Serotonin-immune interactions in detoxified chronic alcoholic patients without apparent liver disease: activation of the inflammatory response system and lower plasma total tryptophan

Psychiatry Research 78 Ž1998. 151]161

Serotonin-immune interactions in detoxified chronic alcoholic patients without apparent liver disease: activation of the inflammatory response system and lower plasma total tryptophan Michael Maes a,b,U , Ai-hua Lin a , Eugene Bosmans c , Eric Vandoolaeghea , Stefania Bonaccorso a,d , Gunter Kenis e , Raf De Jonghe , Robert Verkerk f , Cai Song a,g , Simon Scharpe ´f , Hugo Neels h a

Uni¨ ersity Department of Psychiatry, Clinical Research Center for Mental Health (CRC-MH), AZS, 267 Lange Beeldekensstraat, 2060 Antwerp, Belgium b Department of Psychiatry, Di¨ ision of Psychopharmacology, Vanderbilt Uni¨ ersity, 1601 23rd A¨ enue South, Nash¨ ille, TN 37212, USA c Eurogenetics, Transportstraat, Tessenderlo, Belgium d Department of Psychiatry, Uni¨ ersity of Rome, La Sapienza, Rome, Italy e Department of Anesthesiology, AZ St-Jan, Genk, Belgium f Department of Medical Biochemistry, Uni¨ ersity of Antwerp, Uni¨ ersiteitsplein, 2610 Wilrijk, Belgium g Life Sciences Research Center, Carleton Uni¨ ersity, 1125 Colonel by Dri¨ e, Ottawa, Ont., Canada K1N 5B6 h Laboratory of Clinical Chemistry, OCMW Hospitals, Antwerp, Belgium Received 21 July 1997; revised 19 December 1997; accepted 16 January 1998

Abstract The aims of the present study were to examine Ž1. the inflammatory response system ŽIRS., through measurements of serum interleukin-6 ŽIL-6., soluble IL-6 receptor ŽsIL-6R., sgp130 Žthe soluble form of the IL-6 transducer signal protein., CC16 ŽClara Cell protein; an endogenous anti-cytokine., IL-1R antagonist ŽIL-1RA., IL-8 and sCD14; and Ž2. the availability of plasma total tryptophan to the brain in chronic alcoholic patients without apparent liver disease ŽAWLD.. Detoxified AWLD patients had significantly lower plasma tryptophan and serum CC16 and significantly higher serum IL-1RA and IL-8 concentrations than normal volunteers. There were significant correlations between the availability of tryptophan to the brain and serum IL-6, IL-8 and IL-1RA Žall negative. and CC16 Žpositive.. The results suggest that Ž1. there is, in detoxified AWLD patients, an activation of the monocytic arm of cell-mediated

U

Corresponding author. Tel.: q32 3 4483265; fax: q32 3 4483265; e-mail: [email protected]

0165-1781r98r$19.00 Q 1998 Elsevier Science Ireland Ltd. All rights reserved. PII S0165-1781Ž98.00010-9

152

M. Maes et al. r Psychiatry Research 78 (1998) 151]161

immunity and a lowered anti-inflammatory capacity of the serum; and that Ž2. lower availability of plasma tryptophan to the brain in detoxified AWLD patients is related to activation of the IRS. Lower CC16 may be one factor predisposing chronic alcoholic patients toward infectious disorders. Q 1998 Elsevier Science Ireland Ltd Keywords: Alcoholism; Cytokines; Interleukin-6; Interleukin-8; CC16; Serotonin

1. Introduction In alcoholic patients, either actively drinking, withdrawing, or detoxified, with or without apparent liver disease, a wide range of alterations in the inflammatory response system ŽIRS. and changes in the peripheral and central metabolism of serotonin Ž5-HT. have been found. During all stages of alcoholism, infectious diseases, such as tuberculosis and pneumonia, are more prevalent and virulent ŽBaker and Jerrels, 1993.. Acute or subchronic treatment with alcohol significantly decreases phytohemagglutinin ŽPHA ., concanavalin A ŽCon A. or antigen-induced lymphocytic responses ŽRoselle and Mendenhall, 1982; Grossman et al., 1988.. Alcohol also suppresses total blood cell counts and CD3q , CD4q and CD8q T cells ŽRoselle et al., 1988.. Other alcohol-induced defects in immune regulation are abnormal interleukin-2 ŽIL-2.-supported proliferation of lymphocytes ŽKaplan, 1986., suppression of tumor necrosis factor-a ŽTNFa . and IL-2 production ŽVerma et al., 1993. and decreased antigen-induced production of TNFa , IL-1, IL-6 by monocytes ŽSzabo et al., 1995.. Therefore the increased susceptibility to infections in alcoholic patients has been ascribed to the overall immuno-inhibitory effects of alcohol ŽJerrels et al., 1990; Nelson et al., 1991; Saad et al., 1993.. Chronic alcoholism complicated by liver disease, such as hepatitis or cirrhosis, on the other hand, is characterized by activation of the IRS. Thus, alcoholic liver disease is characterized by Ž1. increased IL-6 production by mononuclear cells ŽMartinez et al., 1993.; Ž2. increased production of TNFa , IL-1 and IL-6 ŽMartinez et al., 1992; Ruiz et al., 1993; Rodriguez-Rodriguez et al., 1995.; and Ž3. increased circulating levels of IL-1, IL-6 and TNFa ŽKhoruts et al., 1991.. Only a few studies have examined the IRS and

