Increased Cerebrospinal Fluid Levels of Neurotrophin 3 (NT-3)

Home , Ventriculitis

Molecular Psychiatry (2000) 5, 510–513  2000 Macmillan Publishers Ltd All rights reserved 1359-4184/00 $15.00 www.nature.com/mp ORIGINAL RESEARCH ARTICLE Increased cerebrospinal ﬂuid levels of neurotrophin 3 (NT-3) in elderly patients with major depression C Hock1, K Heese2,3,FMu¨ller-Spahn4, P Huber5, W Riesen6, RM Nitsch1 and U Otten2

1Department of Psychiatry Research, University of Zu¨rich; Departments of 2Physiology; 4Psychiatry; 5Clinical Chemistry, University of Basel, Switzerland; 3BF Research Institute, National Cardiovascular Center, Osaka, Japan; 6Institute for Clinical Chemistry and Hematology, State Hospital St Gallen, St Gallen, Switzerland

Neurotrophin 3 (NT-3) is a member of the neurotrophin gene family which supports the survival of specific neurons. NT-3 was shown to prevent the death of adult central noradrenergic neurons in vivo, a neuronal population which is associated with the pathophysiology of major depression. We quantitated CSF levels of NT-3 in elderly patients with major depression (DE) and compared them to patients with Alzheimer’s disease (AD), and mentally healthy control subjects (CTR). CSF levels of NT-3 were markedly and significantly elevated in the DE group, as compared to either the AD or the CTR group (P Ͻ 0.01, and P Ͻ 0.001, respectively). In terms of diagnostic accuracy, measurement of NT-3 levels in DE resulted in 73.9% sensitivity, and 89.7% specificity. Increased CSF levels of NT-3 may indicate a disturbance of the central noradrenergic system in patients with DE. NT-3 may constitute a biochemical candidate marker for clinical diagnosis and for the evaluation of therapeutic strategies in DE. Molecular Psychiatry (2000) 5, 510–513. Keywords: noradrenaline; basal forebrain; neurotrophins; ELISA; aging

