PROINFLAMMATORY AND “RESILIENCY” PROTEINS IN THE CSF OF PATIENTS WITH MAJOR DEPRESSION

Research ArticlePROINFLAMMATORY AND ‘‘RESILIENCY’’ PROTEINS

IN THE CSF OF PATIENTS WITH MAJOR DEPRESSION

Jose M. Martinez, M.A.,1� Amir Garakani, M.D.,1 Rachel Yehuda, Ph.D.,1,2 and Jack M. Gorman, M.D.3

Background: A number of studies have shown that elevated levels ofinflammatory cytokines may promote depression and suicidal ideation and thatneuroprotective peptides may decrease the response to stress and depression.In this study, cerebrospinal fluid (CSF) levels of three inflammatory cytokines(IL-1, IL-6, and tumor necrosis factor a (TNFa)) and two putative ‘‘resiliency’’neuropeptides (brain-derived neurotrophic factor (BDNF) and neuropeptide Y(NPY)) were compared between patients with depression and healthy controls.Methods: Eighteen patients with major depression and 25 healthy controlsunderwent a lumbar puncture; CSF samples were withdrawn and assayed forIL-1, IL-6, TNFa, BDNF, and NPY levels. Patients with depression were thenentered into an 8-week treatment protocol and had repeated lumbar punctureprocedures post-treatment. Results: Contrary to prediction, we found that atbaseline depressed patients had higher CSF NPY concentration compared to thenormal comparison group. Within the depressed patients, we found severalstatistically significant correlations between elevated CSF cytokine levels andclinical severity. Conclusion:Despite the small sample size, given the challengesin obtaining CSF from patients with depression these data are of interest inconfirming some aspects of the inflammatory hypothesis of depression.

rrrr 2011 Wiley-Liss, Inc.

Key words: cytokines; brain-derived neurotrophic factor; cerebrospinal fluid;neuropeptide Y; depression

INTRODUCTIONA number of recent studies have suggested thatactivation of the immune system may be involved in thepathophysiology of major depression and of suicidalbehavior.[1–4] At the same time, several neuropeptideshave been linked to resiliency against stress and areconceived therefore to confer antidepressant effects.Among these are neuropeptide Y (NPY) and brain-derived neurotrophic factor (BDNF). NPY is widelypresent throughout the CNS and binds to at least fourdifferent receptors.[5] Many clinical and preclinicalstudies have found a role for NPY in the regulation ofemotional behavior and stress response.[6] Infusion ofNPY directly into the rat hippocampus producedantidepressant-like effects (Ishida et al., 2007).[7] Ithas been suggested that NPY opposes the anxiogeniceffects of CRH in the rodent brain.[8]

The role of NPY in human depression is somewhatless clear, but may be illuminated when attention ispaid to polymorphisms in the gene encoding the

neuropeptide. For example, in a sample of 256 patientswith depression, the presence of a less active NPYallele was associated with failure to respond toantidepressant therapy and to increased activation ofthe bilateral amygdala during presentation of fearful

Published online in Wiley Online Library (wileyonlinelibrary.com).

DOI 10.1002/da.20876

Received for publication 29 January 2011; Revised 28 June 2011;

Accepted 5 July 2011

The authors disclose the following financial relationships within thepast 3 years: Contract grant sponsor: National Institutes of Health

(NIH); Contract grant numbers: R01-MH58808; K08-MH6701.

�Correspondence to: Jose M. Martinez, Department of Psychiatry,

Mount Sinai School of Medicine, One Gustave L. Levy Place,

New York, NY 10029. E-mail: [email protected]

1Department of Psychiatry, Mount Sinai School of Medicine,

New York, New York2James J Peters Veterans Affairs Bronx, New York3Riverdale Behavioral Health Consultants, Bronx, New York

rrrr 2011 Wiley-Liss, Inc.

