JOURNAL OF

www.elsevier.com/locate/jpsychires

Journal of Psychiatric Research 41 (2007) 694–701

PSYCHIATRIC

RESEARCH

Cognitive deterioration in Alzheimer’s disease is accompaniedby increase of plasma neopterin

Imrich Blasko a,*, Gabriele Knaus a, Elisabeth Weiss a, Georg Kemmler a,Christiana Winkler b,c, Gerda Falkensammer d, Andrea Griesmacher d,

Reinhard Wurzner e, Josef Marksteiner a, Dietmar Fuchs b,c

a Department of Psychiatry, Innsbruck Medical University, Anichstrasse 35, A-6020 Innsbruck, Tyrol, Austriab Division of Biological Chemistry, Biocenter Innsbruck Medical University, Austria

c Ludwig Boltzmann Institute of AIDS-Research, Innsbruck, Austriad Department of Medical and Chemical Diagnostic of the Innsbruck Medical University, Austria

e Division of Hygiene and Medical Microbiology, Innsbruck Medical University, Austria

Received 8 November 2005; received in revised form 30 January 2006; accepted 1 February 2006

Abstract

The pro-inflammatory reaction of the immune system is a feature of healthy aging and might influence the progression of Alzhei-mer’s disease (AD). Neopterin is a pteridine derivative, released from macrophages upon stimulation with pro-inflammatory cytokineinterferon-c. Forty-three probable AD patients were investigated at baseline and follow up (14.5 ± 0.5 months; mean ± s.e.m.). Weassessed the clinical progression by the Consortium to Establish a Registry for Alzheimer’s disease (CERAD) battery and comparedcognitive changes to serum concentrations of neopterin, C-reactive protein (CRP) and antibody to cytomegalovirus (CMV). Themean neopterin concentrations increased significantly from 9.8 ± 1.0 to 13.6 ± 2.1 nM (p = 0.04). In contrast, mean CRP concentra-tions at baseline was 0.46 ± 0.1 and non-significantly decreased to 0.28 ± 0.04 mg/dl. Of AD patients 70% were CMV IgG-seropos-itive at baseline and CMV-antibody concentrations correlated with levels of neopterin (Spearman r = 0.386, p = 0.016). CERADscores did not correlate with any of immune parameters at baseline. At follow up, the increase of neopterin correlated significantlywith the decrease in the total CERAD and MMSE scores, according to the clinical progression (r = �0.353, p < 0.05 and r = �0.401,p < 0.01, respectively). Subdividing the sample with respect to baseline MMSE scores, neopterin concentrations significantlyincreased only in the group of MMSE <20. In the multiple testing covariated for age, gender, Apolipoprotein E-e4 allele, time dif-ference between both measurements, neopterin remained significantly associated with cognitive decline. In summary, neopterin con-centrations correlated with cognitive decline in AD patients, which might be due to high CMV seropositivity in that population.� 2006 Elsevier Ltd. All rights reserved.

Keywords: Neopterin; C-reactive protein; Antibody to cytomegalovirus; CERAD; Alzheimer’s disease

1. Introduction

Conditions of enhanced innate immune response withoverproduction of pro-inflammatory proteins are associ-

0022-3956/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.jpsychires.2006.02.001

* Corresponding author. Tel.: +43 512 504 81608; fax: +43 512 50423638.

E-mail address: imrich.blasko@uibk.ac.at (I. Blasko).

ated with healthy aging and Alzheimer’s disease (AD,Mrak and Griffin, 2005; Blasko et al., 2004). Neopterinis produced by monocyte-derived macrophages uponstimulation with pro-inflammatory cytokine interferon-c (IFNc); its levels increase with age and indicate the sta-tus of activated type I immune response (Murr et al.,2003). Healthy aging substantially differs from AD; how-ever, as age is an important risk factor for sporadic AD,

I. Blasko et al. / Journal of Psychiatric Research 41 (2007) 694–701 695

some immune regulations are gradually present in bothconditions. Increase of neopterin concentration indicatesthe course and progression of human immunodeficiencyvirus type 1 (HIV-1) infection (Wirleitner et al., 2005)and other chronic diseases such as malignant and auto-immune diseases (Murr et al., 1999; Schroecksnadelet al., 2003). In vitro studies suggest that IFNc and otherpro-inflammatory cytokines interact with processing andproduction of b-amyloid (Ab), a key peptide linked toAD pathogenesis (Sastre et al., 2003; Blasko et al.,1999). Neopterin concentrations were found to be higherin AD patients as compared to the healthy age-matchedcontrols (Leblhuber et al., 1999; Hull et al., 2000). How-ever, how concentrations of neopterin might shape thetrajectories of cognitive functioning over the course oftime has not been investigated.