Introduction that CSF levels of NT-3 are increased in patients with DE as compared to the AD and CTR groups. Nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin 3 (NT-3), and neurotrophin 4/5 (NT-4/5) are members of the neurotrophin Subjects and methods gene family which supports the survival, differen- Patients with major depression (DE) were diagnosed tiation, maintenance, and repair of vertebrate neurons.1 according to the ICD-10 (F32.0x/1x, F33.0x/1x) and NGF supports the cholinergic neurons of the basal fore- DSM-III-R (296.20–22, 296.30–32) criteria. Diagnosis of brain system that are affected by neuronal loss in Alzh- probable AD was made according to criteria of the eimer’s disease (AD).2 BDNF supports cholinergic, National Institute of Neurological and Communicative dopaminergic and 5-hydroxytryptamine (5-HT) con- Disorders and Stroke–Alzheimer’s disease and Related taining neurons.3–5 NT-3 was shown to prevent the Disorders Association (NINCDS–ADRDA).11 All death of adult central noradrenergic neurons in vivo,6 patients were referred to the research ward from gen- a neuronal population which is associated with the eral practitioners, neurologists and psychiatrists for pathophysiology of major depression (DE).7 In diagnostic purposes and screening for clinical trials. addition, NT-3 reportedly promotes survival of ventral None of the patients was institutionalized. The group mesencephalic dopaminergic neurons,8 cerebellar of healthy control subjects (CTR) consisted of patients granule neurons and Purkinje cells,9 and acts on sen- who underwent lumbar puncture for orthopedic or sory or sympathetic neurons of the dorsal root, nodose neurologic diagnostic purposes and were shown to and sympathetic ganglia.10 Dysfunction of neuro- have normal CSF cell counts, total protein levels, and trophic systems in the adult mammalian brain may be absence of signs of blood–brain barrier dysfunctions or reflected by alterations of neurotrophin levels in the cerebral IgG synthesis, as well as absence of psychiatric cerebrospinal fluid (CSF), and may thus be of diagnos- or central neurological disorders. tic value. Therefore, we quantitated CSF levels of NT- DE, AD and CTR patients were carefully examined 3 in elderly patients with DE and compared them to and received a thorough clinical examination. Psycho- mentally healthy control subjects (CTR), and to metric testing included the Mini Mental State (MMS),12 patients with Alzheimer’s disease (AD). We show here as a global screening instrument for dementia, and the Nurses’ Observation Scale for Geriatric Patients (NOSGER)13 as a functional measure of dementia sever- Correspondence: Dr C Hock, Department of Psychiatry Research, University of Zu¨ rich, Lenggstrasse 31, CH-8029 Zu¨ rich, Switzer- ity. The patients with DE showed no cognitive disturb- land. E-mail: chockȰbli.unizh.ch ances in the clinical examinations and the Mini Mental Received 25 October 1999; revised and accepted 4 February 2000 State scores were within the normal range. Severity of Increased CSF levels of NT-3 in depression C Hock et al 511 depression was rated by using the Montgomery Asberg ApoE genotyping was by INNO-LiPA Apo E, Inno- Depression Rating Scale (MADRS).14 Apolipoprotein genetics, Belgium; ApoE phenotyping according to (ApoE) genotyping, or, if DNA was not available, ApoE McDowell et al.15 The use of the ApoE phenotype syn- phenotyping was included in the laboratory screening onymous with the ApoE genotype in the statistical in the DE and AD patients. analyses seemed to be appropriate, since ApoE geno- CSF was obtained for diagnostic purposes in the DE typing compared with protein phenotyping showed and AD patients in which no lumbar puncture had conflicting results in less than 2%.16 been previously done during the routine diagnostic work-up. Different CSF volumes were available for the Statistical analyses analysis of the neutrophin proteins. This fact explains Statistical analyses of data were performed using the the different sample sizes for the individual measure- Mann-Whitney U test as well as independent samples ments. All available CSF samples were used for the t-tests for group comparisons. Correlation analyses analyses. were performed by multiple regression using CSF lev- NT-3 measurements. One hundred and twenty-five els of neurotrophins, as well as ApoE genotype (or spinal fluid samples were examined. DE group: n = 23, phenotype, respectively), age, duration of the disease 8 men, 15 women, mean age 70.5 ± 11.9 SD yrs, range in AD, MMS, NOSGER and MADRS scores. Regression 47–86 yrs, MMS score: mean 27.2 ± 2.5 SD. AD group: analysis was complemented with analysis of variance n = 39, 20 men, 19 women, mean age 67.2 ± 11.5 SD (ANOVA) by using SPSS for Windows (version 8.0). yrs, range 39–86 yrs, MMS score: mean 19.1 ± 5.3 SD. Statistical significance was assumed at P Ͻ 0.05. CTR group: n = 63, 35 men, 28 women, mean age 56.0 ± 15.0 SD yrs, range 28–84 yrs. Results AD and CTR patients were free of psychotropic medication. Patients with major depression were CSF levels of NT-3 were significantly elevated in the treated with various antidepressants: 11 were treated DE group, as compared to both the AD and the CTR with tricyclics (TCA), two with a monoamine oxidase- group (P Ͻ 0.01, and P Ͻ 0.001, Mann-Whitney U test). A inhibitor (MAOI), one with a combination of TCA CSF levels of NT-3 were not different in the AD group, with MAOI, six with selective serotonin reuptake as compared to the CTR group (Figure 1). NT-3 concen- inhibitors (SSRI), and three were free of antidepress- trations in CSF of the DE group were 25.8 ± 4.3 pg ml−1 ants at the time of lumbar puncture. It is currently (mean ± SEM, range: 0.0–87.00, n = 23), compared to unknown whether these treatments can influence CSF 14.0 ± 1.6 pg ml−1 in the AD-group (range: 0.0–41.0, levels of neurotrophins. Informed consent was taken n = 39), and 10.5 ± 1.6 pg ml−1 in the CTR group (range: from each patient and their caregivers before the inves- 0.00–67.0, n = 63), respectively (Figure 1). To estimate tigation. The study was approved by the local ethics the accuracy of CSF measurements of NT-3 in the diag- committee. All procedures were in accordance with the nosis of DE, we calculated: (a) sensitivities; and (b) 1983 revision of the 1975 Helsinki Declaration. Within specificities, defined as follows: (i) true positives/(true one week of clinical work-up and dementia testing, positives + false negatives); and (ii) true negatives/(true CSF was obtained by lumbar puncture. CSF samples were frozen on dry ice immediately upon withdrawal at the bedside in 0.5-ml aliquots, and were stored at −85°C until biochemical analyses.

