Journal of Alzheimer’s Disease 21 (2010) 1383–1392 1383DOI 10.3233/JAD-2010-100780IOS Press

Vitamin E and Enzymatic/OxidativeStress-Driven Oxysterols in Amnestic MildCognitive Impairment Subtypes andAlzheimer’s Disease

Luigi Iulianoa,∗, Roberto Monticoloa, Giuseppe Strafacea, Ilaria Spoletinib, Walter Giannic,Carlo Caltagironeb,d, Paola Bossub and Gianfranco Spallettab

aVascular Biology and Mass Spectrometry Laboratory, Department of Experimental Medicine, SapienzaUniversity, Rome, ItalybLaboratory of Clinical and Behavioral Neurology, IRCCS Santa Lucia Foundation, Rome, ItalycUnit of Geriatric Oncology, INRCA; Rome, ItalydDepartment of Neuroscience, Tor Vergata University, Rome,Italy

Accepted 27 May 2010

Abstract. Oxidative stress, which contributes to neuronal damage, is thought to be a pathophysiological mechanism of Alzheimer’sdisease (AD). Markers of oxidative stress may appear early in the preclinical, mild cognitive impairment (MCI) phase ofAD. Weinvestigated the interaction among enzymatic-derived oxysterols (24S-hydroxycholesterol and 27-hydroxycholesterol), markersof oxidative stress, including free radical-related oxysterols (7β-hydroxycholesterol and 7-ketocholesterol), and vitaminE inAD patients and two amnestic MCI subtypes, amnestic single-domain MCI (a-MCI) subjects, and multi-domain MCI (md-MCI)subjects, compared to healthy control subjects (HC). The study included 37 patients with AD, 24 with a-MCI, 29 with md-MCI,and 24 HC. Plasma assessments were made using isotope dilution-mass spectrometry.Although we found no significant change in free radical- or enzymatic-derived oxysterol concentrations in AD or MCI patients,vitamin E levels corrected for cholesterol were reduced in AD patients compared to HC. Results suggest that AD patients haveupregulated cerebral oxidative stress or a nutritional deficit of vitamin E. The oxysterols investigated here are not useful markersfor diagnosing AD or MCI.

Keywords: Alzheimer’s disease, amnestic, cholesterol, mild cognitive impairment, oxysterols, vitamin E

INTRODUCTION

Alzheimer’s disease (AD) is the most frequent formof neurodegeneration associated with dementia in theelderly. The prevalence of this disease has created aworldwide socio-economic challenge and affects ap-proximately 4 million individuals in the United States

∗Correspondence to: Luigi Iuliano, M.D., Sapienza Universityof Rome, Department of Experimental Medicine, Unit of VascularMedicine, Vascular Biology & Mass Spectrometry Lab, Corso dellaRepubblica, 79 04100 Latina, Italy. Tel.: +39 0773 474098; Fax:+39 06 62 29 1089; E-mail: luigi.iuliano@uniroma1.it.

and 16 million worldwide. The incidence of AD dou-bles every 5 years after age 65 [1].

AD is influenced by both genetic and epigeneticrisk factors. The strongest genetic risk factor is thepresence of the epsilon4 allele of the apolipoprotein E(APOE) gene, which encodes a protein that has a cru-cial role in cholesterol metabolism. Evidence showsthat APOE4 contributes to AD pathogenesis by mod-ulating the metabolism and aggregation of amyloid-β

(Aβ) peptides in connection with the regulation of lipidmetabolism [2] particularly cholesterol metabolism inthe brain [3].

As a major building block of the mammalian cellmembrane, cholesterol has an important role in the cre-