Further studies have implicated herpes viruses, suchas cytomegalovirus (CMV), as potential contributoryor promoting agents of dementia (Strandberg et al.,2004) as higher viral antibody titers were associated withgreater cognitive decline over four years in 1204 MexicanAmericans (Aiello et al., 2005). Infection with CMV iscommon. In developed countries, CMV seroprevalencesteadily increases after infancy, and 10–20% of childrenare infected before puberty. In adults, prevalence rangesfrom 40% to 100% (de Jong et al., 1998). AD patients andnon-demented controls have similar CMV antibody con-centrations (Renvoize and Hambling, 1984). Using poly-merase chain reaction, CMV was detected at ratherhigher frequencies in AD-patients and age-matched nor-mal controls (36% vs 35%; Lin et al., 2002).

Instead of direct measurement of pro-inflammatorycytokines, the levels of which were found to be diminishedor be under detection limit in AD (Richartz et al., 2005)we decided to measure the concentration of inflammatorymarker C-reactive protein (CRP). CRP is a stable mole-cule produced by hepatocytes by stimulation with cyto-kines IL-1, IL-6 and TNFa (Vermeire et al., 2004).

We hypothesised that changes of the adaptiveimmune system play an important role in AD progres-sion. Neopterin as a marker for cell mediated responseshould therefore increase while AD progresses. In thisstudy, 43 AD out-patient from a memory clinic with het-erogeneous cognition status, concentrations of neop-terin, CRP and CMV IgG-antibody were comparedwith seven cognitive parameters at baseline and in fol-low up 15 months later.

2. Subjects and methods

2.1. Study sample and diagnosis of AD

Forty-three AD patients were assessed by standardizeddiagnostic procedures including neurological, psychiatricand general medical examination. The diagnosis of AD

was established according to the National Institute ofNeurologic and Communicative Disorders and Alzhei-mer’s Disease and Related Disorders Association (NIN-CDS-ADRDA) guidelines (McKhann et al., 1984). Theseverity of disease was assessed at baseline by ClinicalDementia Rating (Davis et al., 1990) and all participantsobtained ratings greater or equal to 1.0 point. Exclusioncriteria included diagnoses of post stroke and vasculardementia according to the criteria of the National Insti-tute of Neurological Disorders and Stroke-AssociationInternationale pour la Recherche et l’Enseignement enNeurosciences (NINDS-AIREN, Roman et al., 1993).In addition, patients with other psychiatric disorders suchschizophrenia, bipolar disorder or severe/recurrentdepression as defined by fourth edition of the Diagnosticand Statistical Manual of Mental Disorders (DSM-IV,American Psychiatric Association, 1994) as well aspatients with any diagnosis of traumatic brain injury,dementia with Levy bodies, Parkinson’s disease, andpatients diagnosed as having frontotemporal lobe demen-tia (FTLD) according to the Manchester-Lund criteria(The Lund and Manchester Groups, 1994) were excludedfrom the study. All patients underwent standard bloodlaboratory tests, excluding hypothyroidism, vitamin B12

and folic acid deficiency. Additional cerebrospinal fluidanalysis was performed as the clinical situation indicated.

The comorbidity associated with activation of theimmune system is high in healthy elderly persons andmight even be higher in frail AD patients. Patients wereexcluded from participating in this study if they dis-played any signs of peripheral inflammatory disease atbaseline or two weeks before collection of bloodsamples. The patient and caregiver provided writteninformed consent prior to the study. This study was con-ducted in accordance with the Declaration of Helsinkiand its subsequent amendments (World Medical Associ-ation declaration of Helsinki, 1997).

2.2. Measurement of cognitive changes with time

A trained neuropsychologist administered a standardbattery of neuropsychological tests. The Consortium toEstablish a Registry for Alzheimer’s Disease (CERAD)battery, as described in detail elsewhere, was used formonitoring of cognition ((Welsh et al., 1991; Welshet al., 1992; Welsh et al., 1994), German version(Monsch, 1997; Berres et al., 2000)). This CERAD-bat-tery consists of seven tests: verbal fluency: animal cate-gory, a short form of the Boston naming test (BNT),mini-mental state examination (MMSE; Folstein et al.,1983), verbal memory test (VMT) consisting of word listlearning with immediate recall (sum of 3 trials), VMT-delayed recall and constructional practice (immediateand delayed recall). Psychometric evaluations took placeat baseline and at follow up, which was scheduled for 18months.

-6

-4

-2

01 2 3 4 5 6 7

**

*

*

* *

*

Fig. 1. The calculations of CERAD cognitive battery in the z-scoresduring study period. Bars represent means of z-score values from 43patients at baseline (h) and second measurement ( ). The z-scores ofall cognitive tests lie under the mean of the normative population(defined as 0) and are therefore expressed in negative values (y-axis).The narrative of data at the x-axis is depicted as follows: 1, verbalfluency; 2, Boston naming test; 3, MMSE; 4, verbal memory testimmediate recall; 5, verbal memory test delayed recall; 6, construc-tional practice – total sum; 7, constructional practice – delayed recall;

*p = 0.000–0.041 in paired T-test.