NT-3 ELISA CSF levels of NT-3 were determined by using commer- cially available systems (Promega, Madison, WI, USA), and were performed according to the manufacturer’s protocol. One hundred and twenty microlitres of undi- luted CSF in carbonate buffer (pH 9.7) were added to 96-well immunoplates (Nunc Inc, Wiesbaden, Germany) at 4°C overnight. Anti-Human-NT-3 poly- clonal antibodies (pAb) were used as capture Ab. Anti- NT-3 monoclonal Ab (mAb) were used as reporter Ab. After incubation with a species-specific Ab (anti-rat IgG) conjugated to horseradish peroxidase (HRP) as a tertiary reactant, and washing, the solution was incu- bated with the chromogenic substrate TMB. Figure 1 Neurotrophin 3 (NT-3) levels in the cerebrospinal Absorbance was measured at 450 nm by using a fluid (CSF) of elderly patients with major depression (DE), microplate reader (Dynatech MR 700, Labsystems, Hel- non-depressed control subjects (CTR) and patients with Alzh- −1 sinki, Finland). Linear range 4.7–300 pg ml ; cross- eimer’s disease (AD). Levels (pg ml−1) are given in −1 reaction with other neurotrophins at 10 ng ml Ͻ3%; mean ± SEM. Asterisks (*, **) indicate significance (P Ͻ 0.05), − detection limit 6.0 pg ml 1. All CSF samples were Mann-Whitney U Test. NT-3: *DE vs AD, P Ͻ 0.01; **DE vs assayed in duplicate determinations. CTR, P Ͻ 0.001.

Molecular Psychiatry Increased CSF levels of NT-3 in depression C Hock et al 512 negatives + false positives). To estimate the probability of disease, we calculated the predictive values of the tests. The positive predictive value (PPV) was defined as true positives/(true positives + false positives). The negative predictive value (NPV) was defined as true negatives/(true negatives + false negatives). Diagnostic accuracy in DE: NT-3 measurement resulted in 73.9% sensitivity, 86.1% specificity, 50.0% PPV, 91.7% NPV using a cut-off value of Ն15.0 pg ml−1 (total n = 125, AD n = 39, DE n = 23, CTR n = 63) (Table 1). Next, we grouped the DE patients (n = 23) according to treatment with substances that affect central noradrenergic neurotransmission (MAOI and TCA, n = 14) and those that selectively affect serotonergic neurotransmission (SSRI, n = 6). CSF levels of NT-3 were significantly lower in the MAOI/TCA group, as compared Figure 2 CSF levels of NT-3 in DE patients (n = 23) grouped Ͻ to the SSRI group (P 0.05, independent samples t- according to treatment with antidepressants: substances that test). NT-3 concentrations in CSF of the MAOI/TCA affect central noradrenergic neurotransmission (MAOI and − group were 19.7 ± 2.9 pg ml 1 (mean ± SEM, range: 0.0– TCA, n = 14); substances that selectively affect serotonergic 35.0, n = 14), compared to 34.9 ± 8.5 pg ml−1 in the neurotransmission (SSRI, n = 6); no treatment with antide- SSRI group (range: 14.0–63.0, n = 6), and 35.7 ± 26.3 pg pressants at the time of lumbar puncture (n = 3). CSF levels of − ml−1 in the group with patients who were not treated NT-3 (pg ml 1) are given in mean ± SEM. Asterix (*) indicates Ͻ with antidepressants at the time of lumbar puncture significance (P 0.05, independent samples t-test, (range: 0.00–87.0, n = 3) (Figure 2). MAOI/TCA vs SSRI). There was no correlation of CSF levels of NT-3 with ApoE genotype (or phenotype, respectively), age, NT-3, but not related neurotrophins, prevents the MADRS, MMS, or NOSGER scores. degeneration of noradrenergic neurons of the locus coeruleus in a 6-hydroxydopamine lesion model.6 Alterations in adrenoreceptor density and function, as Discussion well as changes in adrenoreceptors associated with the This is the first study to demonstrate increased levels pituitary-adrenal axis function, strongly implicate of the neurotrophin NT-3 in spinal fluid of elderly abnormal central noradrenergic transmission in DE (for patients with major depression. CSF measurement of a review see Leonard).17 This dysfunction might be NT-3 levels showed a promising diagnostic accuracy related to the activity of tyrosine hydroxylase, the rate- in terms of sensitivity (73.9%) as well as specificity limiting enzyme in the synthesis of catecholamines. (86.1%). Thus, the NT-3 test constitutes a candidate Alternatively, impaired uptake or transport of target- tool for the biochemical diagnosis of major depression derived neurotrophic factors, including NT-3, may in the elderly. Further experiments will show whether contribute to abnormalities in central noradrenergic these findings may be expanded to depressive popu- function. lations of younger age. This is the first study of CSF levels of NT-3 in DE. Our finding of elevated levels of NT-3 in the CSF of NT-3 was previously determined in the CSF of patients patients with DE might reflect a cerebral alteration of with a variety of neurological conditions including NT-3 production or utilization, since NT-3 is the major hydrocephalus, meningitis, encephalitis, ventriculitis, neurotrophic factor for central noradrenergic neurons.6 brain tumors, and multiple sclerosis, with values rang- ing from 4.8 to 55.9 pg ml−1 determined with the same method as used in this study.18 In the report of Gilmore Table 1 Diagnostic accuracy of cerebrospinal fluid (CSF) et al18 NT-3 was not detectable in the CSF of patients measurements of neurotrophin 3 (NT-3) in major depression with schizophrenia. Another study showed no differ- in the elderly (DE) ence of CSF levels of NT-3 in AD as compared to normal control subjects,19 which is in full agreement with Sensitivity 73.9% Positive 50.0% our findings. However, recently, a possible association predictive of a missense mutation (Gly[-63]Glu) of the neurotro- value (PPV) phin-3 gene with AD was described in a Japanese Specificity 86.1% Negative 91.7% population.20 The mutated type was more frequent predictive = value (NPV) among the AD patients than the controls (P 0.011, odds ratio 1.63, 95% CI 1.11–2.38). Sensitivity, specificity, positive predictive value (PPV), and Treatment with antidepressants may affect NT-3 lev- 21 negative predictive value (NPV) were calculated using a cut- els in the DE group: Smith et al showed that chronic off concentration of 15 pg ml−1 NT-3. NT-3 test: total n = 125; treatment with antidepressants that selectively blocked AD: n = 39; DE: n = 23; CTR: n = 63. neuronal noradrenaline uptake decreased the