cytokine production in newly detoxified Žabstinent. chronic alcoholic patients without apparent liver disease ŽAWLD.. Cook et al. Ž1991. reported signs of immune activation in newly detoxified AWLD patients. Martinez et al. Ž1993., on the other hand, found no significant differences in mitogen-stimulated IL-6 and IL-8 production between newly abstinent alcoholic patients and normal volunteers. The peripheral and central metabolism of serotonin Ž5-HT. has been the subject of many studies examining the effects of alcohol and alcoholism. Because of methodological limitations, human brain 5-HT turnover has to be assessed through indirect methods, such as measurements of plasma tryptophan, the precursor or 5-HT, or the ratio of plasma tryptophan to the sum of amino acids known to compete for the same cerebral uptake mechanism Žvaline, leucine, tyrosine, phenylalanine, isoleucine. ŽBadawy et al., 1995.. The influx of tryptophan into the brain depends on free and total Žbound to proteins. plasma tryptophan and the competing amino acids ŽYuwiler et al., 1977; Pardridge, 1979.. Thus, total and free tryptophan as well as the tryptophanrCAA ratio are indicators for the availability of tryptophan to the brain ŽCurzon and Sarna, 1984; Fernstrom, 1984. and hence, for 5-HT synthesis in the brain ŽMoir and Eccleston, 1968.. There are some reports that acute consumption of alcohol may lower the availability of tryptophan to the brain ŽEriksson et al., 1983; Badawy et al., 1995. and that chronic alcoholic patients and recently abstinent alcoholics have a reduced tryptophan availability ŽBranchey et al., 1981, 1984.. It is known that inflammatory responses are accompanied by decreased plasma total tryptophan concentrations, which may be caused by interferon-g ŽIFNg ., IL-1 and IL-2-stimulated induction of in-

M. Maes et al. r Psychiatry Research 78 (1998) 151]161

doleamine 2,3 dioxygenase ŽIDO., a major tryptophan catabolizing enzyme ŽTakikawa et al., 1984; Brown et al., 1989; Moroni et al., 1991; Hu et al., 1995.. The aims of the present study were to examine Ž1. the IRS through measurements of serum concentration of IL-6, the soluble IL-6 receptor ŽsIL6R., sgp130, CC16 ŽClara Cell protein., IL-1R antagonist ŽIL-1RA. and sCD14 molecule in detoxified AWLD patients; Ž2. whether detoxified patients with AWLD have a lowered availability of total tryptophan to the brain; and Ž3. whether the availability of tryptophan to the brain is inversely related to activation of the IRS. The sIL6R has the potential to mediate IL-6 signals, even in IL-6 insensitive cells, by forming a complex with IL-6, which, in turn, associates with gp130 on the responding cells ŽSaito et al., 1993; Benigni et al., 1996.. The glycoprotein gp130 is the IL-6 transducer signal protein ŽBenigni et al., 1996.. The sgp130 may compete with its membrane bound counterpart for binding of the sIL-6RrIL-6 complex ŽMurakami-Mori et al., 1996.. The CC16 is a natural immunosuppressor and anti-inflammatory secretory protein, which acts as an anti-cytokine through suppression of IL-2 related production of IFNg ŽDierynck et al., 1995.. The IL-1RA is a pure IL-1 receptor antagonist, which may inhibit the biological activity of IL-1 on the target cell, e.g. T cells and macrophages ŽDayer and Burger, 1994; Sims and Dower, 1994.. The sCD14 is a soluble antigen Žlipopolysaccharide, LPS. of monocytes and macrophages ŽNockher and Scherberich, 1995..

153

2. Subjects and methods 2.1. Subjects Twenty-four subjects participated in this study, i.e. 12 inpatients with chronic alcoholism and 12 normal volunteers. Table 1 shows the demographic data of the subjects in this study. The diagnosis of alcohol dependence Ž303.9. was made by means of the DSM-III-R criteria ŽAPA, 1987. using the Structured Clinical Interview for DSMIII-R ŽSpitzer et al., 1990.. All patients had a heavy alcohol consumption Ž) 300 ml pure alcohol daily. for at least 2 years. We excluded the following patients: Ž1. patients with liver disease, such as hepatitis or liver cirrhosis and patients who were HbsAg and HIV positive; Ž2. patients with abnormal blood liverrenzyme tests, such as aspartate aminotransferase ŽAST., alanine aminotransferase ŽALT., alkaline phosphatase ŽALP., gamma-glutamyl transpeptidase Žg GT., lactate dehydrogenase ŽLDH., amylase and creatinine kinase ŽCK.; Ž3. patients with abnormal hematological values, such as hematocrit ŽHt. and mean corpuscular volume ŽMCV.; Ž4. patients with evidence of pulmonary or urinary tract infections; Ž5. patients with a recent or past history of other axis-I diagnoses beside alcohol dependence, such as affective disorders, schizophrenic and paranoid disorders, organic mental disorders including those induced by alcohol dependence and other substance use or abuse disorders; and Ž6. patients treated with drugs known to interfere with endocrine or immune functions.

Table 1 Demographic data and plasma amino acid measurements in detoxified AWLD patients and normal control subjects ŽNC. Groups

MrF

Age Žyears.

Tryptophan

Tyrosine

Valine

Phenylalanine

Isoleucine

Leucine

TryptophanrCAA ratio = 100

NC AWLD x2 or F d.f. P

10r2 11r1 0.4 1 0.5

37.2 Ž9.4. 41.1 Ž8.9. 1.1 1r22 0.3

80 Ž21. 64 Ž6. 5.0 1r22 0.03

67 Ž10. 61 Ž10. 1.6 1r22 0.2

255 Ž56. 219 Ž39. 2.0 1r22 0.2

93 Ž17. 99 Ž16. 0.9 1r22 0.7

84 Ž19. 70 Ž13. 2.4 1r22 0.1

160 Ž36. 143 Ž26. 1.0 1r22 0.3

12.2 Ž2.4. 10.9 Ž1.1. 2.0 1r22 0.2

Notes: All results are shown as mean Ž"S.D..; amino acids are expressed in 10y6 Mrl; F: all results of analyses of variance Žthe amino acid values are processed in ln transformation..