696 I. Blasko et al. / Journal of Psychiatric Research 41 (2007) 694–701

2.3. Quantitation of neopterin, C-reactive protein, CMV

antibody titers and apolipoprotein E phenotype

Serum neopterin concentration was determined byenzyme-linked immunosorbent assay (ELISA,BRAHMS, Berlin, Germany) according to the manufac-turer’s instructions with a detection limit of 2 nM. CRPconcentration was measured by Immunturbidimetricassay (Roche MODULAR, Roche Diagnostics, Mann-heim, Germany). Serum CMV antibody (immunoglobu-lin G) titers were determined by ELISA (Enzygnost [R]anti-CMV/immunoglobulin G; Dade Behring, Vienna,Austria), according to the specifications of the manufac-turer. Apolipoprotein E genotyping was performed witha commercially available DNA assay-kit that uses rapid-cycle PCR and fluorescence resonance energy transferwith the LightCycler (Nauck et al., 2000).

2.4. Acetylcholine-esterase inhibitor treatment

All enrolled patients received acetylcholine-esteraseinhibitor (AChEI) therapy during the study time period,either rivastigmine (up to 12 mg daily), donepezil (up to10 mg) or galantamine (up to 16 mg). The true durationof AChEI treatment in each person was retrospectivelyevaluated with caregiver or patient by completing thefollow up measurements.

2.5. Data analysis

Statistical analysis was performed using the SPSS sta-tistical package, version 11.0. Data collected fromCERAD battery at baseline or follow up were trans-formed to z-scores of the normative sample of 1100 cog-nitively healthy persons (Berres et al., 2000). The z-scoreindicates how many standard deviations an individualperson’s cognitive score is from the mean of the norma-tive sample. z-Scores were used for comparison of theAD group to the normative population. In some calcu-lations, a total sum of all seven CERAD z-scores in eachtime point was also used. For comparison of z-scoresfrom the CERAD battery over time, the paired T-testwas applied. For evaluation of change in time courseD-changes were calculated for all cognitive parametersas well as for neopterin- and CRP-concentrations.D-change is defined as measurement at follow-up minusbaseline measurement. D-change is positive as follow-upvalue increases and vice versa negative as follow-upvalue decreases.

Due to the non-normal distribution of neopterin,CRP and CMV-antibody concentrations, non-paramet-ric tests were required for their analysis. Concentrationsof neopterin and CRP at first and at second measure-ment were compared by means of the Wilcoxon test.Statistical associations between neopterin, CRP, CMV-antibody concentrations and seven cognitive measures

were evaluated using the Spearman rank correlationcoefficient (r). The level of p < 0.05 was considered asstatistically significant.

In a second analysis we studied the combined effect ofseveral covariates on each of the standardised cognitiveparameters using multiple linear regression analysis. Inthese analyses we considered the following independentvariables: neopterin and CRP concentrations (due to theskewed distribution the logarithm of these variables wasused) as well as age at beginning of study, gender,ApoE-e4 presence and time difference between bothmeasurements in the analysis. Significant predictorswere selected using the stepwise backward eliminationmethod. Unfortunately, we were unable to obtain base-line or follow up CRP and CERAD data from all sub-jects; therefore the number of subjects varies between43 and 36 in the data analysis.

3. Results

3.1. Characteristics of study population and cognitivedecline

The study population comprised 30 females and 13males with a mean age of 75.1 years. The mean rawMMSE scores of all 43 AD patients was at baseline18.9 and decreased during the study period to 15.6 points(range 28–4, 30–1, respectively). The mean z-score valuesin each of the CERAD battery tests was below 1 stan-dard deviation of control persons at baseline and furtherdeteriorated during the study period. Paired comparisonof each of the seven psychometric parameters demon-strated significant cognitive decline (Fig. 1, T-test

Table 2BConcentrations of neopterin (nM) and CRP (mg/dl) in AD patientswith baseline MMSE < 20 and MMSE > 20 points

Variable Baseline Follow up p-Value

Group MMSE < 20, n = 21Neopterin 9.4 ± 1.1 15.2 ± 3.3 0.041CRP 0.34 ± 0.09 0.27 ± 0.05 n.s.

Group MMSE > 20, n = 22Neopterin 10.3 ± 1.7 11.9 ± 2.7 n.s.CRP 0.61 ± 0.28 0.29 ± 0.07 n.s.

Mean ± s.e.m.; p-value represents results from Wilcoxon test; n.s. –non-significant.

I. Blasko et al. / Journal of Psychiatric Research 41 (2007) 694–701 697

p = 0.000–0.041). Because of heterogeneous cognitionstatus at baseline, AD-patients were divided into twosubgroups with baseline MMSE more or less than 20(Table 1). For statistical reasons, the median MMSEscore (20) was used as criterion for assigning subjectsto the groups of ‘mild’ dementia (MMSE > 20) or ‘mod-erate/severe’ dementia (MMSE > 20). Using a MMSEscore of 24, only ten subjects would have been in the‘mild’ dementia group.

