The Plasma Levels of Interleukin-12 in Schizophrenia, Major

Home , Interleukin 12

Molecular Psychiatry (2002) 7, 1107–1114  2002 Nature Publishing Group All rights reserved 1359-4184/02 $25.00 www.nature.com/mp ORIGINAL RESEARCH ARTICLE The plasma levels of interleukin-12 in schizophrenia, major depression, and bipolar mania: effects of psychotropic drugs Y-K Kim1, I-B Suh2, H Kim1, C-S Han1, C-S Lim2, S-H Choi3 and J Licinio4

1Department of Psychiatry, Korea University Ansan Hospital, College of Medicine, Korea; 2Clinical Pathology, Korea University Ansan Hospital, Korea; 3Yong-In Mental Hospital, Kusungmyun, YongIn-Kun, Kyunggi-Do, Korea; 4UCLA Neuropsychiatric Institute and Brain Research Institute, CA, USA

Interleukin-12 (IL-12) plays a key role in promoting T helper 1 (Th1) responses and subsequent cell-mediated immunity. Given the role of cytokines in the pathogenesis of psychiatric disorders, the dysregulation of IL-12 in these illnesses would be expected. We measured the plasma levels of IL-12 in 102 psychiatric patients (43 schizophrenia, 34 major depression and 25 bipolar disorder) and 85 normal controls. In addition, IL-12 levels of the patients were measured after an 8-week treatment to assess whether the levels were affected by medication. The IL-12 levels of the patient group with major depression were significantly higher than that of the control group, whereas no differences were found among the other groups. IL-12 values of the three patient groups decreased significantly after 8 weeks of treatment. These findings support the hypothesis that activation of the inflammatory response system and in particular of Th-1-like cells, is involved in the pathophysiology of major depression and that repeated administration of antidepressive and antipsychotic drugs may suppress IL-12 plasma concentrations in psychiatric patients. Molecular Psychiatry (2002) 7, 1107–1114. doi:10.1038/sj.mp.4001084 Keywords: interleukin-12; schizophrenia; major depression; mania; inflammatory response system; cytokine