DEPRESSION AND ANXIETY 29:32–38 (2012)

Depression and Anxiety 29:32–38, 2012.

faces, particularly among those patients with ‘‘anxiousdepression.’’[9] Frisch et al.[10] measured NPY-positiveneuron density in 34 patients with epilepsy undergoingtemporal lobe surgery. They found a positive correla-tion between NPY neuron density and depression andanxiety scores in the basolateral amygdala. NPY hasalso been linked to alcohol dependence and mayrepresent a link between emotional behavior and thepropensity for alcoholism.[11]

In studies of human cerebrospinal fluid (CSF), NPYis reduced in depressed patients and increases withantidepressant therapy.[12–14] Men with combat-relatedPTSD had lower CSF NPY concentrations thannormal volunteers.[15] In a sample of six patients,Nikisch and Mathe[16] reported an increase in CSFNPY level and reduction in corticotropin-releasinghormone (CRH) level following electroconvulsivetherapy. Olsson et al.[17] in a small sample of 7 of 13inpatients with depression (after a recent suicideattempt) being treated with an antidepressant, reporteda significant decrease in CSF NPY between theirsecond and third lumbar punctures and a decrease inCSF substance P across the patient group.Similarly, BDNF has been shown to induce anti-

depressant-like effects in laboratory animals and to bereduced in the serum of depressed patients.[18,19] Chenet al.[20] showed that BDNF concentrations wereincreased in postmortem brains of depressed patientswho had been treated with antidepressants compared tothose who had not received antidepressants. Anotherpostmortem study found reduced BDNF mRNA levelin the prefrontal cortex and hippocampus of subjectswho had committed suicide.[21] A number of studieshave shown that antidepressant therapy increasesserum BDNF levels.[18,22] The relationship betweenbrain, CSF, and blood levels of BDNF is likely to becomplex. Elfving et al.[23] compared BDNF levelsbetween two strains of mice: the Flinders SensitiveLine (FSL), considered to be a genetic model ofdepression, and the Flinders Resistant Line (FRL).FSL mice had higher serum, lower hippocampal, andsimilar CSF BDNF levels compared to FRL mice. CSFBDNF concentration has been reported in otherCNS diseases. Compared to normal controls, CSFBDNF levels were found to be increased in patientswith Parkinson’s disease,[24] decreased in patients withAlzheimer’s disease[25] and equivalent in patients withPTSD[26] and amyotrophic lateral sclerosis.[27] To ourknowledge, there is only one known published study ofCSF BDNF concentrations in patients with depression,a report that evaluated the CSF levels of BDNF, IL-6and corticosterone in patients with MDD and Parkin-son’s Disease and patients with PD or MDD alone.[28]

They reported lower levels of IL-6 and BDNF inpatients with comorbid PD and MDD compared tothose with PD alone after the PD/MDD group andMDD group received 12 weeks of citalopram (thepatients with PD alone did not receive citalopram).Those with MDD alone were reported to have an

increase in CSF BDNF after the treatment withcitalopram and their pretreatment levels of BDNFand IL-6 were higher than those with comorbid PDand MDD.Until the past several decades, the central nervous

system was often conceived as immunological indolentor ‘‘spared.’’ It is now clear that this is far from true.Major advances have been made in understanding thebiology of the brain’s main immunological cell-type,the microglia cell.[29] Microglia serve mainly to protectneurons and other glial cells from toxic and infectiousagents. However, microglia cells are activated by painand peripheral immune stimulation and when activatedcan release inflammatory cytokines directly into thebrain. Hence, stress that affects both the central andperipheral nervous systems is capable of resulting inmicroglial activation and increased central cytokinelevels. Both animal and clinical data suggest thatimmune system activation may be an important aspectin the pathophysiology of depression.A prevailing hypothesis is that peripheral immune

activation triggers cytokine synthesis and release,which then provokes activation of brain neurotrans-mitter and neuroendocrine systems that contribute todepression.[30,31] This hypothesis rests in part onseveral reports of immune system activation duringstress in animals and humans and in depressed patients.Immune system activation by lipopolysaccharide injec-tion increases inflammatory cytokine levels and causesa syndrome of behavioral despair in laboratoryrodents.[32] Inflammatory cytokines may alter hippo-campal plasticity in experimental animals.[33] Slavichet al.[34] recently reported that a social-threat taskincreased both IL-6 and TNF-a in a group of 124normal volunteer subjects. Serum levels of IL-6increased more in response to the Trier Social StressTest in men with major depression and a history ofearly life stress than in normal volunteers.[35] IL-6levels also increased more during the Trier test innondepressed adults with a history of childhoodmaltreatment compared to adults with no such earlylife history.[36]