Molecular Psychiatry Increased CSF levels of NT-3 in depression C Hock et al 513 expression levels of NT-3 in the locus coeruleus in rats. 2 Price DL, Tanzi RE, Borchelt DR, Sisodia SS. Alzheimer’s disease: In contrast, treatment with selective serotonin reuptake genetic studies and transgenic models. Annu Rev Genet 1998; 32: 461–493. inhibitors did not alter NT-3 mRNA levels. These find- 3 Hyman C, Hofer M, Barde YA, Juhasz M, Yancopoulos GD, Squinto ings suggest that some effects of antidepressants on the SP et al. BDNF is a neurotrophic factor for dopaminergic neurons function of the locus coeruleus involve NT-3 of the substantia nigra. Nature 1991; 350: 230–232. expression. In addition, these findings make it rather 4 Knu¨ sel B, Winslow JW, Rosenthal A, Burton LE, Seid DP, Nicolics unlikely that the increased CSF levels of NT-3 in K et al. Promotion of central cholinergic and dopaminergic neuron differentiation by brain-derived neurotrophic factor but not neuro- patients with DE that were observed in the present trophin 3. Proc Natl Acad Sci USA 1991; 88: 961–965. study, are due to the effect of treatment with antide- 5 Mamounas LA, Blue ME, Siuciak JA, Altar CA. Brain-derived neur- pressants. In fact, CSF levels of NT-3 were markedly otrophic factor promotes the survival and sprouting of serotonergic lower within the DE group in patients treated with sub- axons in rat brain. J Neurosci 1995; 15: 7929–7939. stances that affect central noradrenergic neuro- 6 Arenas E, Persson H. Neurotrophin-3 prevents the death of adult central noradrenergic neurons in vivo. Nature 1994; 367: 368–371. transmission (MAOI/TCA) than in patients treated 7 Leonard BE. The role of noradrenaline in depression: a review. J with substances that selectively affect serotonergic Psychopharmacol 1997; 11: S39–S47. neurotransmission (SSRI) as well as in patients who 8 Gall CM, Gold SJ, Isackson PJ, Seroogy KB. Brain-derived neuro- were not treated with antidepressants at the time of trophic factor and neurotrophin-3 mRNAs are expressed in ventral midbrain regions containing dopaminergic neurons. Mol Cell Neu- lumbar puncture. Thus, treatment with antidepressants rosci 1992; 3: 56–63. that block neuronal noradrenaline uptake may reduce 9 Lindholm D, Hamner S, Zirrgiebel U. Neurotrophins and cerebellar previously increased CSF levels of NT-3 in patients development. Perspect Dev Neurobiol 1997; 5: 83–94. with DE. In contrast, in rat brain, chronic (21-day) 10 Maisonpierre PC, Belluscio L, Squinto S, Ip NY, Furth ME, Lindsay administration of several different antidepressant RM et al. Neurotrophin-3: a neurotrophic factor related to NGF and BDNF. Science 1990; 247: 1146–1451. drugs, including tranylcypromine, sertraline, desipra- 11 McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlau mine, significantly increased BDNF mRNA and trkB EM. Clinical diagnosis of Alzheimer’s disease: report of the mRNA in the hippocampus.22 These and other findings NINCDS-ADRDA Work Group under the auspices of Department of constituted the framework for a molecular hypothesis Health and Human Services Task Force on Alzheimer’s Disease. of depression, which postulated that the therapeutic Neurology 1984; 34: 939–944. 