Introduction immune activation in schizophrenia has been based on the findings of increased monocyte numbers, increased Over the past 10 years, it has been hypothesized that activity of macrophage, increased proinflammatory schizophrenia and major depression may be cytokine levels such as IL-1, IL-6, IL-18 and tumor accompanied by an activation of inflammatory necrosis factor(TNF)-␣.7–11 Schizophrenia is also response system, which could be related to the patho- characterized by a decrease in IL-2 production12–14 and physiology or etiology of those illnesses. The macro- interferon(IFN)-␥,7,15 traditionally classified as Th1 phage-T lymphocyte theory of schizophrenia1,2 has cytokines, and an increase in IL-416 and IL-10,17,18 tra- proposed that chronically activated macrophages and ditionally classified as Th2 cytokines. On the other T lymphocytes, along with excessive interleukin-2 and hand, there have been reports of an increased Th1 sys- other cytokine secretions are a cause of some cases of tem in schizophrenia, specifically an increased IL-2 schizophrenia. Similarly, the monocyte-T-lymphocyte level in plasma19 and in cerebrospinal fluid20,21 and hypothesis of major depression3 also suggested that increased production of IL-2 and IFN-␥.22 Interestingly, activation of peripheral blood monocytes and T lym- in our previous study,13 we found that decreased IL-2 phocytes play a role in the pathophysiology or patho- production may be associated with an increased IL-2 genesis of that illness.4,5 serum level in schizophrenic patients, supporting the Schizophrenia is characterized by increased nonspe- T cell exhaustion theory that lower IL-2 production cific innate immunity, decreased type 1 T helper cell may be induced by a consequence of overproduction (Th1)-specific cellular immunity, and a Th1–Th2 of in vivo IL-2. imbalance with a shift to the Th2 system.6 The innate It has been postulated that major depression may be accompanied by significant changes in cell-mediated and humoral immunity, and these changes may be Correspondence: Y-K Kim, Department of Psychiatry, Korea Uni- related to the pathophysiology or pathogenesis of that versity Ansan Hospital, College of Medicine, 516 Go-Jan Dong, illness.23–26 Pro-inflammatory cytokines including IL- Ansan city, Kyunggi Province, 425–070, Korea. E-mail: yong- ␤ 4 9 ␣27 kuȰkorea.ac.kr 1 , IL-6, and TNF- in blood are increased in major Received 7 September 2001; revised 5 November 2001; accepted depression. These findings suggest that innate immun- 16 January 2002 ity is activated by secretion from monocytes and Plasma IL-12 levels in psychiatric disorders Y-K Kim et al 1108 macrophages during major depression. Moreover, Lici- phrenia and major depression, and that upon treatment nio and Wong25 suggested that brain cytokines, princi- these increased levels may return to baseline. pally IL-1␤ and IL-1 receptor antagonist, may have a role in the biology of major depression, and that central and peripheral cytokine compartments are integrated Subjects and methods but differentially regulated. Reports on the assessment of the inflammatory Subjects response system in bipolar mania are limited and Among psychiatric patients newly admitted in closed remain controversial, but cell-mediated immunity acti- wards of the Department of Psychiatry, Korea Univer- vation in bipolar mania has been suggested. There have sity Medical Center, during August 1999 to June 2001, been some reports that bipolar mania is related to the we recruited 102 psychiatric patients (43 schizo- activation of the inflammatory response system, phrenia, 34 major depression, and 25 bipolar mania) characterized by increased plasma concentrations of who met the Diagnostic and Statistical Manual (DSM- soluble interleukin-2 receptor (sIL-2R) and sIL-6R,28–30 IV) criteria.40 Each patient was given a diagnostic and increased acute phase proteins.31 assessment based on clinical interviews using a Struc- If the cytokine abnormalities play a causative role in tured Clinical Interview for DSM-IV.41 All patients had the pathophysiology of psychiatric disorders, it is not psychotic or active symptoms at the time of study unreasonable to hypothesize that psychotropic drugs enrollment. All patients were either medication-naive may be able to suppress the release of cytokines. Actu- (first-onset) or medication-free for at least 4 months. ally, typical antipsychotic drugs (eg haloperidol) seem The patients gave informed consent after the procedure to have negative immunoregulatory effects19,28,32 and had been fully explained. atypical neuroleptics (eg clozapine and risperidone) The psychopathological status of the patients was appear to have complex in vivo immunomodulatory assessed by a trained physician (Y-K Kim) using the effects.33,34 On the one hand, antidepressants impair Brief Psychiatric Rating Scale42 for patients with the release of proinflammatory cytokines from acti- schizophrenia, the Hamilton Rating Scale43 for vated monocytes and macrophages and enhance the Depression for patients with major depression, and the expression of anti-inflammatory cytokines. For Rating Scale44 for Mania for bipolar mania. Each example, clomipramine, sertraline, and trazodone patient’s symptoms were assessed on admission and 8 inhibit the stimulated production of IFN-␥, whereas weeks later by the same physician. clomipramine and sertraline increased the production Twenty-nine of 43 schizophrenic patients, 17 of 34 of IL-10 in an in vitro study.35 major depressive disorder patients, and 17 of 25 Interleukin-12 (IL-12) is a heterodimeric cytokine bipolar manic disorder patients completed the 8-week that is produced primarily by monocytes and macro- study. The high rate of drop-outs in our study was phages.36 IL-12 plays a central role in promoting Th1 caused by the study protocol to include patients only responses and, hence, cell-mediated immunity.37 This in hospitalization for the 8 weeks of study. The reasons function is promoted by the IL-12-induced production for the failure of follow-up included the change to out- of IFN-␥ from both resting and activated natural killer patient department treatment before the termination of (NK) cells and T cells, by enhancing the cytotoxic the study (91.4%) and the switch to other drugs due to activity of NK cells, and by cytotoxic T lymphocyte adverse effect or unresponsiveness (9.6%). Among the generation. Thus, IL-12 is a proinflammatory cytokine schizophrenia patients, 17 patients were medicated with immunoregulatory functions that bridge innate with risperidone (mean 7.5 mg; range 6–14 mg), five resistance and antigen-specific adaptive immunity.38 patients with olanzapine (mean 16 mg; range 10– IL-12 produced during early phases of infection and 20 mg), four patients with nemonapride (mean 13.2 inflammation sets the stage for the ensuing antigen-spe- mg; range 8–18 mg), two patients with clozapine cific immune response, favoring differentiation and (270 mg and 300 mg), and one patient with haloperidol function of Th1 T cells while inhibiting the differen- (21 mg). Among the major depression patients, seven tiation of Th2 T cells.39 Thus, given the role of IL-12 patients were medicated with nefazodone (mean in the cascade of cytokine regulation, we hypothesized 318 mg; range 100–450 mg), six patients with paroxet- that another pro-inflammatory cytokine, IL-12 could be ine (mean 36.6 mg; range 20–40 mg), three patients dysregulated in some psychiatric disorders. To our with fluoxetine (mean 46.6 mg; range 40–60 mg), and knowledge, studies of IL-12 have not been undertaken one patient with venlafaxine (300 mg). Among these in the field of psychiatry. Hence, we measured the major depressive disorder patients, five patients were plasma levels of IL-12 in a large cohort of 102 psychi- also given an antipsychotic medication (haloperidol or atric patients (43 schizophrenia, 34 major depression risperidone) for controlling psychotic symptoms. and 25 bipolar disorder) and 85 normal controls. In Among the patients with bipolar disorder, nine addition, IL-12 levels of the patients were measured patients were medicated with lithium (mean after an 8-week treatment to assess whether the levels 1050 mg day−1; range 900–1500 mg day−1), six patients were affected by medication. It was hypothesized that with valproate sodium (mean 1037.5 mg day−1; range if the monocyte-T lymphocyte hypothesis of schizo- 900–1200 mg day−1), and two patients with both lith- phrenia or major depression is correct, increased ium and valproate sodium. Three of the 17 bipolar dis- plasma levels of IL-12 should be found in schizo- order patients also took an antipsychotic medication