Upon such a subdivision the groups remained stillcomparable for age, education time and frequency ofApolipoprotein E-e4 (ApoE-e4, Table 1) allele. Of allAD patients, 63% had at least one ApoE-e4 allele. Thenumber of ApoE-e4 allele inversely correlated with thecognitive decline as measured on D-change of z-scoresof CERAD VMT-delayed recall (r = �0.601,p < 0.001, n = 31).

3.2. The concentrations of neopterin and CRP at baseline

and follow up

The mean of time difference between baseline and fol-low up measurements was 14.5 ± 0.5 months and it alsoremained comparable in the subgroups as defined byMMSE 20 points (data not shown; mean ± standarderror of mean). Neopterin concentration at baselinewas 9.8 ± 1.0 nM and significantly increased to13.6 ± 2.1 nM at second measurement in the whole 43AD patient population (p = 0.04, Table 2A). The meanCRP level was low at baseline and further non-signifi-cantly decreased at second measurement (0.46 vs0.28 mg/dl). Fourteen percent of the patients at baselineand 5% of the patients at follow up had CRP concentra-tions higher than 0.07 mg/dl. CRP did not correlate withneopterin concentrations at baseline but significant cor-

Table 1Baseline characteristics of studied AD population

Variable Whole group MMSE > 20 MMSE < 20

Number of patients 43 21 22Male 13 9 4Female 30 12 18Age (years) 75.1 ± 1.2 74.8 ± 1.7 75.5 ± 1.7Education years 9.2 ± 0.4 9.7 ± 0.7 8.7 ± 0.3P1 ApoE-e4 allele (%) 63 72 55

Data are presented as mean ± standard error of mean, s.e.m.

Table 2AConcentrations of neopterin (nM) and CRP (mg/dl) during the studyperiod in the whole investigated population; mean ± s.e.m.

Variable Baseline Follow up p-Value

Neopterin 9.8 ± 1.0 (4.4–39.4) 13.6 ± 2.1 (4.4–63.1) 0.040CRP 0.46 ± 0.1 (0.03–5.5) 0.28 ± 0.04 (0.02–1.4) n.s.

Data in parentheses represents range of displayed values; n = 43, p-value represents results from Wilcoxon test; n.s. – non-significant.

relation was observed at follow up measurement(r = 0.416; p < 0.01; n = 42). Upon subdivision of thestudy sample based on baseline MMSE 20 points, neop-terin increase remained significant only in the groupwith baseline MMSE <20 (p < 0.05, Table 2B). CRPconcentrations did not statistically differ in subgroupsas defined by MMSE 20 points.

3.3. CMV-antibody concentrations correlate with

neopterin but not with D-changes of cognitive parameters

CMV-antibody concentrations were measured atbaseline and correlated with corresponding neopterinconcentrations (r = 0.386, p = 0.016, n = 43). The corre-lation of CMV-antibody with baseline CRP, however,failed to reach statistical significance (r = 0.305, p =0.053, n = 41). Compared with baseline data of sevenCERAD z-scores or with their D-changes in the courseof time no statistical correlation in respect to CMV-anti-body concentrations was observed (data not shown).

3.4. Acetylcholine-esterase inhibition therapy and CRP

plasma concentrations

Although AD patients were treated with acetylcho-line-esterase inhibitors (AChEI) for the mean time13.6 ± 0.6 months, they significantly declined in MMSEscores. The duration of AChEI therapy also remainedcomparable in the subgroups as defined by MMSE 20points (data not shown). Because AChEI treatmentcould influence immune parameters we were interestedif D-change of CRP and neopterin correlate with dura-tion of AChEI therapy. The duration of AChEI therapyin months negatively correlated with D-change of CRPconcentrations (r = �0.412, p < 0.01, n = 40, Fig. 2)and positively with D-change of neopterin (r = 0.322,p < 0.05, n = 41). Subdivided with respect to the base-line MMSE score 20, the negative correlation ofD-change of CRP and duration of AChEI therapyremained significant only in the group of MMSE <20(r = �0.569; p < 0.01, n = 21). In these subgroups, nocorrelation according to neopterin concentration wasfound.

change of neopterin in time course (nM)

6050403020100-10-20

chan

ge o

f M

MSE

z-s

core

s in

tim

e co

urse

6

4

2

0

-2

-4

-6

-8

Fig. 3. Scatter plot of D-change of neopterin concentrations versus D-change of MMSE scores during studied time period; r = �0.401,p < 0.01 in Spearman correlation.

change of CRP in time course (mg/dl)

1.0.50.0-.5-1.0-1.5-2.0

dura

tion

of A

ChE

I tre

atm

ent (

mon

ths)

30

20

10

0

Fig. 2. Scatter plot of changes in serum CRP concentrations versusduration of Acetylcholine-esterase inhibition (AChEI) therapy;r = �0.412, p < 0.01 in Spearman correlation.