It is unclear to what degree findings of increasedperipheral levels of inflammatory cytokines in de-pressed patients are related to chronic stress versusdepression itself. For example, in one study IL-6 levelswere increased only in patients with major depressionand post-traumatic stress disorder but not in patientswith major depressive disorder alone.[37] On the otherhand, serum levels of the anti-inflammatory markerIL-1 receptor antagonist were also found in a groupwith elevated depression symptoms.[38] Treatment withantidepressant medication has been shown to reducelevels of inflammatory cytokines,[39,40] and this effectmay only occur in patients whose depression respondsto medication.[41] In fact, response to SSRI antidepres-sants may in part be mediated by polymorphisms onthe gene encoding IL-6.[42] Both depression[43] andelevated serum levels of IL-1 and TNF-a[44] may be

2 Martinez et al.

Depression and Anxiety Depression and Anxiety

33Research Article: Cytokines, Resiliency Proteins and Major Depression

risk factors for Alzheimer’s disease. Elevated inflam-matory markers may be independent contributors tothe excess in cardiovascular disease morbidity seen insome studies of patients with depression.[45]

There has been considerable discussion in theliterature about potential mechanisms by which per-ipheral cytokines may influence the central nervoussystem. Cytokines cannot by themselves cross theblood–brain barrier.[46] However, several hypotheseshave been advanced to explain the relationship betweenthe peripheral and central immune systems,[46–48]

including: (1) Passive transport through areas of thebrain that lack a blood–brain barrier; (2) activation byperipheral cytokines of receptors on pathways thatsignal the synthesis and release of cytokines by theCNS, such as those on the vagus nerve that terminatein the NTS. Cytokines are known to be secreted denovo in the brain by microglia, astrocytes and, undersome circumstances, neurons; (3) binding of peripheralcytokines to receptors in the brain vascular endothe-lium, triggering the release of secondary messengersthat provoke a central immune response; (4) releasefrom peripheral macrophages that can traverse theblood–brain barrier; (5) transport of peripheral cyto-kines into the brain by active carriers; and (6) activationof CRH containing terminals located outside theblood–brain barrier that can then lead to increasedCRH activity within the brain.Only a few studies have examined cytokines in the

CSF of patients with depression. Over a decade ago, astudy involving 13 unmedicated hospitalized patientswith depression and 10 normal volunteers found thatIL-1b levels were higher in the CSF of depressedpatients, but IL-6 levels were lower and there was nodifference between groups in TNF-a level (Levineet al., 1999). Lindqvist et al. (2009) found elevated CSFIL-6 levels in suicide attempters compared to normalcontrols and a significant positive correlation betweenMADRS scores and CSF IL-6 levels in the patients.Interestingly, in this study there was no relationshipbetween serum and CSF cytokine levels. Bonne et al.(in press) found no significant difference of pretreat-ment concentrations of CSF corticotrophic-releasingfactor, IL-6, BDNF, or substance P in chronic PTSDpatients compared to healthy controls, post-treatmentCSF measures did not change significantly. A recentreport from Brundin et al. at the Lund University thatis as yet unpublished found ‘‘high levels of inflamma-tion-related substances (cytokines) inyspinal fluid’’ ofpatients ‘‘who had been diagnosed with major depres-sion or who had made violent suicide attempts’’ (URL:http://www.fiercebiotech.com/press-releases/biologi-cal-changes-suicidal-patients).In this study, we took advantage of unused CSF

samples from a study originally designed to investigatethe role of the thyroid hormone transport proteintransthyretin to compare the levels of three inflamma-tory cytokines (IL-1, IL-6, and TNF-a) and twoputatively protective neuropeptides (brain-derived

neurotrophic factor and NPY) between patients withmajor depression both before and after antidepressanttherapy and a normal comparison sample. We soughtto measure both cytokines and ‘‘resiliency peptides’’(BDNF and NPY) in CSF with the expectation that theformer would be elevated in patients with depressioncompared to healthy controls and that the latter wouldbe lower.

MATERIALS AND METHODS

Subjects for this study were patients with DSM-IIIR defined majordepressive disorder and comparison subjects with no currentpsychiatric illness. All subjects underwent a thorough medicalevaluation prior to baseline testing. This included a physicalexamination, neurological examination with visualization of thefundi; blood tests for routine chemistries, T3, T4, TSH, CBC, andpregnancy test; urinalysis; urine screen for substances of abuse andEKG. All subjects had a psychiatric evaluation and SCID to confirmdiagnosis.