12 Folstein MF, Folstein SE, McHugh PR. Mini Mental State. A practi- action of antidepressant treatments occurs via intra- cal method for grading the cognitive state of patients for the clin- cellular mechanisms that decrease or increase, respect- ician. J Psychiatry Res 1975; 12: 189–198. ively, neurotrophic factors.23,24 To date, neither 13 Spiegel R, Brunner C, Ermini-Funfschilling D, Monsch A, Notter measurements of cerebral protein or expression levels M, Puxty J et al. A new behavioral assessment scale for geriatric of neurotrophins nor of their specific receptors and out- and in-patients: the NOSGER (Nurses’ Observation Scale for Geriatric Patients). J Am Geriatr Soc 1991; 39: 339–347. their signaling have been reported of patients with 14 Montgomery SA, Asberg M. A new depression scale designed to major depression. Thus, the present results may stimu- be sensitive to change. Br J Psychiatry 1979; 134: 382–389. late further studies to elucidate the role of NT-3 and 15 McDowell IF, Wisdom GB, Trimble ER. Apolipoprotein E pheno- related neurotrophins in the pathophysiology of type determined by agarose gel electrofocusing and immunoblot- ting. Clin Chem 1989; 35: 2070–2073. major depression. 16 Hansen PS, Gerdes LU, Klausen IC, Gregersen N, Faergeman O. In conclusion, CSF levels of NT-3 were markedly Genotyping compared with protein phenotyping of the common increased in elderly patients with DE, as compared to apolipoprotein E polymorphism. Clin Chim Acta 1994; 224: 131– mentally healthy controls and patients with AD. 137. Increased CSF levels of NT-3 may indicate a disturb- 17 Leonard BE. The role of noradrenaline in depression: a review. J Psychopharmacol 1997; 11: S39–S47. ance of the central noradrenergic system in patients 18 Gilmore JH, Jarskog LF, Lindgren JC, McEvoy JP, Xiao H. Neurotro- with DE. NT-3 may constitute a biochemical candidate phin-3 levels in the cerebrospinal fluid of patients with schizo- marker for clinical diagnosis and for the evaluation of phrenia or medical illness. Psychiatry Res 1997; 73: 109–113. therapeutic strategies in DE. 19 Murase K, Igarashi K, Hayashi K. Neurotrophin-3 (NT-3) levels in the developing rat nervous system and in human samples. Clin Chim Acta 1994; 227: 23–36. 20 Kunugi H, Hattori M, Ueki A, Isse K, Hirasawa H, Nanko S. Possible Acknowledgements association of missense mutation (Gly[-63]Glu) of the neurotrophin-3 gene with Alzheimer’s disease in Japanese. Neurosci Lett This work was supported by Grants 3100–049397.96/1 1998; 241: 65–67. (Schweizerischer Nationalfonds) (UO, CH, FM-S), and 21 Smith MA, Makino S, Altemus M, Michelson D, Hong SK, Kvetnan- SFB505 (Deutsche Forschungsgemeinschaft) (UO). We sky R et al. Stress and antidepressants differentially regulate neuro- thank F Jancu for technical support and K Kraüchi, trophin 3 mRNA expression in the locus coeruleus. Proc Natl Acad Sci USA 1995; 92: 8788–8792. PhD, for help in statistical analyses. 22 Nibuya M, Morinobu S, Duman RS. Regulation of BDNF and trkB mRNA in rat brain by chronic electroconvulsive seizure and antidepressant drug treatments. J Neurosci 1995; 15: 7539–7547. References 23 Duman RS, Heninger GR, Nestler EJ. A molecular and cellular theory of depression. Arch Gen Psychiatry 1997; 54: 597–606. 1 Bothwell M. Functional interactions of neurotrophins and neuro- 24 Altar CA. Neurotrophins and depression. Trends Pharmacol Sci trophin receptors. Ann Rev Neurosci 1995; 8: 223–253. 1999; 20: 59–61.

Molecular Psychiatry