Molecular Psychiatry Plasma IL-12 levels in psychiatric disorders Y-K Kim et al 1109 (risperidone or olanzapine) for controlling psychotic of the optical density, and the best fit line was deter- symptoms. mined by regression analysis. The IL-12 level of each Patients with a history of any concomitant psychi- sample was determined by the concentration read from atric illness, such as substance or alcohol abuse, a his- the standard curve. The intra- and inter-assay coef- tory of infection, or a known autoimmune disease were ficients of variance were less than 5%. All assays for excluded. Patients were found to have a normal physi- both patients and controls were carried out in the same cal state as seen from normal values of blood and urine run in duplicate by the same operator (I-B Suh). tests, such as SGOT, SGPT, hemoglobin, hematocrit, serum electrolytes, blood urea, and creatine. Patients Statistical analysis were given a standard set of tests by the Venereal Dis- Study groups were compared for continuous covariates ease Research Laboratory (VDRL) and had normal elec- by a two-tailed t-test or an analysis of variance with trocardiograms (EKG) and electroencephalograms the Dunnett or Scheffe test. A multiple comparison (EEG). The laboratory tests were rechecked after 8 method of the Dunnett test was used when three dis- weeks and were not significantly different compared ease groups were compared to the control group, and with the admission. the Scheffe test was used for all pairwise comparisons. Eighty-five healthy controls were recruited among For discrete covariates, study groups were compared healthy individuals who visited the same hospital for by a chi-square test. The pre- and post-treatment IL-12 regular health screening during the same time period values of patient groups were compared by repeated as the patients. Each patient was given a non-struc- measures analysis of variance. Pearson’s product tured clinical interview. Patients with any personal or moment correlation coefficients were calculated to familial history of psychiatric illness, diagnosed auto- examine the relationships between IL-12 level and immune disease, or substance or alcohol abuse were clinical variables. The null hypothesis was rejected at excluded. All subjects were free of chronic and acute P Ͻ 0.05. physical illness (such as infectious or allergic diseases) associated with abnormal cell-mediated immunity within the 2 weeks before the study. They showed nor- Results mal laboratory findings in blood chemistry, renal function, thyroid function, liver function, VDRL tests, chest Demographic data X-ray of the lungs and heart, EKG, and EEG. Table 1 lists the demographic data of the 187 study subjects. There were no significant differences in the Methods male:female ratio between the four study groups ␹2 = = = Ten ml of fasting blood were withdrawn with a lithium ( 6.55; df 3; P 0.09), and there were no signifi- heparin vacuum tube between 8.00 am and 9.00 am cant differences in body mass index (BMI) between the = = = and immediately centrifuged at 3800 rpm for 10 min. three diagnostic groups (F 0.53; df 3; P 0.66). Plasma was stored at −70°C until thawed for assay. For However, the control and bipolar mania patient groups the three patient groups, blood was sampled both on were significantly younger than the depression patient = = = admission and 8 weeks later. group (F 6.23; df 3; P 0.0005), and in each group The plasma concentrations of IL-12 were measured male patients were significantly younger than female = = = by an enzyme-linked immunosorbent assay (ELISA) patients (F 8.02; df 1; P 0.005). technique using a commercially available kit (Quantikine; R&D Systems, USA) according to the Plasma IL-12 levels manufacturer’s instructions. In brief, these assays are After adjusting for age differences (F = 0.29; P = 0.59), performed in a microplate using the quantitative sand- there were highly significant differences among the IL- wich enzyme immunoassay technique with a mono- 12 values among the four study groups (F = 3.49; df = 3; clonal antibody specific for IL-12. Ninety-six-well P = 0.017). The IL-12 values of the depression group polystyrene microtiter plates were pre-coated with a were significantly higher than those of the control murine monoclonal antibody specific for IL-12. Stan- group (Scheffe test, P Ͻ 0.05), whereas no differences dards and samples were pipetted into the wells, incu- were found between other groups (Figure 1). Within bated 2 h at room temperature, washed each well with each diagnostic group, there were no significant differ- wash buffer twice, added 200 ␮l of IL-12 conjugate to ences of IL-12 levels among subgroups of patients with each well, incubated 1.5 h at room temperature, different disease subtypes or taking different medi- washed twice, added 200 ␮l of substrate solution, incu- cations (data not shown). Concerning the effect of bated for 20 min at room temperature, added 50 ␮lof smoking, IL-12 level did not differ between smokers stop solution. The optical density of each well was and non-smokers among the 102 patients (smokers, n measured using a microplate reader (Coda automatic = 35, 12.6 ± 11.7 pg ml−1; non-smokers, n = 67, 11.6 ± EIA analyzer; Bio-Rad, USA) set to 450 nm and cor- 9.7 pg ml−1, t = 0.43, df = 100, P = 0.6), nor among the rected wavelength set to 540 nm. Duplicate readings 85 controls (smokers, n = 22, 9.7 ± 4.3 pg ml−1; non- were averaged for each standard, control, and sample, smokers, n = 63, 9.2 ± 3.3 pg ml−1, t = 0.47, df = 83, P subtracting the average zero standard optical density. = 0.6). In the 102 patients, no significant correlations For the standard curve, the data points were linearized were found between IL-12 levels and any clinical vari- by plotting the log of the IL-12 concentration vs the log able including age, duration of illness, age of onset,