154

M. Maes et al. r Psychiatry Research 78 (1998) 151]161

The normal volunteers were free of any medication during at least 1 month previous to blood sampling. They were non-drinkers and none had ever been taking psychotropic drugs. Exclusionary criteria for normal volunteers were a current, past or family history Žfirst-degree relatives. of psychiatric disorders, including substance abuse. Inclusion criteria for patients and normal control subjects included: Ž1. normal blood analyses, e.g. hematologic measures, such as hematocrit ŽHt. and mean corpuscular volume ŽMCV., serum electrolytes and renal function tests, such as blood urea and serum creatinine; and Ž2. absence of chronic illnesses known to affect the endocrine or immune status and of acute infectious or allergic reactions for at least 2 weeks prior to the study. 2.2. Experimental design In patients and healthy volunteers, serum for the assay of the IRS variables and plasma for the assay of tryptophan and the CAA Žvaline, leucine, isoleucine, phenylalanine and tyrosine. were sampled at 07.30 h Ž"30 min. after an overnight fast. In both patients and normal volunteers blood was collected some minutes after awakening. In normal volunteers blood was collected at their homes. In AWLD patients blood was collected in the hospital, 32]34 days after admission. The samples were frozen at y708C until thawed for assay. Immediately after admission into hospital, alcoholic patients received diazepam Ž20]40 mgr day., which was tapered off during a 2-week period. Consequently, patients were drug free for 18]20 days before blood samplings were carried out. All patients participated in an alcoholism treatment program and remained hospitalized during this study. At the time blood collections were carried out, some Ž n s 5. patients were taking disulfiram, 400 mg daily, whose treatment was started 21 days after admission into hospital. All blood specimens for the assay of immune and amino acid variables in detoxified AWLD patients and normal volunteers were assayed the same day, in a single run with a single lot number of reagents and consumables employed by a sin-

gle operator. The IL-6, sIL-6R, IL-1RA, IL-8, sgp130, CC16 and sCD14 are quantified by means of ELISA methods ŽEurogenetics, Tessenderlo, Belgium. based on appropriate and validated sets of monoclonal antibodies. The intra-assay coefficients of variation are - 8% for all assays. Plasma amino acids were determined by means of a HPLC method as described previously ŽTurnell and Cooper, 1982; Maes et al., 1990.. The intraassay CV values computed in our laboratory are: tryptophan 3.3%; tyrosine 3.8%; valine 3.0%; phenylalanine 3.2%; isoleucine 3.4%; and leucine 3.7%. The L-TRPrvalineq leucine q isoleucine q tyrosine q phenylalanine ŽCAA. ratio was computed and multiplied by 100. The dry chemistry method ŽOrtho Clinical Diagnostics. was employed to measure serum AST, ALT, ALP, g GT, LDH, amylase and CK. Hematocrit and MCV were determined by means of a Coulter STKS fully automated total blood cell counter. The analytical interassay CV values obtained in our laboratory were 0.8% for Ht and 0.6% for MCV. 2.3. Statistical analyses The independence of classification systems was ascertained by means of analysis of contingence Ž x 2-test.. Relationships between variables were ascertained by means of Pearson’s product moment and Spearman’s rank order correlation coefficients or through Žmultiple. regression analysis. Group mean differences were checked by means of analysis of variance ŽANOVA., analysis of covariance ŽANCOVA. and linear discriminant analysis ŽLDA.. The diagnostic performance of a test result was ascertained by means of receiver operating characteristics ŽROC. analysis with computation of the area under the ROC curve, sensitivity, specificity and predictive value of a positive ŽPVq. and negative ŽPVy. test result and with kappa statistics ŽZweig and Campbell, 1993.. Transformations Žln. were used to reach normality of distribution or to adjust for heterogeneity of variance between study groups. 3. Results Table 1 shows the demographic data of the

M. Maes et al. r Psychiatry Research 78 (1998) 151]161

subjects in this study. There were no significant differences in the malerfemale ratio or in age between the two study groups. There were significant correlations between age and the tryptophanrCAA ratio Ž r s y0.45, Ps 0.02.. There were no significant correlations between age and the other amino acids, e.g. plasma tryptophan: r s y0.05, P s 0.8 Žall results of Pearson’s product moment correlation, which were pooled over the two study groups.. There were no significant correlations between age and any of the immune-inflammatory variables Žnot even at the Ps 0.05 level.. Table 2 shows that plasma tryptophan was significantly lower in detoxified AWLD patients than in normal control subjects. Covarying for age in an ANCOVA did not change these results Ž F s 4.7, d.f.s 1r21, Ps 0.04.. All detoxified AWLD patients had plasma tryptophan values lower than the mean of the normal volunteers. The area under the ROC curve was 77.8% for plasma tryptophan. At the maximal threshold value the diagnostic performance was: sensitivity 83.3%, specificity 75.0%, PVq 76.9% and PVy 81.8% Ž k s 0.58, t s 3.5, Ps 0.002.. There was a trend toward lower plasma tyrosine, valine, isoleucine and leucine in detoxified AWLD patients than in normal volunteers, but the differences did not reach significance. There were no significant differences in the tryptophanrCAA ratio between both groups. Table 2 shows the measurements of serum IL-6, sIL-6R, sgp130, CC16, IL-1RA, IL-8 and sCD14 in detoxified AWLD patients. There was a trend toward higher serum IL-6 in detoxified AWLD

155

patients than in normal control subjects. Serum IL-8 and IL-1RA were significantly higher and CC16 significantly lower in detoxified AWLD patients than in normal volunteers. Covarying for age in ANCOVAs did not change these results, i.e. IL-8 Ž F s 11.3, d.f.s 1r21, Ps 0.003., IL1RA Ž F s 7.6, d.f.s 1r21, Ps 0.01. and CC16 Ž F s 7.2, d.f.s 1r21, Ps 0.01.. The areas under the ROC curve for the immune-inflammatory variables were: CC16 81.0%; IL-1RA 80.2%; and IL-8 84.4%. At the optimal IL-8 threshold value Ži.e.G 150 pgrml., the diagnostic performance was: sensitivity 58.3%, specificity 83.3% and PVq 78.0% Ž k s 0.42, t s 2.2, Ps 0.03.. When serum IL-6, CC16, IL-1RA and IL-8 were used as discriminatory variables in an LDA, a significant separation of detoxified AWLD patients from normal volunteers was obtained Ž F s 19.8, d.f.s 1r22, Ps 0.0004.. The LDA score showed an area under the ROC curve of 92.0%. At the maximal threshold value, the diagnostic performance of this LDA score was: sensitivity 83.3%, specificity 91.7% and PVq 91.0% Ž k s 0.75, t s 5.5, Ps 0.00007.. In both study groups combined, Pearson’s product moment correlation analyses showed significant relationships between plasma tryptophan and IL-8 Ž r s y0.60, Ps 0.002. and serum CC16 Ž r s 0.58, Ps 0.003., whereas there was a trend toward significant correlations between plasma tryptophan and serum IL-6 Ž r s y0.34, Ps 0.09. and IL-1RA Ž r s y0.37, Ps 0.07.. In both study groups combined, Pearson’s product moment correlation analyses showed significant inverse correlations between the tryptophanrCAA ratio and

Table 2 Measurements of serum concentrations of immune-inflammatory markers in detoxified AWLD patients and normal control subjects ŽNC. Groups

IL-6 Žpgrml.

sIL-6R Žngrml.

sgp130 Žngrml.