698 I. Blasko et al. / Journal of Psychiatric Research 41 (2007) 694–701

3.5. Comparison of neopterin and CRP concentrations

with CERAD cognitive battery

The baseline, follow up or D-changes of CRP con-centrations did not correlate with the correspondingCERAD z-scores or with their D-changes in the time per-iod. In the first calculation with neopterin we tested thehypothesis if it could predict the rate of cognitive decline.When compared to baseline neopterin, from sevenz-scores of CERAD cognitive battery, only D-change inverbal fluency inversely correlated with the concentra-tions of neopterin (r = �0.322, p < 0.05, n = 39).

In a second set of calculation, correlation analysis ofthe D-change of neopterin concentrations with theD-change of each cognitive parameter in CERADz-scores was performed. Increase of neopterin correlatedwith higher loss in MMSE z-scores (r = �0.401, p < 0.01,n = 42; Fig. 3) as well as with greater loss in the sum of allseven CERAD z-scores (r = �0.353, p < 0.05, n = 36).No further CERAD battery subscores correlated withneopterin or CRP concentrations.

Finally, multiple regression analysis considering theinfluence of D-changes of neopterin and CRP on cogni-tive CERAD battery controlled for age, gender, ApoE-e4presence and time interval between both measurementswas performed. D-change of z-scores in verbal fluency,BNT, MMSE, VMT-immediate and VMT-delayedrecall, constructional practice immediate or delayedrecall and the total sum of all seven CERAD z-scoreswere considered as dependent variables. The regressionwith variable MMSE, constructive praxis immediaterecall and sum of all CERAD subscores were signifi-cantly associated with an increase of neopterin (p =0.009–0.036) and surprisingly also with a decrease of

CRP (p = 0.008–0.037). Interestingly, significant associ-ation with dependent variables VMT-immediate recalland VMT-delayed recall with the presence of oneApoE-e4 allele was found (p = 0.038 and p = 0.003,respectively). The dependent variable VMT-immediaterecall was significantly associated with age and timedifference between both measurements (p < 0.01 forboth).

4. Discussion

In this AD out-patient study the levels of neopterinincreased as cognition declined. The mean neopterinconcentration in this patient sample was comparablewith other AD patients and control populations (Lebl-huber et al., 1999; Hull et al., 2000). The concentrationof neopterin and also almost that of CRP correlatedwith CMV antibody concentration indicating accord-ingly the state of immune system activation. We foundthat rather cumulative cognitive parameters such asMMSE and sum of CERAD scores correlated with theincrease of neopterin in comparison to single cognitivedomains. Subdividing the group based on the baselineMMSE, the increase of neopterin remained significantonly in the more deteriorated group. This is in agree-ment with study of Casal et al. (2003), showing thatthe neopterin increased only in the subgroup of moreseverely demented patients as defined by the cumulativeclinical scale, global deterioration scale. The changeover time in a susceptible psychometric parameter forAD such as verbal memory test immediate and delayedrecall (Knopman and Ryberg, 1989) associated in multi-ple testing with ApoE-e4 distribution.

I. Blasko et al. / Journal of Psychiatric Research 41 (2007) 694–701 699

The correlation of baseline neopterin with worseningof cognitive domain verbal fluency over time (D-change)is intriguing; however, multiple testing did not supportthe association. Multiple testing confirmed the associa-tion of neopterin with dependent variable MMSE andthe sum of all CERAD subscores. Interestingly, thedecrease of CRP concentrations was associated withthe same cognitive domains as neopterin although thisassociation was not detected in paired comparison. Wesuggest that this might rather reflect the improvedperipheral inflammatory conditions (CRP decrease) inour AD patients or secondly the anti-inflammatoryeffect of AChEI therapy (Reale et al., 2004; Realeet al., 2005). The last could be due to in vitro observa-tion in which peripheral blood mononuclear cells(PBMC) from AD patients treated with AChEI showeddownregulated production of IL-1a, IL-6 and TNFa(Reale et al., 2004), which all stimulate CRP production(Vermeire et al., 2004). Accordingly, in our AD patientstreated with AChEI an inverse correlation betweenduration of treatment with AChEI and CRP was found.These results were again limited to the more deterioratedgroup (MMSE < 20).

In contrast to previous studies, cognitive decline ofAD-patients did not correlate with CMV antibody con-centration at baseline nor at follow up (Lin et al., 2002;Strandberg et al., 2004). This fact could be due to thelow number of cases in our study or short time intervalbetween both measurements.

To distinguish between activation of innate oracquired immune system based on measurements ofCRP and neopterin is virtually impossible; however, incontrast to CRP, neopterin seems to be more associatedwith cell-mediated immune response. In a previousstudy, negative correlation of an acute phase proteina1-antichymotrypsin (ACT) with concentration ofneopterin in AD patients was described (Licastro,2001). ACT production is stimulated by inflammatorycytokine IL-1b (Kordula et al., 1998), and this cytokinestimulates also the CRP production. Therefore, oppositeregulation of CRP and neopterin concentrations in thestudy period might suggests an increase in activity of cellmediated immunity.