ISSN 1387-2877/10/$27.50 2010 – IOS Press and the authors. All rights reserved

1384 L. Iuliano et al. / Vitamin E and Oxysterols in MCI and AD

ation of lipid rafts, which are involved in the traffickingof Aβ in the cell membrane [4]. In fact, some actions ofcholesterol may be mediated by the oxidized productsof Aβ. Cholesterol acts as a signaling molecule in neu-ronal functioning and memory and is implicated in neu-rodegeneration via oxidative stress-mediated mecha-nisms [4–7]. Upregulated oxidative stress is reportedlyincreased in mild cognitive impairment (MCI) patientswho are considered at higher risk of developing AD [8],coherently with the notion that oxidative stress couldbe an early event in the pathophysiology of AD [9].There are also reports that Aβ catalyzes the conversionof cholesterol to 7β-hydroxycholesterol (7β-OHC), ahighly neurotoxic [10] oxysterol that may participatein mechanisms of neurodegeneration. 7β-OHC caus-es rearrangement of the liquid-ordered phase [11,12],which results in the creation of lipid rafts, and is a po-tent inhibitor ofα-protein kinase C, an enzyme criticalfor memory consolidation and synaptic plasticity andimplicated in AD [10]. Furthermore 5,6-secosterol, anewly discovered cholesterol ozonolysis product pro-duced during inflammation, has been detected in thehuman brain; it covalently modifies Aβ and acceleratesamyloidogenesisin vitro [13].

Cholesterol is unable to pass the blood-brain barrierand its level in the brain depends exclusively onde no-vo synthesis and elimination. These two processes arecontrolled by oxysterols, which are oxygenated deriva-tives of cholesterol produced both enzymatically andin a free radical-mediated reaction [6].

Among the enzymatic-driven oxysterols, 24S-hydr-oxycholesterol (24S-OHC) and 27-hydroxycholesterol(27-OHC) are able to pass the blood-brain barrier andcontrol brain cholesterol levels [6]. The enzymatic for-mation of 24S-OHC is the most important mechanismfor eliminating cholesterol from the central nervoussystem [14].

Numerous studies have demonstrated elevated mark-ers of oxidative stress, including lipid peroxidation,protein, and nucleic acid oxidation, in the cerebrospinalfluid and blood of AD patients [15]. Free radical-drivenoxysterols, including 7β-OHC and 7-ketocholesterol(7-KC), are validated markers of lipid peroxidation;they have been described as elevated in several condi-tions associated with increased oxidative stress, suchas the vascular disorders [16] associated with AD [17].

Evidence shows that vitamin E is reduced in AD andMCI [18–20] and that high antioxidant intake (vitaminE, vitamin C and carotene) may delay cognitive declinein the elderly [21]. Although the neuroprotective roleof the antioxidant vitamin E has been tested in clini-

cal trials, a recent Cochrane systematic review showedno evidence of the efficacy of vitamin E in preventingAD or MCI or in treating patients with these illness-es and emphasized the need for additional research toidentify the role of vitamin E in managing cognitiveimpairment [22].

To our knowledge, the interplay between enzymatic-and free-radical-derived oxysterols and between the lat-ter and vitamin E levels has yet to be investigated inAD. Therefore, the present study investigated whetheroxysterols and vitamin E are altered in AD and MCIpatientsin vivo and whether these abnormalities areassociated with each other and with the state of thedisease.

MATERIALS AND METHODS

Subjects

We consecutively recruited 37 patients with a diag-nosis of probable AD, 24 patients with a diagnosis ofamnestic single-domain MCI (a-MCI), and 29 patientswith a diagnosis of amnestic multi-domain MCI (md-MCI) in outpatient clinics in central Italy. We recruit-ed 24 healthy controls (HC) in the same geographicalarea after their medical history had been reviewed. Allsubjects (or their proxies) signed written informed con-sent forms. The study procedures were developed ac-cording to the guidelines of the Santa Lucia FoundationEthics Committee, which approved the protocol, andthe Helsinki Declaration of 1975.

Inclusion and exclusion criteria

Inclusion criteria for the two MCI groups were: 1) di-agnostic evidence of a-MCI based on the guidelines ofPetersen et al. [23]: i) complaint of defective memory,ii) normal activities of daily living, iii) normal generalcognitive functioning, iv) abnormal memory function-ing for age, and v) absence of dementia; and 2) a Clini-cal Dementia Rating [24] score= 0.5. Specifically, di-agnoses of a-MCI or md-MCI were made according tothe guidelines of Winblad et al. [25]: if patients’ non-memory domains were intact, they were classified asa-MCI; if they had deficits in one or more non-memorydomains, they were diagnosed as md-MCI.

Inclusion criteria specific for AD were based on theindications of the NINCDS-ADRDA work group [26].All patients were in the initial stages of cognitive im-pairment; they had not taken psychotropic or acetil-

L. Iuliano et al. / Vitamin E and Oxysterols in MCI and AD 1385

cholinesterase inhibitor drugs for at least 6 months andhad undergone their first clinical examination for thediagnosis of AD or MCI.