Molecular Psychiatry Plasma IL-12 levels in psychiatric disorders Y-K Kim et al 1110 Table 1 Demographic data of study subjects

Schizophrenia Major depression Bipolar mania Normal controls (n = 43) (n = 34) (n = 25) (n = 85)

Sex (male/female) 12/31 6/28 12/13 28/57 Age (years) 33.6 ± 8.8 (14–58) 39.6 ± 13.7 (21–67) 28.6 ± 8.0 (17–49) 32.5 ± 9.6 (18–66) BMI (kg m−2) 21.8 ± 3.0 (14.6–28.2) 21.4 ± 2.3 (17.1–28.0) 21.9 ± 2.7 (17.0–28.7) 22.1 ± 2.6 (16.9–30.2) Age of onset (years) 29.2 ± 8.8 (14–57) 37.9 ± 15.4 (20–63) 26.7 ± 9.1 (15–49) Duration of total illness 51.5 ± 47.8 (6–180) 21.3 ± 36.9 (1–144) 18.7 ± 39.9 (1–168) (months) Duration of current episode — 2.0 ± 0.7 (1–3) 1.6 ± 1.1 (1–6) (months) Number of previous 1.2 ± 2.0 (0–7) 0.4 ± 0.9 (0–4) 0.7 ± 1.2 (0–5) admissions (n) Number of episodes (n) — 0.5 ± 1.2 (0–6) 0.7 ± 1.2 (0–5) Family history (yes/no) 4/39 4/30 2/23 Medication status on admission Medication-naive 22 16 18 Medication-free 21 18 7 Subtypes Schizophrenia paranoid 26 undifferentiated 11 disorganized 6 Major depression melancholic 18 atypical 9 psychotic 7 Bipolar mania psychotic 16 non-psychotic 9 Psychopathology scorea On admission 26.5 ± 9.6 (n = 43) 25.4 ± 6.8 (n = 34) 32.6 ± 6.9 (n = 25) At 8 weeks 11.3 ± 9.6 (n = 29) 5.2 ± 4.1 (n = 17) 1.8 ± 2.6 (n = 17)

aThe psychopathological scores were based on the Brief Psychiatric Rating Scale for patients with schizophrenia, the Hamilton Rating Scale for Depression for patients with major depression, and the Rating Scale for Mania for patients with bipolar mania.

BMI, number of admissions, number of episodes, and psychopathology scores (data not shown).

Changes in IL-12 levels before and after 8 weeks of treatment Sixty-three patients (29 schizophrenia, 17 major depression, and 17 bipolar mania) completed the 8- week trial and were included in a second analysis. Within each diagnostic group, there were no significant differences of IL-12 levels between the drop-outs and the patients who complete the 8 week trial (data not shown). The changes in IL-12 levels between before and after treatment were analyzed by repeated measures analysis of variance. Body weight was used as a covariate, since there were significant differences in weight gains among the three patients groups, with more weight gains for bipolar mania patients compared to schizophrenia patients (F = 3.63; df = 2; P = 0.03). IL-12 values Figure 1 Plasma interleukin-12 levels in schizophrenia, of the three patient groups were significantly lower major depression, bipolar mania and normal controls. after 8 weeks of treatment (F = 4.40; df = 1; P = 0.04) *P Ͻ 0.05 by Scheffe test compared with normal controls. (Figure 2). The IL-12 values of the three groups were

Molecular Psychiatry Plasma IL-12 levels in psychiatric disorders Y-K Kim et al 1111