In addition to meeting major depression criteria, depressedsubjects had to have a Hamilton Depression Rating Scale score ofgreater than 15, indicating at least moderate severity. Exclusioncriteria for the depressed patient group were history of bipolardisorder or schizophrenia, history in the past year of substance abuseor dependence, panic disorder, obsessive compulsive disorder oreating disorder, any history of thyroid disease, significant medicaldisorders or use of medications that may distort study measures,significant history of clinical depression that is treatment refractoryand unlikely to respond to standard clinical care and pregnantwomen. Healthy controls needed to have a Hamilton DepressionScale score of less than 8. Healthy controls were excluded if they meta history of major depressive disorder, bipolar disorder, schizo-phrenia, panic disorder, obsessive compulsive disorder, eatingdisorder, or substance use disorder, any history of thyroid diseaseor use of thyroid medication, significant medical disorder or the useof medication that would distort study measures and pregnantwomen.

All subjects were maintained free of psychotropic medication for2 weeks (6 weeks for fluoxetine) prior to study. As-needed use ofbenzodiazepine anxiolytics or hypnotics, such as zolpidem and chloralhydrate, was permitted.

After 2 weeks of being medication-free and after an overnight fast,a lumbar puncture was performed between the hours of 8 and 9 a.m.Ten cc. of CSF were withdrawn and aliquoted in 1 cc portions. CSFaliquots were stored at �701C until assay. Patients with depressionwere then entered in the 8-week treatment protocol.

Standards enzyme-linked immunosorbent assay (ELISA) wereused with a lower limit of detection around 10–15 pg/ml, translatingto roughly an absorbance of 0.01. If a level was not detected it wasrecorded as zero. Assays were run in a blinded fashion so that theidentities of the samples (patient or control) were not known.

All patients were started on extended release venlafaxine 37.5mgdaily. Treatment consisted of a fixed flexible dose regimen with aminimum target dose of 225mg per day and a maximum dose of375mg. Dose escalation above 225mg daily occurred after 4 weeksfor patients who were nonresponsive at this dose. For patients whodid not tolerate 225mg daily, we permitted dose reduction to 150mgdaily. The following rating scales and measures were obtained weeklythroughout the study: the Hamilton Depression Rating Scale,Clinical Global Severity Scale, and Clinical Global Improvement(CGI) Scale. Suicide ideation occurring in the 2 weeks preceding thelumbar puncture procedure was measured using the Scale for Suicide


Depression and Anxiety

rr


34 Martinez et al.

Ideation (SSI).[49] Suicide attempt was defined as a self-destructive actwith some degree of intent to end one’s life. Details of suicideattempts, including documentation of the number, method, andlethality rating of attempts, were recorded using the ColumbiaSuicide History Form.[50] At the conclusion of the 8-week treatmentprotocol and while still taking venlafaxine, the patients underwentrepeated lumbar puncture and blood draw for all measures listedduring baseline testing.

The differences between patients and comparison subjects for allcontinuous measures were calculated using standard t-tests orMann–Whitney U Test. Differences within the patient group betweenthe pre- and post-treatment tests were calculated using WilcoxonSigned Rank Test. Correlations were calculated using Pearson’scoefficients. Significance levels were set at Po.05, two tailed. Samplesizes vary for the different measures because of missing data.

The research was approved by the Internal Review Boards ofColumbia University and Mount Sinai School of Medicine and allsubjects signed written, informed consent forms prior to participation.

RESULTSThe 18 patients with major depression included 8

males and 10 females with a mean age of 40.4 years(710.0), while the 25 nondepressed comparisonsubjects included 13 males and 12 females with a meanage of 29.9 years (76.7). The difference in age betweenthe two groups was statistically significant (t5�4.1,df5 41, Po.00). Body Mass Index between the twogroups was not significantly different (healthy controlshad a mean BMI5 24.4 and MDD patients had a meanBMI of 24.1). None of the healthy control subjectsmoked, among the depressed patients two werelight smokers (less than 10 cigarettes/day, 2 weremoderate smokers (between 10 and 20 cigarettes/day)and the remainder did not smoke. At baseline,depressed patients had mean scores on the HamiltonDepression Scale (HAMD), Clinical Global SeverityScale (CGS), and SSI of 19.273.8 (N5 13), 4.370.50(N5 12), and 1.872.3 (n5 13), respectively (Table 1).After treatment with venlafaxine, the patients CGIscale score was mean 2.3370.99, n5 12. Of these 12,two had no clinical change, two experienced onlyminimal improvement, and eight met criteria forimprovement by CGI criteria of much better or verymuch better (two or one).At baseline, before antidepressant treatment, de-