Common exclusion criteria for the three patientgroups and the HC group were: 1) major medical ill-nesses (i.e., diabetes, obstructive pulmonary disease orasthma); hematological/oncological disorders; vitaminB12 or folate deficiency, as evidenced by blood concen-trations below the lower limits of the reference inter-vals; pernicious anemia; clinically significant and un-stable active gastrointestinal, renal, hepatic, endocrineor cardiovascular system diseases; newly treated hy-pothyroidism; 2) comorbidity of primary psychiatric orneurological disorders (i.e., schizophrenia, major de-pression, stroke, Parkinson’s disease, seizure disorder,head injury with loss of consciousness) and any othersignificant mental or neurological disorder; 3) knownor suspected history of alcoholism or drug dependenceand abuse during lifetime; 4) magnetic resonance imag-ing evidence of focal parenchymal abnormalities; and5) magnetic resonance imaging evidence of neoplasm.A specific exclusion criterion for HC was use of antiox-idant and hypolipemic drugs because of their effects onoxidative stress [27].

Diagnostic and cognitive examination

A trained clinical neurologist interviewed patientsand caregivers using the above-mentioned criteria forthe diagnosis of AD or MCI. All patients and HCunderwent a thorough neuropsychological assessment.To obtain a global index of cognitive impairment,we administered the Mini Mental State Examination(MMSE) [28]. To assess individual cognitive do-mains, we administered the Mental Deterioration Bat-tery (MDB) [29], which is a standardized and vali-dated neuropsychological battery composed of sevenneuropsychological tests: verbal memory (MDB Rey’s15-word Immediate Recall and Delayed Recall); short-term visual memory (MDB Immediate Visual Memo-ry); logical reasoning (MDB Raven’s Progressive Ma-trices’ 47); language (MDB Phonological Verbal Flu-ency, and MDB Sentence Construction); simple con-structional praxis (MDB Copying Drawings, and MDBCopying Drawings with Landmarks). To assess othercognitive domains, we administered the following testsin addition to the MDB: complex constructional prax-is (Copy of Rey-Osterrieth picture) [30,31]; long-termvisual memory (Delayed Recall of Rey-Osterrieth pic-ture) [30,31]; and frontal attention shifting and controlabilities (Stroop test interference time) [32].

Laboratory methods

Peripheral venous blood was obtained from partici-pants after overnight fasting. The plasma or serum wasthen removed and stored at−80◦C until assay.

For cholesterol oxidation markers, we measured 7β-OHC and 7-KC, which are formed by a free radicalmechanism, and the two enzymatic oxysterols, 24S-OHC and 27-OHC, which are formed by CYP46 andCYP27 respectively. To assess vitamin E status wemeasuredα-tocopherol, the main and most active vi-tamin E stereoisomer. Oxysterols and vitamin E wereanalyzed by mass spectrometry using isotope dilutionmethods, as previously reported [33–35]. Because 7β-OHC and 7-KC undergo mutual interconversionin vi-vo [36] and oxysterols and vitamin E are associatedwith cholesterol [37], results are also presented as rela-tive values corrected for cholesterol concentration. To-tal cholesterol and triglycerides were analyzed usingcommercial kits, as previously reported [38].

Statistical analysis

Comparison among groups for the gender variablewas performed using theχ2 test. Comparisons amonggroups for continuous sociodemographic characteris-tics (i.e., age and educational level) and global cogni-tive level (i.e., MMSE score) were made using a seriesof factorial analyses of variance (ANOVA) followed byFisher’s Protected Least Significant Difference (PLSD)post-hoc test.

Comparisons among groups on the crude values ofoxysterols and Vitamin E concentrations were per-formed using a series of ANOVAs.

To obtain values of oxysterols and vitamin E con-centrations corrected for cholesterol, we first carriedout among-group comparisons using overall multivari-ate analyses of variance (MANOVA) to minimize thelikelihood of type-I errors. In this analysis, the valuesof oxysterols and vitamin E concentrations correctedfor cholesterol (ratios) were used as dependent vari-ables and groups were used as independent variables.If the model was significant, ANOVAs followed byFisher’s Protected Least Significant Difference (PLSD)post-hoctests were performed to measure differencesamong groups. The level of statistical significance usedthroughout the study was defined asp < 0.05.