CC16 Žngrml.

IL-1RA Žngrml.

IL-8 Žpgrml.

CD14 ŽUrml.

NC AWLD F d.f. P

4.1 Ž5.1. 16.1 Ž25.4. 3.4 1r22 0.07

182 Ž65. 196 Ž50. 0.6 1r22 0.5

1078 Ž138. 1147 Ž184. 1.0 1r22 0.7

42 Ž16. 25 Ž8. 9.3 1r22 0.006

0.22 Ž0.09. 0.35 Ž0.16. 9.1 1r22 0.006

70 Ž111. 300 Ž302. 11.6 1r22 0.002

496 Ž77. 486 Ž90. 0.1 1r22 0.7

Notes: All results are shown as mean Ž"S.D..; F: all results of ANOVAs performed on the ln-transformed data.

156

M. Maes et al. r Psychiatry Research 78 (1998) 151]161

IL-8 Ž r s y0.42, Ps 0.04., IL-6 Ž r s y0.44, Ps 0.03. and IL-1RA Ž r s y0.50, P s 0.01.. In healthy volunteers, but not in AWLD patients, there were significant Spearman rank order correlations between tryptophan and IL-8 Ž r s y0.83, Ps 0.001., IL-1RA Ž r s y0.63, Ps 0.02. and CC16 Ž r s 0.63, Ps 0.02. and between the tryptophanrCAA ratio and IL-8 Ž F s y0.63, Ps 0.02. and IL-1RA Ž r s y0.83, Ps 0.001.. Since some of the above group differences or significant relationships could be related to the use of disulfiram and a possibly poorer health status of the AWLD patients or long-lasting effects of alcohol abuse, we have controlled our results for possible effects of use of disulfiram and for differences in g GT and MCV values. We did not control our results for effects of possible differences in Ht, since changes in Ht, but not MCV, may be a consequence of IRS activation, through increased secretion of IL-6 ŽMaes et al., 1996a.. There were no significant differences in MCV Žmeans s 89.9" 4.3 fl vs. 91.3" 5.7 fl; F s 0.4, d.f.s 1r20, Ps 0.5., Ht Žmeans s 45.2" 3.5% vs. 43.1" 2.1%; F s 3.1, d.f.s 1r20, Ps 0.09. and g GT Žmeans s 35.2" 15.2 Urml vs. 29.7 " 6.2 Url; F s 1.3, d.f.s 1r22, P s 0.2. between normal volunteers and AWLD patients, respectively. There were no significant differences in serum IL-6 Ž F s 1.1, Ps 0.3., sIL-6R Ž F s 2.2, Ps 0.2., sgp130 Ž F s 0.1, Ps 0.7., CC16 Ž F s 0.8, Ps 0.6., IL-1RA Ž F s 0.7, Ps 0.6., IL-8 Ž F s 0.9, Ps 0.6. or sCD14 Ž F s 0.01; Ps 0.9; all d.f.s 1r10. between detoxified AWLD patients, who were or were not treated with disulfiram. There were no significant differences in plasma tryptophan Ž F s 0.1, Ps 0.7., tyrosine Ž F s 3.6, Ps 0.08., valine Ž F s 2.4, P s 0.1., phenylalanine Ž F s 0.5, Ps 0.5., leucine Ž F s 0.01, P s 0.9., iso le u cin e Ž F s 1 .5 , P s 0 .2 . a n d th e tryptophanrCAA ratio Ž F s 2.6, P s 0.1; all d.f.s 1r10. between detoxified AWLD patients who were or were not treated with disulfiram. There were no significant correlations between AST, ALT, g GT, or ALP and any of the plasma amino acids or serum IRS variables, except for sIL-6R and sgp130. There were significant and positive correlations between sgp130 and AST Ž r s 0.69, Ps 0.02. and ALT Ž r s 0.68, Ps 0.02. and between serum sIL-6R and ALT Ž r s 0.66,

Ps 0.03.. There was a significant and inverse correlation between MCV and plasma tryptophan Ž r s y0.51, Ps 0.01., but not between MCV and the tryptophanrCAA ratio Ž r s y0.20, Ps 0.6.. After adjustment for possible effects of MCV and g GT Žforced entry as covariates in ANCOVAs., the differences between healthy volunteers and AWLD patients in tryptophan Ž F s 7.0, d.f.s 1 r 1 7 , P s 0 .0 1 . , C C 1 6 Ž F s 6.1, d.f.s 1r17, Ps 0.02., IL-1RA Ž F s 5.6, d.f.s 1r17, Ps 0.03. and IL-8 Ž F s 8.0, d.f.s 1r17, Ps 0.01. remained significant. After adjustment for MCV and g GT Žforced entry as explanatory variables in regression analyses ., the relationships between tryptophan and IL-8 Ž F s 9.8, d.f.s 1r18, Ps 0.006. and CC16 Ž F s 6.3, d.f.s 1r18, P s 0.02. and between the tryptophanrCAA ratio and IL-8 Ž F s 5.2, d.f.s 1r18, Ps 0.03., IL-6 Ž F s 5.2, d.f.s 1r18, Ps 0.03. and IL-1RA Ž F s 10.7, d.f.s 1r18, P s 0.004. remained significant. 4. Discussion The first major finding of this study is that detoxified AWLD patients show a significant activation of the IRS. To the best of our knowledge, this is the first study to show that serum CC16 is significantly lower in detoxified AWLD patients than in normal control subjects. CC16 has immunosuppressive, anti-inflammatory and anti-cytokine activities with inhibitory effects on phospholipase A2 activity and on the IL-2-related production of IFNg ŽMiele et al., 1987; Mantile et al., 1993; Peri et al., 1993; Dierynck et al., 1995.. Therefore it may be hypothesized that lower serum levels of this endogenous anti-cytokine may increase the risk for inflammatory or infectious responses in detoxified AWLD patients. Detoxified AWLD patients also showed significantly increased serum IL-8 concentrations and a trend toward higher IL-6 concentrations. IL-8 is produced by IL-1- and TNFa-induced monocytes, macrophages, neutrophils, endothelial cells, fibroblasts and hepatocytes ŽCavaillon, 1996.. IL-8 exerts a chemotaxic activity not only on neutrophils, but also on eosinophils, basophils, natural killer cells and lymphocytes ŽCavaillon, 1996.. Thus, increased serum IL-8 in detoxified AWLD