pressed patients (N5 16) had significantly higher CSFconcentration of NPY than comparison subjects(N5 17) (176.1747.3 pg/ml versus 137.7724.1 pg/ml;

t5�2.9, df5 22.0, Po.007). There were no statisti-cally significant differences at baseline between groupsfor BDNF, IL-1, IL5 6, or TNF-a (Table 2).After 8 weeks of treatment with venlafaxine patients

with depression experienced statistically significantimprovements in HAMD, CGS, and SSI scale scores(Table 1). However, there were no significant changesbetween pre- and post-treatment time points in any ofthe CSF measures (Table 3).Correlation coefficients between clinical severity and

CSF measures showed that among patients at baselinethere were statistically significant positive correlationsbetween CSF IL-1 concentration and history of suicideattempts (r5 .53, P5.041, N5 15) and between CSFBDNF concentration and SSI score (r5 .62, P5.033,N5 12). After treatment there were significant positivecorrelations between CSF IL-6 concentration and SSIscale (r5 .85, P5.008, N5 8) and between CSF TNFaconcentration and SSI scale (r5 .81, P5.008, N5 9).Finally, there was a significant positive correlationbetween baseline CSF TNFa and post-treatment CGIscore (r5 .68, P5.030, N5 10).

TABLE 1. Comparison of behavioral measures pre- andpost-treatment

VariablePre-

treatmentPost-

treatmentPaired

sample test

HAM-D 19.273.8 10.674.3 t5 5.78; Po.001; df5 12CGS 4.370.5 3.371.4 t5 2.87; Po.015; df5 11Suicide index 1.872.4 0.771.4 t5 2.42; Po.032; df5 12

CGS, Clinical Global Severity Scale; HAMD, Hamilton DepressionScale.

TABLE 2. Comparison of baseline biological variablesbetween depressed patients and controls

VariablePatients Healthy controls Mann–Whitney U test

BDNF 4.2871.88 4.8672.22 NSIL-1 0.07370.022 0.06470.004 NSIL-6 0.06670.010 0.06070.007 NSTNF-a 0.10570.113 0.09570.057 NSNPY 176.13747.26 137.66724.14 Po.031

BDNF, brain-derived neurotrophic factor; IL, interleukin; TNF,tumor necrosis factor; NPY, neuropeptide Y.

TABLE 3. Comparison of pre- and post-treatment biological measures in depressed patients

Variable Pre-treatment Post-treatment Wilcoxon signed ranks test

BDNF (n5 12) 4.0372.030 4.4972.005 NSIL-1 (n5 9) 0.07870.027 0.06970.012 NSIL-6 (n5 8) 0.05970.004 0.09170.070 NSTNF-a (n5 9) 0.07870.035 0.07570.023 NSNPY (n5 10) 187.01740.23 179.68753.97 NS

BDNF, brain-derived neurotrophic factor; IL, interleukin; TNF, tumor necrosis factor; NPY, neuropeptide Y.

4 Martinez et al.



DISCUSSION

The study reported here has many obvious short-comings, primary among which are the small samplesize, the fluctuating sample sizes among variousmeasures because of missing data, and the lack of asecond lumbar puncture from the normal comparisonsubjects, which would have enabled us to have a betterpicture of whether there was any ‘‘normalization’’ ofCSF indices in patients following antidepressanttreatment. It is also never clear to what extent thelevel of any neuropeptide or neurotransmitter in theCSF reflects its actual activity at critical synaptic andcellular levels. Finally, there is a statistically significant,nearly 10-year age difference between patients andcomparison subjects. Although this may affect theresults we obtained in the comparison of baseline NPYconcentrations between the two groups, it should notbe relevant to the significant correlations we observedwithin the patient group.Nevertheless, given the well-known difficulties of

obtaining CSF samples from patients with psychiatricillnesses, we believe that these data are worth reporting,even though some of the results are not in the predicteddirection. First, we found an unexpectedly higher CSFNPY concentration in patients than comparisonsubjects at baseline and did not detect any significantincrease in NPY concentration with antidepressanttreatment as has been reported before. Although thesedata support the notion that NPY may be involved inthe pathophysiology of depression, they are equivocalas to its role in promoting ‘‘resiliency.’’ One could, ofcourse, speculate that higher NPY levels in depressedpatients represent an adaptive change to the illness.Similarly, the positive correlation between baselineCSF BDNF concentration and severity of suicidalideation appears, like our NPY finding, to be theopposite of what would be expected if BDNF is indeeda ‘‘resiliency’’ neuropeptide.Second, we found several positive correlations