RESULTS

Table 1 shows the sociodemographic characteristicsof the study population. The a-MCI, md-MCI, and HC

1386 L. Iuliano et al. / Vitamin E and Oxysterols in MCI and AD

Tabl

e1

Soc

iode

mog

raph

ican

dco

gniti

vech

arac

teris

tics

ofa-

MC

I,m

d-M

CI,

AD

and

HC

grou

ps∗

Cha

ract

eris

tics

HC

a-M

CI

md-

MC

IA

DA

NO

VAA

Dvs

.A

Dvs

.A

Dvs

.a-M

CIv

s.a-

MC

Ivs.

md-

MC

Ivs.

(n=

24)

(n=

24)

(n=

29)

(n=

37)

(df=

3.11

)a-

MC

Im

d-M

CI

HC

md-

MC

IH

CH

C

Age

(yea

rs)

69.8

3±6.

468

.42±

5.4

70.8

6±6.

676

.03±

7.9

F7.

54p0

.000

1p§

<0.

0001

p§0.

0029

p§0.

0008

p§0.

1969

p§0.

4737

p§0.

5861

Edu

catio

nall

evel

(yea

rs)

11.2

5±4.

311

.0±

4.5

6.69

±3.

55.

54±

4.3

14.0

3<

0.00

01<

0.00

010.

2713

<0.

0001

0.00

030.

8367

0.00

01M

MS

Esc

ore

28.9

2±1.

428

.78±

1.2

26.2

8±1.

918

.05±

7.2

44.3

9<

0.00

01<

0.00

01<

0.00

01<

0.00

010.

0413

0.91

600.

0298

Mal

ege

nder

n(%

)9

(37.

5)18

(78)

12(4

1)10

(27)

χ2:

15.8

08df

=3

p=

0.00

12S

tatin

use

n(%

)0

(0)

3(1

2.5)

6(2

0.7)

6(1

6.2)

χ2:

5.38

8df

=3

p=

0.14

5∗P

lus-

min

usva

lues

deno

tes

mea

n±

SD

.§F

ishe

rpo

st-h

octe

st.

a-M

CI:

amne

stic

sing

le-d

omai

nm

ildco

gniti

veim

pairm

ent;

md-

MC

I:am

nest

icm

ulti-

dom

ain

mildco

gniti

veim

pairm

ent;

AD

:Alz

heim

er’s

dise

ase;

HC

:hea

lthy

cont

rols

;df:

degr

ees

offr

eedo

m;M

MS

E:M

ini-M

enta

lSta

teE

xam

inat

ion.

Sig

nific

ant

pva

lues

(p<

0.05

)ar

ein

dica

ted

inbo

ld.

L. Iuliano et al. / Vitamin E and Oxysterols in MCI and AD 1387

Table 2Crude values of biochemical routine variable concentrations in a-MCI, md-MCI, AD and HC groups

HC (n = 24) a-MCI (n = 24) md-MCI (n = 29) AD (n = 37) ANOVA (df = 3.110)Biochemical parameters mean± SD mean± SD mean± SD mean± SD F p

Total Cholesterol (mg/dL) 200.62± 37.69 200.96± 53.45 202.03± 40.25 209.84± 41.33 0.374 0.7716LDL Cholesterol (mg/dL) 126.54± 27.96 123.96± 36.47 122.00± 31.39 128.00± 32.69 0.213 0.8872HDL Cholesterol (mg/dL) 50.92± 12.02 48.96± 10.71 51.97± 11.57 53.86± 12.87 0.867 0.4608Triglycerides (mg/dL) 116.17± 49.12 139.62± 83.10 140.45± 77.38 139.86± 69.75 0.970 0.4095Glucose (mg/dL) 88.29± 11.80 90.04± 13.03 96.38± 14.84 92.16± 23.49 1.083 0.3594Creatinine (mg/dL) 0.809± 0.205 0.906± 0.172 0.890± 0.203 0.839± 0.191 1.401 0.2465Uric acid (mg/dL) 4.82± 0.91 5.12± 1.19 4.95± 1.63 4.95± 1.38 0.195 0.8997

a-MCI: amnestic single-domain mild cognitive impairment;md-MCI: amnestic multi-domain mild cognitive impairment;AD: Alzheimer’sdisease; HC: healthy controls. df: degrees of freedom. To convert from conventional unit to a SI unit, multiply by the conversion factor; 0.0259for cholesterol to mmol/L; 0.0113 for triglycerides to mmol/L; 0.0555 fro glucose to mmol/L; 88.4 for creatinine toµmol/L; 59.48 for uric acidto µmol/L.