M. Maes et al. r Psychiatry Research 78 (1998) 151]161

patients is another indicant of activation of the IRS. These findings extend those of Cook et al. Ž1991., who found signs of T cell activation in newly detoxified alcoholic patients. Martinez et al. Ž1993., on the other hand, were unable to detect significant alterations in mitogen-induced production of IL-6 and IL-8 by peripheral blood mononuclear cells ŽPBMC. of chronic alcoholic patients 30 days after abstinence. However, in vitro assays of cytokines after mitogen stimulation of PBMC, probably, do not reflect the in vivo situation as can be evaluated by direct measurements of serum cytokine concentration or LPS q PHA-stimulated production of cytokines by whole blood ŽDe Groote et al., 1992.. In detoxified AWLD patients, no significant alterations in sIL6R and sgp130 could be found. This is also the first study to find that serum IL-1RA is significantly higher in detoxified AWLD patients than in non-alcoholic control subjects. IL-1RA is mainly derived form monocytes and is secreted, following activation along with IL-1 ŽDayer and Burger, 1994.. Thus, increased IL-8 IL-6 and IL-1RA levels support the thesis that detoxification in AWLD patients is accompanied by an activation of the monocytic arm of cellmediated immunity. It is thought that IL-1RA antagonizes the biological activities of IL-1, e.g. on brain 5-HT and catecholamine turnover ŽDinarello, 1994., through binding to IL-1 receptors ŽDayer and Burger, 1994; Sims and Dower, 1994.. Thus, raised IL-1RA production in detoxified AWLD patients may be one mechanism to protect the body from the detrimental effects of exaggerated production of proinflammatory cytokines. Detoxified AWLD patients show lower plasma tryptophan, without significant changes in the tryptophanrCAA ratio, than normal volunteers. These findings extend those of Branchey et al. Ž1984., who reported lower tryptophan availability in recently abstinent alcoholics. Studies in rodents, baboons and humans show that acute administration of alcohol results in reductions in plasma tryptophan ŽBranchey et al., 1981; Branchey and Lieber, 1982; Eriksson et al., 1983; Badawy et al., 1995.. Other authors, however, could not find any effects of acute alcohol treatment on plasma tryptophan ŽSiegal et al., 1964;

157

Pietraszek et al., 1991.. Also, patients with chronic alcoholism show lower plasma tryptophan and a lower tryptophanrCAA ratio ŽBranchey et al., 1981.. These effects of alcohol are ascribed to enhanced activity of liver tryptophan pyrrolase ŽBuydens-Branchey et al., 1988; Badawy et al., 1995., the rate limiting enzyme for the catabolism of tryptophan along the kynurenine pathway ŽWolf, 1974.. Other authors found reduced liver tryptophan pyrrolase activity during chronic alcohol administration ŽMorland, 1974a,b.. Accordingly, Friedman et al. Ž1988. observed a reduced conversion of tryptophan to kynurenine within 1 week following detoxification, which suggests reduced liver pyrrolase activity. Thus, there is no firm evidence that induction of pyrrolase activity by chronic abuse of alcohol may play a role in lower plasma tryptophan in detoxified AWLD patients. A third major finding of this study is that there were significant and inverse relationships between the availability of plasma tryptophan to the brain Žeither plasma tryptophan or the tryptophanr CAA ratio. and inflammatory markers, such as serum IL-6, IL-8 and IL-1RA. Moreover, a significant positive relationship between the availability of tryptophan to the brain and serum CC16 was found. These results may suggest that the lowered availability of plasma tryptophan in detoxified AWLD patients reflects, at least in part, the inflammatory response in that condition. There were also significant relationships between signs of IRS activation and lowered availability of plasma tryptophan in healthy volunteers, but not in AWLD patients. The negative results in the subgroup of AWLD patients may be explained by the major upheavals in IRS function in AWLD patients, which could have blurred any relationships between the IRS and plasma tryptophan. Reduced plasma tryptophan in patients showing IRS activation may be caused by Ž1. induction of IDO by proinflammatory monocytic and Th1-like cytokines ŽHu et al., 1995.; and Ž2. increased use of peripheral amino acids for leukocyte activity and synthesis of acute phase proteins in the liver ŽMoldawer et al., 1987.. Similar findings have been reported to occur in major depression. That disorder is characterized by a significantly re-