between CSF inflammatory cytokine concentrationsand clinical measures. The significant correlationsmust be taken into context with the fact that wepreformed multiple tests of correlation, most of whichwere not statistically significant. The relationshipbetween baseline IL-1 concentration and history ofsuicide attempts appears to be in keeping with thegeneral notion that inflammatory cytokines are asso-ciated with severity of depression and suicidal idea-tion.[51] Similarly, the significant correlations betweenIL-6 and TNFa concentrations collected from patientsat the second lumbar puncture when they had beenreceiving venlafaxine and score on the suicidal ideationscale are also compatible with the inflammatorycytokine theory, although it is unclear why these samerelationships were not statistically significant at base-line, before treatment.Third, the significant positive correlation between

baseline TNFa concentration and post-treatment CGI

suggest that the former is a predictor of poor treat-ment outcome, as high CSF TNFa before treat-ment predicted poorer response to antidepressantmedication.

CONCLUSIONAlthough some of the findings from these analyses of

CSF indices in depressed patients may seem counter-intuitive, in general they support the hypotheses thatinflammatory cytokines may play a role in depressionand that both NPY and BDNF are dysregulated inpatients with depression. These appear to be researchavenues well worth pursuing.

Acknowledgments. The authors acknowledgeDr. Lloyd Mayer, for his help with the assay of CSFsamples. The authors declare that they have noconflicting interest relevant to this manuscript.

REFERENCES1. Raison Cl, Capuron L, Miller AH. Cytokines sing the blues:

inflammation and the pathogenesis of depression. TrendsImmunol 2008;27:24–31.

2. Adler UC, Marques AH, Calil HM. Inflammatory aspects ofdepression. Inflamm Allergy Drug Targets 2008;7:19–23.

3. Miller AH, Maletic V, Raison CL. Inflammation and itsdiscontents: the role of cytokines in the pathophysiology ofmajor depression. Biol Psychiatry 2009;65:732–741.

4. Dowlati Y, Herrmann N, Swardfager W. A meta-analysisof cytokines in major depression. Biol Psychiatry 2010;67:446–457.

5. Morales-Medina JC, Dumont Y, Quirion R. A possible role ofneuropeptide Y in depression and stress. Brain Res 2010;1414:194–205.

6. Eaton K, Sallee FR, Sah R. Relevance of neuropeptide Y (NPY)in psychiatry. Curr Top Med Chem 2007;7:1645–1659.

7. Ishida H, Shirayama Y, Iwata M, et al. Infusion of neuro-peptide Y into CA3 region of hippocampus produces anti-depressant-like effect via Y1 receptor. Hippocampus 2007;17:271–280.

8. Giesbrecht CJ, Jackay JP, Silveira HB, Urban JH, Colmers WF.Countervailing modulation of Ih by neuropeptide Y andcorticotrophin-releasing factor in basolateral amygdala as apossible mechanism for their effects on stress-related behaviors.J Neurosci 2010;30:16970–16982.

9. Donschke K, Dannolowski U, Hohoff C, et al. Neuropeptide Y(NPY) gene: impact on emotional processing and treatmentresponse in anxious depression. Eur Neuropsychopharmacol2010;20:301–309.

10. Frisch C, Hanke J, Kleineruschkamp S, et al. Positive correlationbetween the density of neuropeptide y positive neurons in theamygdala and parameters of self-reported anxiety and depressionin mesiotemporal lobe epilepsy patients. Biol Psychiatry2009;66:433–440.

11. Carvajal C, Dumont Y, Quirion R. Neuropeptide Y: role inemotion and alcohol dependence. CNS Neurol Disord DrugTargets 2008;5:181–195.

12. Nikisch G, Agren H, Eap CB, Czernik A, Baumann P, Mathe AA.Neuropeptide Y and corticotropin-releasing hormone in CSF



rr


36 Martinez et al.

mark response to antidepressive treatment with citalopram. Int JNeuropsychopharmacol 2005;8:403–410.

13. Matte AA, Husuam H, El Koury A, et al. Search for biologicalcorrelates of depression and mechanisms of action of anti-depressant treatment modalities? Do neuropeptides play a role?Physiol Behav 2007;92:226–231.