Figure 2 Course of plasma interleukin-12 levels from admission to re-investigation at 8 weeks in 29 schizophrenic patients (left), in 17 major depressive patients (middle) and 17 bipolar manic patients (right). significantly different (F = 7.36; df = 2; P = 0.0015). The observed in depressed patients.46 IL-12 is produced IL-12 values between male and female patients were primarily by the antigen-presenting cells and exerts significantly different (F = 9.90; df = 1; P = 0.0026). The immunoregulatory effects on T and NK cells, and regu- interaction effect of three patients groups and sex was lates the balance between type I (producing IL-2 and significant (F = 7.06; df = 2; P = 0.0018). Since the inter- IFN-␥) and type 2 helper T cells (producing IL-4 and action of three groups and sex was significant, we made IL-10).36,37 Thus, the finding of increased IL-12 in our a separate analysis for each sex; among male patients, major depressed patients may be considered as the IL-12 values of the depression group were the high- additional evidence for an inflammatory response sys- est of all patient groups (F = 5.83; df = 2; P = 0.0134), tem during major depression. but among female patients no significant differences in For schizophrenia and bipolar mania, we did not IL-12 values between the three patient groups were find significant differences in IL-12 levels between found (F = 0.46; df = 2; P = 0.6329). There were no sig- these two diagnostic groups and controls. These find- nificant correlations between the changes of IL-12 and ings were contrary to our expectation that IL-12 would the changes of the psychopathology scores (data not be higher in schizophrenia, since the signs of an shown). inflammatory disease process have been documented in schizophrenia.6,47 However, it is worthy of note that though the means of IL-12 levels of two diagnostic Discussion groups were not different from those of normal con- The first major finding of this study is that major trols, the standard deviation between schizophrenia depressive disorder patients had significantly higher and normal controls was quite large (10.3 vs 2.5, plasma levels of IL-12 than normal controls. Moreover, respectively). This finding suggests a considerable het- this result remained significant, even after adjusting for erogeneity in the levels of IL-12 among individual age. The finding of increased IL-12 levels in major schizophrenic patients. Therefore, it is possible that depression may contribute to the activation of immune there is, in fact, a subgroup of schizophrenic patients response in that illness. Indeed, there is now strong who have markedly elevated IL-12 during episodes, evidence that major depression may be accompanied while others have normal levels even during episodes. by significant changes in cell-mediated and humoral This result could further suggest that the immune immunity. There is some evidence for increased pro- abnormalities may occur only in a subgroup of patients duction of pro-inflammatory cytokines such as IL-1␤, with schizophrenia. This concept is supported by a IL-6, TNF-␣, and IFN-␥ in culture supernatant of previous study48 that treatment-resistant schizophrenia mitogen-stimulated peripheral blood mononuclear (TRS) had significantly higher serum IL-6 levels than cells (PBMC), increased serum concentrations of IL-6 healthy volunteers, whereas patients with non-TRS did and IL-2, and changes in the concentrations of some of not differ in serum IL-6 levels from controls. In con- the respective receptors, such as the IL-2 and the IL-6 trast, bipolar mania patients did not differ much from receptor in major depression.5,24,45 Moreover, increased normal controls regarding the standard deviation (3.5 neutrophil and monocyte phagocytosis has been vs 2.5, respectively). Although there have been some

Molecular Psychiatry Plasma IL-12 levels in psychiatric disorders Y-K Kim et al 1112 recent studies29–31 showing an inflammatory response Th-2 cytokines, and a decrease in IL-2 and IFN-␥, system in bipolar mania, our findings did not support classified as Th1 cytokines.58 Taken together, it is the hypothesis of immune abnormalities in IL-12 in speculated that psychotropic drugs are related to their that illness. negative immunoregulatory effects, and in this context, The second major finding of this study is that the 8 IL-12 levels would be decreased via the reduced weeks of treatment with psychotropic drugs resulted monocyte/macrophage activity caused by the psycho- in significant decreases of IL-12 levels. Moreover, these tropic drugs. findings were not affected by the weight changes in Whether IL-12 in our major depression patients plays these patients. At the present time, we cannot clearly a causative role in the pathophysiology of that illness explain the mechanism of decreased IL-12 levels by the is uncertain. Even if IL-12 levels in our depressed various drugs. However, considering the effects of psy- patients decreased significantly over the 8-week hospi- chotropic drugs on cytokines, it is possible that the talization period, during the same time as a decrease immunosuppressive effects, mediated by decreased in HAM-D scores was apparent, we did not find a sig- monocyte/macrophage functions, may be responsible nificant correlation between the changes of IL-12 and for the reduced IL-12 levels and subsequent Th1–Th2 those of HAM-D scores. Therefore, further investi- imbalance. As for antidepressants, it is reported that gations regarding the temporal relationship between IL- increased monocyte phagocytosis in depressed patients 12 and mood changes are needed to clarify its clini- was reversed upon successful antidepressant treat- cal significance. ments.46 Also, levels of pro-inflammatory cytokines, Some potential limitations of the present study such as IL-1␤ and IL-6, which are released mainly from should be noted. First, because the plasma samples of monocytes and macrophages, returned to control levels our patients had different storage times (from 1 month over a 6-week period upon successful antidepressant to about 2 years), it is possible that this might have treatment in major depression patients.49,50 Alterna- affected the results. However, in our patients, the stor- tively, in an in vitro study,35 antidepressants inhibited age time was not correlated with IL-12 concentrations the stimulated production of IFN-␥, while increasing (data not shown). Furthermore, a recent study53 the production of IL-10, an anti-inflammatory cytokine, reported that the storage time within 3 years after the in whole blood of healthy volunteers. Moreover, a var- blood was collected did not affect the inflammatory iety of antidepressants including imipramine, ven- response system. Second, our patients were prescribed lafaxine, and fluoxetine are recently reported to have a a variety of psychotropic drugs. Though our result sug- common negative immunoregulatory effect by sup- gests that IL-12 had inhibiting effects independent of pressing the IFN-␥:IL-10 ratio.51 the class of psychotropic drugs, the immune system is It has been reported that typical and atypical anti- composed of complex regulatory mechanisms. Due to psychotic drugs can modulate the production of cyto- the variable medication and the small groups of kines or cytokine receptors. Typical antipsychotic patients using similar drugs in this study, a further rep- drugs such as haloperidol are known to have immuno- lication study that examines the effects of each psycho- suppressive activities through stimulation of the pro- tropic drug on IL-12 in the larger members of patients duction of the IL-1 receptor antagonist and suppression will be warranted. Third, it has been suggested that of the production of inflammatory cytokines, such as numerous confounding factors such as age, BMI, gen- IL-2, IL-6, and IFN-␥.19,32,33,52 Atypical antipsychotics der, smoking habits, ongoing or recent infectious dis- such as risperidone and clozapine also appear to have eases, or prior medication may affect the plasma levels anti-inflammatory activities,32,52,53 although short-term of cytokines.59 Coagulation can activate monocytes treatment induces the production of pro-inflammatory during blood sampling.60 Thawing also increases IL-6 cytokines such as IL-6, IFN-␥ and TNF-␣, but the effect levels, and the type of tubes can affect cytokine disappears upon prolonged treatment.31,54,55 In measurements.61 Hence, these factors deserve parti- addition, the proportion of mononuclear phagocytes cular attention. and macrophages in the CSF of schizophrenic patients was significantly higher before treatment, and returned Acknowledgements to normal during treatment with conventional neuroleptic medication.56 We would like to thank Professor Hae-Hiang Song in The effects of mood stabilizers such as lithium or the Department of Biostatistics, Catholic Medical Col- valproate sodium on immune function are not clear, lege for her assistance in data analysis. but some studies have suggested that lithium may have profound immunomodulatory effects in humans. Effec- References tive lithium treatment diminished the plasma concen- 1 Smith RS. Is schizophrenia caused by excessive production of tration of acute phase proteins, a marker for the acti- interleukin-2 and interleukin-2 receptors by gastrointestinal lym- 31,57 vation of the inflammatory response system. phocyte? Med Hypotheses 1991; 34:225–229. Lithium also reduced the soluble IL-6 receptors and the 2 Smith RS, Maes M. The macrophage-T lymphocyte theory of soluble IL-2 receptors to a normal level in symptomatic schizophrenia: additional evidence. Med Hypotheses 1995; 45: 135–141. rapid cycling bipolar patients with a mild inflamma- 3 Maes M, Smith RS, Scharpe S. The monocyte-T-lymphocyte 29 tory response system. Recently, it was found that lith- hypothesis of major depression. Psychoneuroendocrinology 1995; ium caused an increase in IL-4 and IL-10, classified as 20:116–120.