158

M. Maes et al. r Psychiatry Research 78 (1998) 151]161

duced availability of tryptophan to the brain, IRS activation with increased production of proinflammatory cytokines and significant relationships between plasma tryptophan or the tryptophanr CAA ratio and immune or IRS markers, such as serum IL-6, neopterin, haptoglobin and the CD4 q rCD8q T cell ratio Žall negative. and serum zinc and transferrin Žall positive. ŽMaes et al., 1993, 1994, 1996b, 1997.. In the present study we have controlled for or otherwise may disregard the possible effects of intervening variables, such as use of disulfiram, age and effects of liver pathologies. First, no significant effects of disulfiram on plasma tryptophan availability or IRS variables could be found. Previously, it has been reported that disulfiram had no significant effects on the CSF concentrations of 5-hydroxyindoleacetic acid, the major metabolite of 5-HT ŽBeck et al., 1980.. Second, it should be underscored that patients were selected to participate in this study if no apparent liver disease was present and if routine blood tests, such as serum g GT, MCV and Ht, were within the normal ranges. In AWLD patients, no significant correlations between IL-6, IL-8, CC16 or IL-1RA and serum liver enzyme concentrations could be found. Third, all intergroup differences reported herein and the significant relationships between IRS variables and plasma tryptophan availability remained significant after controlling for possible effects of differences in MCV and g GT. Thus, the changes in the IRS in detoxified AWLD patients do not seem to result from a liver pathology. Therefore the mechanisms by which chronic use of alcohol may induce IRS activation several weeks after detoxification have remained elusive. Damage to tissues other than the liver cannot be excluded as a cause for IRS activation in detoxified AWLD patients. In any case, tissue damage cannot explain the reduced serum CC16 concentrations in detoxified AWLD patients. Fourth, it could be argued that possible changes in the diet of AWLD patients could have affected plasma tryptophan concentrations. However, the dietary impact on the availability of plasma tryptophan to the brain is rather small ŽCurzon, 1990.: Ž1. physiologically relevant differences in protein intake do not induce alter-

ations in the tryptophanrCAA ratio ŽFernstrom et al., 1987.; Ž2. plasma CAA concentrations are unaffected by breakfast in freely feeding adults ŽMilson et al., 1979; Scriver et al., 1985.; Ž3. the tryptophanrCAA ratio is not significantly altered in subjects given meals of considerably different composition ŽAshley et al., 1982, 1985.; and Ž4. fasting for up to 5 days had little effect on plasma tryptophan concentrations in the rat ŽCurzon, 1990.. Acknowledgements The research reported was supported in part by the Staglin Investigator Award ŽNARSAD. to Dr M. Maes; and the Clinical Research Center for Mental Health, Antwerp, Belgium. The secretarial assistance of Mrs. M. Maes is greatly appreciated. References APA, 1987. DSM-III-R: Diagnostic and Statistical Manual of Mental Disorders. American Psychiatric Association, Washington, DC. Ashley, D.V.M., Barclay, D.V., Chauffard, F.A., 1982. Plasma amino acid responses in humans to evening meals of differing nutritional composition. American Journal of Clinical Nutrition 36, 143]153. Ashley, D.V.M., Liardon, R., Leathwood, P.D., 1985. Breakfast meal composition influences plasma tryptophan to large neutral amino acid ratios of healthy lean young men. Journal of Neural Transmission 63, 271]283. Badawy, A.A.-B., Morgan, C.J., Lovett, J.W., Bradley, D.M., Thomas, R., 1995. Decrease in circulating tryptophan availability to the brain after acute ethanol consumption by normal volunteers: implications for alcohol-induced aggressive behavior and depression. Pharmacopsychiatry 28, 93]97. Baker, R.C., Jerrels, T.R., 1993. Recent developments in alcoholism: immunological aspects. Recent Advances in Alcoholism 11, 249]271. Beck, O., Borg, S., Holmstedt, B., Kvande, H., Skroder, R., 1980. Concentration of serotonin metabolites in the cerebrospinal fluid from alcoholics before and during disulfiram therapy. Acta Pharmacologica et Toxicologica 47, 305]307. Benigni, F., Fantuzzi, G., Sacco, S., Sironi, M., Pozzi, P., Dinarello, C.A., Sipe, J.D., Poli, V., Cappelletti, M., 1996. Six different cytokines that share GP 130 as receptor subunit. Blood 87, 1851]1854. Branchey, L., Lieber, C.S., 1982. Activation of tryptophan pyrrolase after chronic alcohol administration. Substance and Alcohol Actions of Misuse 2, 225]229. Branchey, L., Branchey, M., Shaw, S., Lieber, C.S., 1984.

M. Maes et al. r Psychiatry Research 78 (1998) 151]161 Depression, suicide and aggression in alcoholics and their relationship to plasma amino acids. Psychiatry Research 12, 219]226. Branchey, L., Shaw, S., Lieber, C.S., 1981. Ethanol impairs tryptophan transport into the brain and depresses serotonin. Life Sciences 29, 2751]2755. Brown, R.R., Lee, C.M., Kohler, P.C., Hank, J.A., Storer, B.E., Sondel, P.M., 1989. Altered tryptophan and neopterin metabolism in cancer patients treated with recombinant interleukin-2. Cancer Research 49, 4941]4944. Buydens-Branchey, L., Branchey, M., Worner, T.M., Zucker, D., Aramsombatdee, E., Lieber, C.S., 1988. Increase in tryptophan oxygenase activity in alcoholic patients. Alcoholism: Clinical and Experimental Research 12, 163]167. Cavaillon, J.-M., 1996. Les Cytokines, Masson, Paris. Cook, R.T., Garvey, M.J., Booth, B.M., Goeken, J.A., Stewart, B., Noel, M., 1991. Activated CD-8 cells and HLA DR expression in alcoholics without overt liver disease. Journal of Clinical Immunology 11, 246]253. Curzon, G., 1990. Availability of amino acids to the brain and implication for transmitter amine function. In: Richardson, M.A. ŽEd.., Amino Acids in Psychiatric Disease. American Psychiatric Press, Washington, DC, pp. 33-48. Curzon, G., Sarna, G.S., 1984. Tryptophan transport to the brain: newer findings and older ones reconsidered. In: Schlossberger, H.G., Kochen, W., Linzen, B., Steinhart, H. ŽEds.., Progress in Tryptophan and Serotonin Research. Walter De Gruyter, Berlin]New York, pp. 145]157. Dayer, J.M., Burger, D., 1994. Interleukin-1, tumour necrosis factor and their specific inhibitors. European Cytokine Network 5, 563]571. De Groote, D., Zangerle, P.F., Gevaert, Y., Fassotte, M.F., Beguin, Y., Noizat-Pirenne, F., Pirenne, J., Gathy, R., Lopez, M., Dehart, I., Igot, D., Baudrihaye, M., Delacroix, D., Franchimont, P., 1992. Direct stimulation of cytokines ŽIL-1 b , TNF-a , IL-6, IL-2, IFN-g and GM-CSF. in whole blood. I. Comparison with isolated PBMC stimulation. Cytokine 4, 239]248. Dierynck, I., Bernard, A., Roels, H., De Ley, M., 1995. Potent inhibition of both human interferon-gamma production and biologic activity by the Clara cell protein CC16. American Journal of Respiratory Cell and Molecular Biology 12, 205]210. Dinarello, C.A., 1994. The biological properties of interleukin1. European Cytokine Network 5, 517]531. Eriksson, T., Magnusson, T., Carlson, A., Hagman, M., Jagenburg, R., 1983. Decrease in plasma amino acids in man after an acute dose of ethanol. Journal of Studies on Alcohol 44, 215]221. Fernstrom, J.D., 1984. Tryptophan availability and serotonin synthesis in rat brain: effects of experimental diabetes. In: Schlossberger, H.G., Kochen, W., Linzen, B., Steinhart, H. ŽEds.., Progress in Tryptophan and Serotonin Research. Walter de Gruyter, Berlin]New York, pp. 161]172. Fernstrom, J.D., Fernstrom, M.H., Grubb, P.E., 1987. Twenty-four hour variations in rat blood and brain levels of