14. Madaan V, Wilson DR. Neuropeptides: relevance in treatment ofdepression and anxiety disorders. Drug News Perspect2009;22:319–324.

15. Sah R, Ekhator NN, Strawn JR, et al. Low cerebrospinal fluidneuropeptide Y concentrations in posttraumatic stress disorder.Biol Psychiatry 2009;66:705–707.

16. Nikisch G, Mathe AA. CSF monoamine metabolites andneuropeptides in depressed patients before and after electro-convulsive therapy. Eur Psychiatry 2008;23:356–359.

17. Olsson A, Regnell G, Traskman-Bendz L, Ekman R, Westrin A.Cerebrospinal neuropeptide Y and substance P in suicideattempters during long-term antidepressant treatment. EurNeuropsychopharmacol 2004;14:479–485.

18. Sen S, Duman R, Sanacora G. Serum brain-derived neurotrophicfactor, depression, and antidepressant medications: meta-analysesand implications. Biol Psychiatry 2008;64:527–532.

19. Dwivedi Y. Brain-derived neurotrophic factor: role in depressionand suicide. Neuropsychiatr Dis Treat 2009;5:433–449.

20. Chen B, Dowlatshani D, MacQueen GM, Wang JF, Young LT.Increased hippocampal BDNF immunoreactivity in subjectstreated with antidepressant medication. Biol Psychiatry2001;50:260–265.

21. Dwivedi YY, Rizavi HS, Conley RR, Roberts RC, Tamminga CA,Pandey GN. Altered gene expression of brain-derivedneurotrophic factor and receptor tyrosine kinas B inpostmortem brain of suicide subjects. Arch Gen Psychiatry2003;60:804–815.

22. Brunoni AR, Lopes M, Fregni F. A systematic review and meta-analysis of clinical studies on major depression and BDNF levels:implications for the role of neuroplasticity in depression. Int JNeuropsychopharmacol 2008;11:1169–1180.

23. Elfving B, Plougmann PH, Muller HK, Mathe AA, Rosenberg R,Wegener G. Inverse correlation of brain and blood BDNF levelsin a genetic rat model of depression. Int J Neuropsychopharmacol2010;13:563–572.

24. Salehi Z, Mashayekhi F. Brain-derived neurotrophic factorconcentrations in the cerebrospinal fluid of patients withParkinson’s disease. J Clin Neurosci 2009;16:90–93.

25. Laske C, Stransky E, Leyhe T, et al. BDNF serum andCSF concentrations in Alzheimer’s disease, normal pressurehydrocephalus and healthy controls. J Psychiatr Res 2007;41:387–394.

26. Bonne O, Gill JM, Luckenbaugh DA, et al. Corticotropin-releasing factor, interleukin-6, brain-derived neurotrophic factor,insulin-like growth factor-1, and substance P in the cerebrospinalfluid of civilians with posttraumatic stress disorder before andafter treatment with paroxetine. J Clin Psychiatry 2010; [Epubahead of print].

27. Grundstrom E, Lindholm D, Johansson A, Blennow K,Ashmark H. GDNF but not BDNF is increased in cerebrospinalfluid in amyotrophic lateral sclerosis. Neuroreport 2000;11:1781–1783.

28. Palhagen S, Qi H, Martensson B, Walinder J, Granerus AK,Svenningsson P. Monoamines, BDNF, IL-6 and corticosteronein CSF in patients with Parkinson’s disease and major depression.J Neurol 2010;257:524–532.

29. Graeber MB. Changing face of microglia. Science2010;330:783–788.

30. Anisman H, Merali Z. Cytokines, stress and depressive illness:brain-immune interactions. Ann Med 2003;35:2–11.

31. Hayley S, Poulter MO, Merali Z, Anisman H. Thepathogenesis of clinical depression: stressor-and cytokine-induced alterations of neuroplasticity. Neuroscience 2005:135:659–678.

32. Zhu B-B, Lindler KM, Owens AW, Daws LC, Blakely RD,Hewlett WA. Interleukin-1 receptor activation by systemiclipopolysaccharide induces behavioral despair linked to MAPKregulation of CNS serotonin transporters. Neuropsychopharma-cology 2010;35:2510–2520.

