazono K, Yuki K, Takaku F, Wernstedt C, Kanzaki T, Olofsson A, et al. Latent forms of TGF-beta: structure and biology. Ann N Y Acad Sci 1990;593:51-58.

46. Mizuno T, Sawada M, Suzumura A, Marunouchi T. Expression of cytokines during glial differentiation. Brain Res 1994;656:141-146.

47. Montague PR, Gancayco CD, Winn MJ, Marchase RB, Friedlander J. Role of NO production in NMDA receptor-mediated neurotransmitter release in cerebral cortex. Science 1994;263:973-977.

48. Myint AM, Kim YK. Cytokine-serotonin interaction through IDO: a neurodegeneration hypothesis of depression. Med Hypotheses 2003;61:519-525.

49. Nestler EJ, Barrot M, DiLeone RJ, Eisch AJ, Gold SJ, Monteggia LM. Neurobiology of depression. Neuron 2002;34:13-25.

50. Nordstrom P, Samuelsson M, Asberg M, Traskman-Bendz L, Aberg-Wistedt A, Nordin C, et al. CSF 5-HIAA predicts suicide risk after attempted suicide. Suicide Life Threat Behav 1994;24:1-9.

51. Ordway GA, Klimek V. Noradrenergic pathology in psychiatric disorders: postmortem studies. CNS Spectr 2001;6:697-703.

52. Pae CU, Yoon CH, Kim TS, Kim JJ, Park SH, Lee CU, et al. Anti-psychotic treatment may alter T-helper (TH) 2 arm cytokines. Int Immunopharmacol 2006;6:666-671.

53. Penttinen J. Hypothesis: low serum cholesterol, suicide, and interleukin-2. Am J Epidemiol 1995;141:716-718.

54. Pfennig A, Kunzel HE, Kern N, Ising M, Majer M, Fuchs B, et al. Hypothalamus-pituitary-adrenal system regulation and suicidal behavior in depression. Biol Psychiatry 2005;57:336-342.

55. Prud'homme GJ, Piccirillo CA. The inhibitory effects of transforming growth factor-beta-1 (TGF-beta1) in autoimmune diseases. J Autoimmun 2000;14:23-42.

56. Roy A. Hypothalamic-pituitary-adrenal axis function and suicidal behavior in depression. Biol Psychiatry 1992;32:812-816.

57. Russo-Neustadt A, Ha T, Ramirez R, Kesslak JP. Physical activity-antidepressant treatment combination: impact on brain-derived neurotrophic factor and behavior in an animal model. Behav Brain Res 2001;120:87-95.

58. Schiepers OJ, Wichers MC, Maes M. Cytokines and major depression. Prog Neuropsychopharmacol Biol Psychiatry 2005;29:201-217.

59. Schmitt E, Hoehn P, Huels C, Goedert S, Palm N, Rude E, et al. T helper type 1 development of naive CD4+ T cells requires the co-ordinate action of interleukin-12 and interferon-gamma and is inhibited by transforming growth factor-beta. Eur J Immunol 1994;24:793-798.

60. Schwarz MJ, Chiang S, Muller N, Ackenheil M. T-helper-1 and T-helper-2 responses in psychiatric disorders. Brain Behav Immun 2001;15:340-370.

61. Sher L, Mann JJ, Traskman-Bendz L, Winchel R, Huang YY, Fertuck E, et al. Lower cerebrospinal fluid homovanillic acid levels in depressed suicide attempters. J Affect Disord 2006;190:83-89.

62. Smith K, Conroy RW, Ehler BD. Lethality of suicide attempt rating scale. Suicide Life Threat Behav 1984;14:215-242.

63. Smith MA, Makino S, Kvetnansky R, Post RM. Stress and glucocorticoids affect the expression of brain-derived neurotrophic factor and neurotrophin-3 mRNAs in the hippocampus. J Neurosci 1995;15:1768-1777.

64. Spitzer RL, Williams JB, Gibbon M, First MB. The Structured Clinical Interview for DSM-III-R (SCID). I: history, rationale, and description. Arch Gen Psychiatry 1992;49:624-629.

65. Thomas AJ, Davis S, Morris C, Jackson E, Harrison R, O'Brien JT. Increase in interleukin-1beta in late-life depression. Am J Psychiatry 2005;162:175-177.

66. Traskman L, Asberg M, Bertilsson L, Sjostrand L. Monoamine metabolites in CSF and suicidal behavior. Arch Gen Psychiatry 1981;38:631-636.

67. Tuglu C, Kara SH, Caliyurt O, Vardar E, Abay E. Increased serum tumor necrosis factor-alpha levels and treatment response in major depressive disorder. Psychopharmacology (Berl) 2003;170:429-433.

68. van Heeringen K, Audenaert K, Van de Wiele L, Verstraete A. Cortisol in violent suicidal behaviour: association with personality and monoaminergic activity. J Affect Disord 2000;60:181-189.

69. Viljoen M, Panzer A. Proinflammatory cytokines: a common denominator in depression and somatic symptoms? Can J Psychiatry 2005;50:128.

70. Weisman AD, Worden JW. Risk-rescue rating in suicide assessment. Arch Gen Psychiatry 1972;26:553-560.

71. Westling S, Ahren B, Traskman-Bendz L, Westrin A. High CSF-insulin in violent suicide attempters. Psychiatry Res 2004;129:249-255.

72. Xing G, Chavko M, Zhang LX, Yang S, Post RM. Decreased calcium-dependent constitutive nitric oxide synthase (cNOS) activity in prefrontal cortex in schizophrenia and depression. Schizophr Res 2002;58:21-30.

73. Xu H, Qing H, Lu W, Keegan D, Richardson JS, Chlan-Fourney J, Li XM. Quetiapine attenuates the immobilization stress-induced decrease of brain-derived neurotrophic factor expression in rat hippocampus. Neurosci Lett 2002;321:65-68.

74. Yamada K, Noda Y, Nakayama S, Komori Y, Sugihara H, Hasegawa T, et al. Role of nitric oxide in learning and memory and in monoamine metabolism in the rat brain. Br J Pharmacol 1995;115:852-858.

75. Yerevanian BI, Feusner JD, Boek RJ, Mintz J. The dexamethasone suppression test as a predictor of suicidal behavior in unipolar depression. J Affect Disord 2004;83:103-108.