Molecular Psychiatry Plasma IL-12 levels in psychiatric disorders Y-K Kim et al 1113 4 Maes M, Bosmans E, Meltzer H, Scharpe S, Suy E. Interleukin-1: 26 Dantzer R, Wollman E, Vitkovic L, Yirmiya R. Cytokines and a putative mediator of HPA axis hyperactivity in major depression? depression: fortuitous or causative association? Mol Psychiatry Am J Psychiatry 1993; 150: 1189–1193. 1999; 4:328–332. 5 Maes M. Evidence for an immune response in major depression: a 27 Mikova O, Yakimova R, Bosmans E, Kenis G, Maes M. Increased review and hypothesis. Prog NeuroPsychopharmacol Biol Psy- serum tumor necrosis factor alpha concentrations in major chiatry 1995; 19:11–38. depression and multiple sclerosis. Eur Neuropsychopharmacol 6 Muller N, Riedel M, Ackenheil M, Schwarz MJ. The role of immune 2001; 11:203–208. function in schizophrenia: an overview. Eur Arch Psychiatry Clin 28 Maes M, Bosmans E, Calabrese J, Smith R, Meltzer HY. Interleukin- Neurosci 1999; 249:62–68. 2 and interlkin-6 in schizophrenia and mania: effects of neurolept- 7 Wilke I, Arolt V, Rothermundt M, Weitzsch CH, Hornberg M, ics and mood stabilizers. J Psychatr Res 1995; 29: 141–152. Kirchner H. Investigations of cytokine production in whole blood 29 Rapaport MH, Guylai L, Whybrow P. Immune parameters in rapid cultures of paranoid and residual schizophrenic patients. Eur Arch cycling bipolar patients before and after lithium treatment. J Psy- Psychiatry Clin Neurosci 1996; 246:279–284. chiatry Res 1999; 33:335–340. 8 Ganguli R, Yang Z, Shurin G, Chengappa KNR, Brar JS, Gubbi AV et 30 Tsai SY, Chen KP, Yang YY, Chen CC, Lee JC, Singh VK et al. al. Serum interleukin-6 concentration in schizophrenia: elevation Activation of indices of cell-mediated immunity in bipolar mania. associated with duration of illness. Psychiatry Res 1994; 51:1–10. Biol Psychiatry 1999; 45:989–994. 9 Frommberger UH, Bauer J, Haselbauer P, Fraulin A, Riemann D, 31 Maes M, Delange J, Ranjan R, Meltzer HY, Desnyder R, Cooremans Berger M. Interleukin-6 (IL-6) plasma levels in depression and W et al. Acute phase proteins in schizophrenia, mania and major schizophrenia: comparison between the acute state and after depression: modulation by psychotropic drugs. Psychiatry Res remission. Eur Arch Psychiatry Clin Neurosci 1997; 247:228–233. 1997; 66:1–11. 10 van Kammen DP, McAllister-Sistilli CG, Kelley ME, Gurklis JA, 32 Leykin I, Mayer R, Shinitzky M. Short and long term immunosup- Yao JK. Elevated interleukin-6 in schizophrenia. Psychiatry Res pressive effects of clozapine and haloperidol. Immunopharmacol- 1999; 87:129–136. ogy 1997; 37:75–86. 11 Tananka KF, Shintani F, Fujii Y, Yagi G, Asai M. Serum interleu- 33 Maes M, Bosmans E, Ranjan R, Vandoolaeghe E, Meltzer HY, De kin-18 levels are elevated in schizophrenia. Psychiatry Res 2000; Ley M et al. Lower plasma CC16, a natural anti-inflammatory pro- 96:75–80. tein, and increased plasma interleukin-1 receptor antagonist in 12 Ganguli R, Brar JS, Chengappa KNR, Deleo M, Yang ZW, Shurin G schizophrenia: effects of antipsychotic drugs. Schizophr Res 1996; et al. Mitogen-stimulated interleukin-2 production in never-medi- 21:39–50. cated, first-episode schizophrenic patients. Arch Gen Psychiatry 34 Maes M, Bosman E, Kenis G, De Jongh R, Smith RS, Meltzer HY. 1995; 52:668–672. In vivo immunomodulatory effects of clozapine in schizophrenia. 13 Kim YK, Lee MS, Suh KY. Decreased IL-2 production in Korean Schizophr Res 1997; 26: 221–225. schizophrenic patients. Biol Psychiatry 1998; 43:701–704. 35 Maes M, Song C, Lin AH, Bonaccorso S, Kenis G, DeJongh R et al. 14 Arolt V, Rothermundt M, Wandinger KP, Kirchner H. Decreased in Negative immunoregulatory effects of antidepressants: inhibition of vitro production of interferon-gamma and interleukin-2 in whole interferon- and stimulation of interleukin-10 secretion. Neuropsy- blood of patients with schizophrenia during treatment. Mol Psy- chopharmacology 1999; 20:370–379. chiatry 2000; 5:150–158. 