Neopterin is produced by monocyte-derived macro-phages upon stimulation with IFNc, which is producedmainly by activated T-cells. The strength of our studymight lie in the fact that our AD-patients were withoutsignificant periphery inflammation as documented bylow CRP levels. In AD patients, an impairment of mito-genic activation (Stieler et al., 2001), calcium response(Sulger et al., 1999) and higher telomere shortening(Panossian et al., 2003) of PBMC and T cells havealready been demonstrated. Independently, longitudinalstudies of the healthy elderly population have led to theemerging concept of an ‘‘immune risk phenotype’’,which was itself found to be associated with CMV sero-

positivity (Pawelec et al., 2005). The negative impact ofCMV seropositivity on survival in the elderly may bedue, in part, to congestion of the immune system withclonally expanded, highly differentiated, non-functionalCMV-specific T cells. Similarly to the immune cells, sucha congestion might occur also in brain cells as in a his-topathological study 36% of AD brain specimens werepositive for CMV (Lin et al., 2002).

Our AD patients were in 70% of the cases positive forCMV antibody and CMV antibody concentrations cor-related with concentrations of neopterin and CRP.CMV infection itself leads to an increase in the numberof CD8 + T cells, which afterward produce IFNc butlack substantial growth potential (Almanzar et al.,2005). Similarly, in response to influenza vaccinationthe predominance of CD8 + T cells and their clonalexpansion was described (Saurwein-Teissl et al., 2002).These cells again were unable to proliferate, butin vitro produced high amounts of IFNc. Therefore,we suggest that the enhanced neopterin productionobserved in our moderately to severely impaired ADpatients might be a consequence of senescence ofCMV-expanded T cell clones in the investigated popula-tion. These T cells could in turn produce IFNc and as aresult, we observed the neopterin increase. Further stud-ies with CMV proliferating T cell clones isolated fromAD patients are in progress in our laboratory.

Based on the above-mentioned discussion one couldconsider that neopterin increase in the studied AD pop-ulation is solely due to high seropositivity for CMV. Asa result, neopterin increase observed in the course ofcognitive decline would be accidental and independent.This question still needs to be clarified.

Another possibility for explaining the link betweenincreased neopterin production and cognitive deteriora-tion of AD patients could be through the influencing ofAb production. IFNc is in vitro a strong stimulator ofAb40 and Ab42 production in neuronal and extraneuro-nal cells (Sastre et al., 2003; Blasko et al., 1999) andincreased peripheral production of Ab could vice versainfluence the central content of Ab in brain (Mackicet al., 2002; Zlokovic, 2005). However, plasma Ab42decreases during disease progression (Mayeux et al.,2003) and in recent studies neopterin levels did not cor-relate with the plasma Ab40 or Ab42 levels betweenyoung and old patients with Down Syndrome (Mehtaet al., 2005). It was suggested that the increase of neop-terin through age reflects inflammatory cell activationrather than AD neuropathology.

In summary, we provide the evidence that increasedconcentration of neopterin in moderately to severelydeteriorated AD patients might be due to the high prev-alence of CMV infection in the investigated population.We propose that the mechanism underlying thesechanges is due to the altered immune phenotype duringsenescence.

700 I. Blasko et al. / Journal of Psychiatric Research 41 (2007) 694–701

Acknowledgements

We acknowledge Edith Stadelmann, Thomas Bodner,and Thomas Walch for their assistance in collectingblood samples and neuropsychological data, AstridHaara’s excellent technical assistance, and HartmannHinterhuber for his continuous support of our work.This work was supported by the Austrian Federal Minis-try of Social Affairs and Generations and by the ‘‘Land-esgedachtnisstiftung’’ of the Tyrolean government.

References

Aiello AE, Hann MN, Kumar S, Moore KA, Blythe l. Novel infectiousbiomarkers of cognitive impairment. In: 1st International Confer-ence on Prevention of Dementia, Washington DC, June 18–21,2005.

Almanzar G, Schwaiger S, Jenewein B, Keller M, Herndler-Brand-stetter D, Wurzner R, Schonitzer D, Grubeck-Loebenstein B.Long-term cytomegalovirus infection leads to significant changes inthe composition of the CD8 + T-cell repertoire, which may be thebasis for an imbalance in the cytokine production profile in elderlypersons. Journal of Virology 2005;79:3675–83.

American Psychiatric Association. Diagnostic and statistical manualof mental disorders. 4th ed. Washington DC: American Psychiat-ric Association; 1994.

Berres M, Monsch AU, Bernasconi F, Thalmann B, Stahelin HB.Normal ranges of neuropsychological tests for the diagnosis ofAlzheimer’s disease. Studies in Health Technology and Informatics2000;77:195–9.