159

the aromatic and branched-chain amino acids: chronic effects of dietary protein content. Metabolism 36, 643]650. Friedman, M.J., Krstulovic, A.M., Severinghaus, J.M., Brown, S.J., 1988. Altered conversion of tryptophan to kynurenine in newly abstinent alcoholics. Biological Psychiatry 23, 89]93. Grossman, C.J., Mendenhall, C.L., Roselle, G.A., 1988. Alcohol and immune regulation. In vivo effects of ethanol on concanavalin A sensitive thymic lymphocyte function. International Journal of Immunopharmacology 10, 187]195. Hu, B., Hissong, B.D., Carlin, J.M., 1995. Interleukin-1 enhances indoleamine 2,3-dioxygenase activity by increasing specific mRNA expression in human mononuclear phagocytes. Journal of Interferon and Cytokine Research 15, 617]624. Jerrels, T.R., Perritt, D., Eckardt, M.J., Marietta, C., 1990. Alterations in interleukin-2 utilization by T cells from rats treated with an ethanol-containing diet. Alcoholism: Clinical and Experimental Research 14, 245]249. Kaplan, D.R., 1986. A novel mechanism of immunosuppression mediated by ethanol. Cellular Immunology 102, 1]9. Khoruts, A., Stahnke, L., McClain, C.J., Logan, G., Allen, J.I., 1991. Circulating tumor necrosis factor, interleukin-1 and interleukin-6 concentrations in chronic alcoholic patients. Hepatology 13, 267]276. Maes, M., Meltzer, H.Y., Scharpe, ´ S., Bosmans, E., Suy, E., Minner, B., De Meester, I., Calabrese, J., Vandervorst, C., Raus, J., Cosyns, P., 1993. Relationships between lower plasma L-tryptophan levels and immune variables in depression. Psychiatry Research 49, 11]27. Maes, M., Scharpe, ´ S., Meltzer, H.Y., Okayli, G., Bosmans, E., D’Hondt, P., Vanden Bossche, B., Cosyns, P., 1994. Increased neopterin and interferon g secretion and lower availability of L-tryptophan in major depression: further evidence for activation of cell-mediated immunity. Psychiatry Research 54, 143]160. Maes, M., Schotte, C., Scharpe, ´ S., Martin, M., Blockx, P., 1990. The effects of glucocorticoids on the availability of L-tryptophan and tyrosine in the plasma of depressed patients. Journal of Affective Disorders 18, 121]127. Maes, M., Vandevyvere, J., Vandoolaeghe, E., Bril, T., Vandoolaeghe, E., Neels, H., 1996a. Disorders in iron metabolism and the erythron in major depression: further evidence for an inflammatory response. Journal of Affective Disorders 40, 23]33. Maes, M., Verkerk, R., Vandoolaeghe, E., Van Hunsel, F., Neels, H., Wauters, A., Demedts, P., Scharpe, ´ S., 1997. Serotonin-immune interactions in major depression: lower serum tryptophan as a marker of an immune-inflammatory response. European Archives of Psychiatry and Clinical Neuroscience 247, 154]161. Maes, M., Wauters, A., Verkerk, R., Neels, H., VanGastel, A., Cosyns, P., Scharpe, ´ S., Desnyder, R., 1996b. Lower L-tryptophan availability in depression: a marker of a more generalized disorder in protein metabolism. Neuropsychopharmacology 15, 243]251. Mantile, G., Miele, L., Cordella-Miele, E., Singh, G., Katyal,