36 Gately MK, Renzetti LM, Magram J, Stern AS, Adorini L, Gubler 15 Rothermundt M, Arolt V, Weitzsch C, Eckhoff D, Kirchner H. U et al. The interleukin-12/interleukin-12-receptor system: role in Immunological dysfunction in schizophrenia: a systematic normal and pathologic immune responses. Annu Rev Immunol approach. Neuropsychobiology 1998; 37:186–193. 1998; 16:495–521. 16 Mittleman BB, Castellanos FX, Jacobsen LK, Rapoport JL, Swedo 37 Stern AS, Magram J, Presky DH. Interlukin-12: an integral cytokine SE, Shearer GM. Cerebrospinal fluid cytokines in pediatric neuro- immune response. Life Sci 1996; 58:639–654. psychiatric disease. J Immunol 1997; 159: 2994–2999. 38 Trinchieri G. Interleukin-12: a proinflammatory cytokine with 17 van Kammen DP, McAllister-Sistilli CG, Kelley ME. Relationship immunoregulatory functions that bridge innate resistance and anti- between immune and behavioral measures in schizophrenia. In: gen-specific adaptive immunity. Annu Rev Immunol 1995; 13: Wieselmann G (ed). Current Update in Psychoimmunology. 251–276. Springer Verlag Wien: NY, 1997, pp 51–55. 39 Trinchieri G. Immunobiology of interleukin-12. Immunol Res 1998; 18 Cazzullo CL, Scarone S, Grassi B, Vismara C, Trabattoni D, Clerici 17:269–278. M et al. Cytokine production in chronic schizophrenia patients 40 American Psychiatric Association. Diagnostic and Statistical Man- with or without paranoid behavior. Prog NeuroPsychoPharmacol ual of Mental Disorders, 4th edn). American Psychiatric Press: Biol Psychiatry 1998; 22: 947–957. Washington DC, 1994. 19 Kim YK, Kim L, Lee MS. Relationships between interleukins, neur- 41 First MB, Gibbon M, Spitzer RL, Williams JBW. Structural Clinical otransmitters and psychopathology in drug-free male schizo- Interview for DSM-IV Research Version. Biometric Research phrenics. Schizophr Res 2000; 44:165–175. Department, New York Psychiatric Institute: New York, 1998. 20 Licinio J, Seibyl JP, Altemus M, Charney DS, Krystal JH. Elevated 42 Overall JE, Gorham DR. The brief psychiatric rating scale. Psychol CSF levels of interleukin-2 in neuroleptic-free schizophrenic Rep 1962; 10:799–812. patients. Am J Psychiatry 1993; 150: 1408–1410. 43 Hamilton M. A rating scale for depression. J Neurol Neurosurg Psy- 21 McAllister CG, van Kammen DP, Rehn TJ, Miller AL, Gurklis J, chiatry 1960; 23:56–62. Kelley ME et al. Increases in CSF levels of IL-2 in schizophrenia: 44 Young RC, Biggs JT, Ziegler VE, Meyer DA. A rating scale for effects of recurrence of psychosis and medication status. Am J Psy- mania: reliability, validity and sensitivity. Br J Psychiatry 1978; chiatry 1995; 152: 1291–1297. 133:429–435. 22 Cazzullo CL, Sacchetti E, Galluzzo A, Panariello A, Colombo F, 45 Muller N, Ackenheil M. Psychoneuroimmunology and the cytokine Zagliani A et al. Cytokine profiles in drug-naive schizophrenic action in the CNS: implications for psychiatric disorders. Prog Neu- patients. Schizophr Res 2001; 47:293–298. roPsychopharmacol Biol Psychiatry 1998; 22:1–33. 23 Maes M. Major depression and activation of the inflammatory 46 McAdams C, Leonard BE. Neutrophil and monocyte phagocytosis response system. In: Dantzer R et al (eds). Cytokines, Stress and in depressed patients. Prog NeuroPsychopharmacol Biol Psychiatry Depression. Plenum: New York, 1999, pp 1–24. 1993; 17:971–984. 24 Connor TJ, Leonard BE. Depression, stress and immunological acti- 47 Schwarz MJ, Ackenheil M, Riedel M, Muller N. Blood-CSF barrier vation: the role of cytokines in depressive disorders. Life Sci 1998; impairment as indicator for an immune process in schizophrenia. 62:583–606. Neurosci Lett 1998; 253:201–203. 25 Licinio J, Wong ML. The role of inflammatory mediators in the 48 Lin A, Kenis G, Bignotti S, Tura GJB, DeJong R, Bosmans E et al. biology of major depression: central nervous system cytokines The inflammatory response system in treatment-resistant schizo- modulate the biological substrate of depressive symptoms, regulate phrenia: increased serum interleukin-6. Schizophr Res 1998; 32: stress-responsive systems, and contribute to neurotoxicity and neu- 9–15. roprotection. Mol Psychiatry 1999; 4: 317–327. 49 Sluzewska A, Rybakowski JK, Laciak M, Mackiewicz A, Sobieska