Blasko I, Marx F, Steiner E, Hartmann T, Grubeck-Loebenstein B.TNFalpha plus IFNgamma induce the production of Alzheimerbeta-amyloid peptides and decrease the secretion of APPs. FASEBJournal 1999;13:63–8.

Blasko I, Stampfer-Kountchev M, Robatscher P, Veerhuis R, Eike-lenboom P, Grubeck-Loebenstein B. How chronic inflammationcan affect the brain and support the development of Alzheimer’sdisease in old age: the role of microglia and astrocytes. Aging Cell2004;3:169–76.

Casal JA, Robles A, Tutor JC. Serum markers of monocyte/macro-phage activation in patients with Alzheimer’s disease and othertypes of dementia. Clinical Biochemistry 2003;36:553–6.

Davis PB, Morris JC, Grant E. Brief screening tests versus clinicalstaging in senile dementia of the Alzheimer type. Journal of theAmerican Geriatrics Society 1990;38:129–35.

de Jong MD, Galasso GJ, Gazzard B, Griffiths PD, Jabs DA, KernER, Spector SA. Summary of the II international symposium oncytomegalovirus. Antiviral Research 1998;39:141–62.

Folstein MF, Robins LN, Helzer JE. The mini-mental state examina-tion. Archives of General Psychiatry 1983;40:812.

Hull M, Pasinetti GM, Aisen PS. Elevated plasma neopterin levels inAlzheimer disease. Alzheimer Disease and Associated Disorders2000;14:228–30.

Knopman DS, Ryberg S. A verbal memory test with high predictiveaccuracy for dementia of the Alzheimer type. Archives of Neurol-ogy 1989;46:141–5.

Kordula T, Rydel RE, Brigham EF, Horn F, Heinrich PC, Travis J.Oncostatin M and the interleukin-6 and soluble interleukin-6receptor complex regulate a1-antichymotrypsin expression inhuman cortical astrocytes. Journal of Biological Chemistry1998;273:4112–8.

Leblhuber F, Walli J, Demel U, Tilz GP, Widner B, Fuchs D.Increased serum neopterin concentrations in patients with

Alzheimer’s disease. Clinical Chemistry and Laboratory Medi-cine 1999;37:429–31.

Licastro F. Genetic background of inflammatory molecules affectsprotein expression, the risk of developing Alzheimer’s disease andcognitive decline. Archives of Gerontology and Geriatrics2001(Suppl. 7):227–34.

Lin WR, Wozniak MA, Cooper RJ, Wilcock GK, Itzhaki RF.Herpesviruses in brain and Alzheimer’s disease. Journal ofPathology 2002;197:395–402.

The Lund and Manchester Groups. Clinical and neuropathologicalcriteria for frontotemporal dementia. Journal of Neurology Neu-rosurgery and Psychiatry 1994;57:416–8.

Mackic JB, Bading J, Ghiso J, Walker L, Wisniewski T, Frangione B,Zlokovic BV. Circulating amyloid-beta peptide crosses the blood-brain barrier in aged monkeys and contributes to Alzheimer’sdisease lesions. Vascular Pharmacology 2002;38:303–13.

Mayeux R, Honig LS, Tang MX, Manly J, Stern Y, Schupf N, MehtaPD. Plasma A[beta]40 and A[beta]42 and Alzheimer’s disease:relation to age, mortality, and risk. Neurology 2003;61:1185–90.

McKhann G, Drachman D, Folstein M, Katzman R, Price D, StadlanEM. Clinical diagnosis of Alzheimer’s disease: report of theNINCDS-ADRDA Work Group under the auspices of Depart-ment of Health and Human Services Task Force on Alzheimer’sDisease. Neurology 1984;34:939–44.

Mehta PD, Patrick BA, Dalton AJ, Patel B, Mehta SP, Pirttila T,Coyle PK. Increased serum neopterin levels in adults with Downsyndrome. Journal of Neuroimmunology 2005;164:129–33.

Monsch AU. Neuropsychological examination in evaluating dementia.Schweizerische Rundschau fur Medizin Praxis 1997;86:1340–2.

Mrak RE, Griffin WS. Glia and their cytokines in progression ofneurodegeneration. Neurobiology of Aging 2005;26:349–54.

Murr C, Fuith LC, Widner B, Wirleitner B, Baier-Bitterlich G, FuchsD. Increased neopterin concentrations in patients with cancer:indicator of oxidative stress? Anticancer Research 1999;19:1721–8.

Murr C, Hainz U, Asch E, Berger P, Jenewein B, Saurwein-Teissl M,Grubeck-Loebenstein B, Fuchs D. Association of increased neop-terin production with decreased humoral immunity in the elderly.Experimental Gerontology 2003;38:583–7.

Nauck M, Hoffmann MM, Wieland H, Marz W. Evaluation of the apoE genotyping kit on the LightCycler. Clinical Chemistry2000;46:722–4.

Panossian LA, Porter VR, Valenzuela HF, Zhu X, Reback E,Masterman D, Cummings JL, Effros RB. Telomere shortening inT cells correlates with Alzheimer’s disease status. Neurobiology ofAging 2003;24:77–84.