160

M. Maes et al. r Psychiatry Research 78 (1998) 151]161

L., Mukherjee, A.B., 1993. Human Clara cell 10-kDa protein is the counterpart of rabbit uteroglobin. Journal of Biological Chemistry 268, 20343]20351. Martinez, F., Abril, E.R., Earnest, D.L., Watson, R.R., 1992. Ethanol and cytokine secretion. Alcohol 9, 455]458. Martinez, F., Thomas, N.M., Darban, H., Cox, T.J., Wood, S., Watson, R.R., 1993. Interleukin-6 and interleukin-8 production by mononuclear cells of chronic alcoholics during treatment. Alcoholism: Clinical and Experimental Research 17, 1193]1197. Miele, L., Cordella-Miele, E., Mukherjee, A.B., 1987. Uteroglobin: structure, molecular biology and new perspectives on its function as a phospholipase A2 inhibitor. Endocrine Reviews 8, 474]490. Milson, J.P., Morgan, M.Y., Sherlock, S., 1979. Factors affecting plasma amino acid concentrations in control subjects. Metabolism 28, 313]319. Moir, A.T.B., Eccleston, D., 1968. The effect of precursor loading in the cerebral metabolism of 5-hydroxyindoles. Journal of Neurochemistry 15, 1093]1108. Moldawer, L.L., Georgieff, M., Lundholm, K., 1987. Interleukin-1, tumour necrosis factor-alpha Žcachectin . and the pathogenesis of cancer cachexia. Clinical Physiology 7, 263]274. Morland, J., 1974a. Hepatic tryptophan oxygenase activity as a marker of changes in protein metabolism during chronic ethanol treatment. Acta Pharmacologica et Toxicologica 35, 155]159. Morland, J., 1974b. Effect of chronic ethanol treatment on tryptophan oxygenase, tyrosine aminotransferase and general protein metabolism in the intact and perfused rat liver. Biochemical Pharmacology 23, 21]267. Moroni, F., Russi, P., Gallo-Mezo, M.A., Moneti, G., Pellicciari, R., 1991. Modulation of quinolinic and kynurenic acid content in the rat brain: effects of endotoxins and nicotinylalanine. Journal of Neurochemistry 57, 1630]1635. Murakami-Mori, K., Taga, T., Kishimoto, T., Nakamura, S., 1996. The soluble form of the IL-6 receptor ŽsIL-6R a . is a potent growth factor for AIDS-associated Kaposi’s sarcoma ŽKS. cells; the soluble form of gp130 is antagonistic for IL-6R a-induced AIDS-KS cell growth. International Immunology 8, 595]602. Nelson, S., Bagby, G., Andersen, J., Nakamura, C., Shellito, J., Summer, W., 1991. The effects of ethanol, TNF and G-CSF on lung antibacterial defences. In: Friedman, H. ŽEd.., Drugs of Abuse, Immunity and Immunodeficiency. Plenum, New York, pp. 24]36. Nockher, W.A., Scherberich, J.E., 1995. Monocyte cell-surface CD14 expression and soluble CD14 antigen in hemodialysis: evidence for chronic exposure to LPS. Kidney International 48, 1467]1469. Pardridge, W.M., 1979. Tryptophan transport through the blood-brain barrier: in vivo measurement of free and albumin-bound amino acid. Life Sciences 25, 1519]1528. Peri, A., Cordella-Miele, E., Miele, L., Mukherjee, A.B., 1993. Tissue specific expression of the gene coding for human

Clara cell 10-kDa protein, a phospholipase A2-inhibitory protein. Journal of Clinical Investigation 92, 2099]2109. Pietraszek, M.H., Urano, T., Sumioshi, K., Serizawa, K., Takahashi, S., Takada, Y., Takada, A., 1991. Alcohol-induced depression: involvement of serotonin. Alcohol and Alcoholism 26, 155]159. Rodriguez-Rodriguez, E., Gonzales-Reimers, E., SantolariaFernandez, F., Milena-Abril, A., Rodriguez-Moreno, F., Oramas-Rodriguez, J., Martinez-Riera, A., 1995. Cytokine levels in acute alcoholic hepatitis: a sequential study. Drug and Alcohol Dependence 39, 23]27. Roselle, G.A., Mendenhall, C.L., 1982. Alteration of in vitro lymphocyte function by ethanol, acetaldehyde and acetate. Journal of Clinical and Laboratory Immunology 9, 33]37. Roselle, G.A., Mendenhall, C.L., Grossman, C.J., Weesner, R.E., 1988. Lymphocyte subset alterations in patients with alcoholic hepatitis. Journal of Clinical and Laboratory Immunology 26, 169]173. Ruiz, A.D., Perez, J.L.S., Martinez, G.L., Calvin, J.G., Extremera, B.G., Gea, F.G., 1993. Tumour necrosis factor, interleukin-1 and interleukin-6 in alcoholic cirrhosis. Alcohol and Alcoholism 28, 319]323. Saad, A.J., Domiati-Saad, R., Jerrels, T., 1993. Ethanol ingestion increases susceptibility of mice to Listeria monocytogenes. Alcoholism: Clinical and Experimental Research 17, 75]85. Saito, T., Taga, T., Miki, D., Futatsugi, K., Yawata, H., Kishimoto, T., Yasukawa, K., 1993. Preparation of monoclonal antibodies against the IL-6 signal transducer, gp130, that can inhibit IL-6-mediated function. Journal of Immunological Methods 163, 217]223. Scriver, C.R., Gregory, D.M., Sovetts, D., 1985. Normal plasma free amino acid values in adults: the influence of some common physiological variables. Metabolism 34, 868]873. Siegal, F.L., Roach, M.K., Pomeroy, L.R., 1964. Plasma amino acid patterns in alcoholism: the effects of ethanol loading. Proceedings of the National Academy of Sciences of the United States of America 51, 605-611. Sims, J.E., Dower, S.K., 1994. Interleukin-1 receptors. European Cytokine Network 5, 539]546. Spitzer, R.L., Williams, J.B.W., Gibbon, M., First, M.B., 1990. Structured Clinical Interview for DSM-III-R. American Psychiatric Press, Washington, DC. Szabo, G., Mandrekar, P., Catalano, D., 1995. Inhibition of superantigen-induced T cell proliferation and monocyte IL-1 b , TNF-a and IL-6 production by acute ethanol treatment. Journal of Leukocyte Biology 58, 342]350. Takikawa, O., Yoshida, R., Yasui, H., Hayaishi, O., 1984. The relationship between plasma kynurenine and indoleamine 2,3-dioxygenase activity in the extrahepatic tissues. In: Schlossberger, H.G., Kochen, W., Linzen, B., Steinhart, H. ŽEds.., Progress in Tryptophan and Serotonin Research. Walter de Gruyter, Berlin]New York, pp. 517]520. Turnell, D., Cooper, J., 1982. Rapid assay for amino acids in serum or urine by precolumn derivatisation and reversedphase liquid chromatography. Clinical Chemistry 28, 527]531.

M. Maes et al. r Psychiatry Research 78 (1998) 151]161 Verma, B.K., Fogarasi, M., Szabo, G., 1993. Down-regulation of tumor necrosis factor a activity by acute ethanol treatment in human peripheral blood monocytes. Journal of Clinical Immunology 13, 8]22. Wolf, H., 1974. Studies on tryptophan metabolism in man. Scandinavian Journal of Clinical and Laboratory Investigation 33 ŽSuppl..

161

Yuwiler, A., Oldendorf, W.H., Geller, E., et al., 1977. The effect of albumin binding and amino acid composition on tryptophan uptake into the brain. Journal of Neurochemistry 28, 1015]1023. Zweig, M.H., Campbell, G., 1993. Receiver operating characteristic ŽROC. plots: a fundamental evaluation tool in clinical medicine. Clinical Chemistry 39, 561]577.