Molecular Psychiatry Plasma IL-12 levels in psychiatric disorders Y-K Kim et al 1114 M, Wiktorowicz K. Interleukin-6 serum levels in depressed patients plasma cytokine and soluble cytokine receptor levels. J Clin Psych- before and after treatment with fluoxetine. Ann NY Acad Sci 1995; opharmacol 1996; 16:403–409. 762:474–476. 56 Nikkila HV, Muller K, Ahokas A, Miettien K, Rimon R, Andersson 50 Seidel A, Arolt V, Hunstiger M, Rink L, Behnisch A, Kirchner H. LC. Accumulation of macrophages in the CSF of schizophrenic Cytokine production and serum proteins in depression. Scand J patients during acute psychotic episodes. Am J Psychiatry 1999; Immunol 1995; 41:535–538. 156: 1725–1729. 51 Kubera M, Lin AH, Kenis G, Bosmans E, van Bockstaele D, Maes 57 Sluzewska A, Sobieska M, Rybakowski JK. Changes in acute phase M. Anti-inflammatory effects of antidepressants through sup- proteins during lithium potentiation of antidepressants in refrac- pression of the interferon-gamma/interleukin-10 production ratio. tory depression. Neuropsychobiology 1997; 35:123–127. J Clin Psychopharmacol 2001; 21: 199–206. 58 Rapaport MH, Manji HK. The effects of lithium on ex vivo cytokine 52 Song C, Lin Ah, Kenis G, Bosmans E, Maes M. Immunosuppressive production. Biol Psychiatry 2001; 50:217–224. effects of clozapine and haloperidol: enhanced production of the 59 Haack M, Hinze-Selch D, Fenzel T, Kraus T, Kuhn M, Schuld A et interleukin-1 receptor antagonist. Schizophr Res 2000; 42:157– al. Plasma levels of cytokines and soluble cytokine receptors in 164. psychiatric patients upon hospital admission: effects of con- 53 Maes M, Chiavetto LB, Bignotti S, Tura GJB, Pioli R, Boin F et founding factors and diagnosis. J Psychiatric Res 1999; 33:407–418. al. Effects of atypical antipsychotics on the inflammatory response 60 Riches P, Gooding R, Millar BC, Rowbottom AW. Influence of col- system in schizophrenic patients resistant to treatment with typical lection and separation of blood samples on plasma IL-1, IL-6 and neuroleptics. Eur Neuropsychopharmacol 2000; 10: 119–124. TNF-␣ concentrations. J Immunol Meth 1992; 153: 125–131. 54 Maes M, Meltzer HY, Bosmans E. Immune-inflammatory markers 61 De Jongh R, Vranken J, Vundelinckx G, Bosmans E, Maes M, Heylen in schizophrenia: comparison to normal controls and effects of clo- R. The effects of anticoagulation and processing on assays of IL-6, zapine. Acta Psychiatr Scand 1994; 89:346–351. sIL-6R, sIL-2R and soluble transferrin receptor. Cytokine 1997; 9: 55 Pollmacher T, Hinze-Selch D, Mullington J. Effects of clozapine on 696–701.

Molecular Psychiatry