33. Seguin JA, Brrennan J, Mangano E, Hayley S. Proinflammatorycytokines differentially influence adult hippocampal cellproliferation depending upon the route and chronicity ofadministration. Neuropsychiatr Dis Treat 2009;5:5–14.

34. Slavich GM, Way BM, Eisenberger NI, Taylor SE. Neuralsensitivity to social rejection is associated with inflammatoryresponses to social stress. Proc Natl Acad Sci USA2010;107:14817–14822.

35. Pace TW, Mletzko TC, Algabe O, et al. Increased stress-inducedinflammatory responses in male patients with major depressionand increased early life stress. Am J Psychiatry 2006;163:1630–1633.

36. Carpenter LL, Gawuga C, Tyrka AR, Lee JK, Anderson GM,Price LH. Association between plasma IL-6 response to acutestress and early-life adversity in healthy adults. Neuropsycho-pharmacology 2010;35:2617–2623.

37. Gill GP, Luckenbaugh D, Charney C, Vythilingham M.Sustained elevation of serum interleukin-6 and relative insensi-tivity to hydrocortisone differentiates posttraumatic stressdisorder with and without depression. Biol Psychiatry2010;68:999–1006.

38. Lehton SM, Niskanen L, Miettola J, Tolmunen T, Viinamaki H,Mantyselka P. Serum anti-inflammatory markers in generalpopulation subjects with elevated depressive symptoms. NeurosciLett 2010;484:201–205.

39. Basterzi AD, Aydemir C, Kisa C, et al. IL-6 levels decrease withSSRI treatment in patients with major depression. HumPsychopharmacol 2005;20:473–476.

40. Janssen DGA, Caniato RN, Verster JC, Baune BT.A psychoneuroimmunological review on cytokines involved inantidepressant treatment response. Hum Psychopharmacol2010;25:201–215.

41. O’Brien SM, Scully P, Fitzgeral P, Scott LV, Dinan TG. Plasmacytokine profiles in depressed patients who fail to respond toselective serotonin reuptake inhibitor therapy. J Psychiatr Res2007;41:326–331.

42. Uher R, Perroud N, Ng MY, et al. Genome–w-wide pharma-cogenetics of antidepressants response in the GENDEP project.Am J Psychiatry 2010;167:555–564.

43. Rapp MA, Schnaider-Berri M, Grossman HT, et al. Increasedhippocampal plaques and tangles in patients with Alzheimerdisease with a lifetime history of major depression. Arch GenPsychiatry 2006;63:161–167.

44. Tan ZS, Beiser AS, Vasan RS, et al. Inflammatory markers andthe risk of Alzheimer disease: the Framingham study. Neurology2007;68:1902–1908.

45. Kop WJ, Stein PK, Tracy RP, Barzilay JI, Schulz R,Gottdiener JS. Autonomic nervous system dysfunction andinflammation contribute to the increased cardiovascularmortality risk associated with depression. Psychosom Med2010;72:626–635.

46. Dunn AJ. Effects of cytokines and infections on brain neuro-chemistry. Clin Neurosci Res 2006;6:52–68.

6 Martinez et al.



47. Kronfol Z, Remick DG. Cytokines and the brain: implicationsfor clinical psychiatry. Am J Psychiatry 2000;157:683–694.

48. Schiepers OJ, Wichers MC, Maes M. Cytokines and majordepression. Prog Neuropsychopharmacol Biol Psychiatry2006;29:201–217.

49. Beck AT, Kovacs M, Weissman A. Assessment of suicidalintention: the scale for suicide ideation. J Consult Clin Psychol1979;47:343–352.

50. Oquendo MA, Halberstam B, Mann JJ. Risk factors forsuicidal behavior: utility and limitations of research instruments.In: First MB, editor. Standardized Evaluation in ClinicalPractice. Vol. 22. Washington, DC: American PsychiatricPublishing, Inc.; 2003.

51. Lindqvist D, Janelidze S, Hagell P, et al. Interleukin-6 is elevatedin the cerebrospinal fluid of suicide attempters and related tosymptom severity. Biol Psychiatry 2009;66:287–292.



rr


38 Martinez et al.

PROINFLAMMATORY AND “RESILIENCY” PROTEINS IN THE CSF OF PATIENTS WITH MAJOR DEPRESSION

Documents

Transcript of PROINFLAMMATORY AND “RESILIENCY” PROTEINS IN THE CSF OF PATIENTS WITH MAJOR DEPRESSION