Pawelec G, Akbar A, Caruso C, Solana R, Grubeck-Loebenstein B,Wikby A. Human immunosenescence: is it infectious? Immuno-logical Reviews 2005;205:257–68.

Reale M, Iarlori C, Gambi F, Feliciani C, Salone A, Toma L, DeLucaG, Salvatore M, Conti P, Gambi D. Treatment with an acetylcho-linesterase inhibitor in Alzheimer patients modulates the expressionand production of the pro-inflammatory and anti-inflammatorycytokines. Journal of Neuroimmunology 2004;148:162–71.

Reale M, Iarlori C, Gambi F, Lucci I, Salvatore M, Gambi D.Acetylcholinesterase inhibitors effects on oncostatin-M, interleu-kin-1 beta and interleukin-6 release from lymphocytes of Alzhei-mer’s disease patients. Experimental Gerontology 2005;40:165–71.

Renvoize EB, Hambling MH. Cytomegalovirus and Alzheimer’sdisease. Age Ageing 1984;13:205–9.

Richartz E, Stransky E, Batra A, Simon P, Lewczuk P, Buchkremer G,Bartels M, Schott K. Decline of immune responsiveness: apathogenetic factor in Alzheimer’s disease? Journal of PsychiatricResearch 2005;39:535–43.

Roman GC, Tatemichi TK, Erkinjuntti T, Cummings JL, Masdeu JC,Garcia JH, et al.. Vascular dementia: diagnostic criteria forresearch studies: report of the NINDS-AIREN InternationalWorkshop. Neurology 1993;43:250–60.

I. Blasko et al. / Journal of Psychiatric Research 41 (2007) 694–701 701

Sastre M, Dewachter I, Landreth GE, Willson TM, Klockgether T,Van Leuven F, Heneka MT. Nonsteroidal anti-inflammatory drugsand peroxisome proliferator-activated receptor-gamma agonistsmodulate immunostimulated processing of amyloid precursorprotein through regulation of beta-secretase. Journal of Neurosci-ence 2003;23:9796–804.

Saurwein-Teissl M, Lung TL, Marx F, Gschosser C, Asch E, Blasko I,Parson W, Bock G, Schonitzer D, Trannoy E, Grubeck-Loeben-stein B. Lack of antibody production following immunization inold age: association with CD8(+)CD28(�) T cell clonal expansionsand an imbalance in the production of Th1 and Th2 cytokines.Journal of Immunology 2002;168:5893–9.

Schroecksnadel K, Kaser S, Ledochowski M, Neurauter G, Mur E,Herold M, Fuchs D. Increased degradation of tryptophan in bloodof patients with rheumatoid arthritis. Journal of Rheumatology2003;30:1935–9.

Stieler JT, Lederer C, Bruckner MK, Wolf H, Holzer M, GertzHJ, Arendt T. Impairment of mitogenic activation of peripheralblood lymphocytes in Alzheimer’s disease. Neuroreport 2001;12:3969–72.

Strandberg TE, Pitkala KH, Linnavuori K, Tilvis RS. Cognitiveimpairment and infectious burden in the elderly. Archives ofGerontology and Geriatrics Supplement 2004:419–23.

Sulger J, Dumais-Huber C, Zerfass R, Henn FA, Aldenhoff JB.The calcium response of human T lymphocytes is decreased in

aging but increased in Alzheimer’s dementia. Biological Psychi-atry 1999;45:737–42.

Vermeire S, Van Assche G, Rutgeerts P. C-reactive protein as a markerfor inflammatory bowel disease. Inflammatory Bowel Diseases2004;10:661–5.

Welsh K, Butters N, Hughes J, Mohs R, Heyman A. Detection ofabnormal memory decline in mild cases of Alzheimer’s diseaseusing CERAD neuropsychological measures. Archives of Neurol-ogy 1991;48:278–81.

Welsh KA, Butters N, Hughes JP, Mohs RC, Heyman A. Detectionand staging of dementia in Alzheimer’s disease. Use of theneuropsychological measures developed for the Consortium toestablish a registry for Alzheimer’s Disease. Archives of Neurology1992;49:448–52.

Welsh KA, Butters N, Mohs RC, Beekly D, Edland S, Fillenbaum G,Heyman A. The Consortium to establish a registry for Alzheimer’sDisease (CERAD). Part V. A normative study of the neuropsy-chological battery. Neurology 1994;44:609–14.

Wirleitner B, Schroecksnadel K, Winkler C, Fuchs D. Neopterin inHIV-1 infection. Molecular Immunology 2005;42:183–94.

World Medical Association declaration of Helsinki. Recommenda-tions guiding physicians in biomedical research involving humansubjects. JAMA 1997;277:925–6.

Zlokovic BV. Neurovascular mechanisms of Alzheimer’s neurodegen-eration. Trends Neuroscience 2005;28:202–8.