Table 3Crude values of oxysterols and Vitamin E concentrations in a-MCI, md-MCI, AD and HC groups

HC (n = 24) a-MCI (n = 24) md-MCI (n = 29) AD (n = 37) ANOVA (df = 3.11)Biochemical parameters mean± SD mean± SD mean± SD mean± SD F p

24S-OHC (ng/mL) 57.26± 24.54 61.02± 20.73 58.76± 19.66 68.66± 19.08 1.930 0.128927-OHC (ng/mL) 190.53± 51.42 176.24± 84.62 195.22± 66.65 221.85± 76.42 2.218 0.090027-OHC/24S-OH 4.05± 2.55 3.18± 1.71 3.53± 1.56 3.40± 1.32 1.041 0.37777β-OHC (ng/mL) 6.66± 1.73 8.66± 6.68 8.85± 5.18 9.33± 8.07 0.970 0.40957-KC (ng/mL) 13.19± 5.17 16.42± 16.09 14.74± 11.61 16.79± 13.25 0.500 0.68337β-OHC plus 7-KC (ng/mL) 19.85± 6.69 25.08± 22.66 23.59± 16.66 26.12± 20.90 0.619 0.6041Vitamin Eµmol/L 33.10± 7.79 31.65± 8.67 30.27± 10.33 28.85± 11.15 1.010 0.3914

a-MCI: amnestic single-domain mild cognitive impairment;md-MCI: amnestic multi-domain mild cognitive impairment;AD: Alzheimer’sdisease; HC: healthy controls; SD: standard deviation.

groups did not differ significantly for age. As expect-ed, however, the AD participants were older than theparticipants in the other groups. Thus, to verify thepotential effect of age on our results, we performed aseries of correlation analyses for age and all biochem-ical variables considered in this study and found nosignificant correlations either in the HC group or in thewhole patient group (correctedp > 0.2 for all com-parisons). The chi-square analysis showed that the a-MCI group contained a significantly higher percentage(78%) of males than all other groups (about 30–40%).Thus, we investigated whether the gender variable hadan effect on our results. Results showed no differences(correctedp > 0.2 for all comparisons) between malesand females for Vitamin E concentrations corrected forcholesterol or for any other variable considered.

No difference was found between patients and con-trols for selected routine blood chemistry parameters,including lipids, glucose, creatinine, and uric acid val-ues (Table 2).

ANOVA results (Table 3) showed no significant dif-ferences among groups for absolute concentrations ofthe 24S-OHC, 27-OHC, 27-OHC/24S-OHC ratio, 7β-OHC, 7-KC, 7β-OHC plus 7-KC, or vitamin E. Theplasma concentration of 27-OHC was 3-4 times higher

than that of 24S-OHC in the single groups. Free-radicaloxysterols and enzymatic oxysterols were highly corre-lated (7β-OHC vs. 7-KCr = 0.924.,p < 0.0001; 24S-OHC vs. 27-OHCr = 0.382,p < 0.0001); however,free radical oxysterols were not significantly correlatedwith enzymatic oxysterols, which were not significant-ly correlated with cholesterol. By contrast, vitamin Ewas positively correlated with cholesterol in the wholepopulation (r = 0.372, p< 0.0001), whereas no corre-lation was observed between vitamin E and oxysterolconcentrations. Also, no significant correlation wasfound between oxysterol concentrations and vitaminE corrected for cholesterol (ratios) either in the wholesample or in the individual groups (i.e., AD, MCI, orHC).

A MANOVA using the values of oxysterols and Vi-tamin E concentrations corrected for cholesterol (ra-tios) as dependent variables indicated that the groupsdiffered globally for these parameters (Roy’s GreatestRoot = 0.135; F= 2.927, df= 5,108,p = 0.0162).ANOVA results, presented in Table 4, show that vita-min E/cholesterol ratios were significantly reduced inAD patients compared to HC.

No statistically significant difference in 24S-OHC/cholesterol or in the 27-OHC/cholesterol ratio was ob-

1388 L. Iuliano et al. / Vitamin E and Oxysterols in MCI and AD

Tabl

e4

Val

ues

ofox

yste

rols

and

Vita

min

Eco

rrec

ted

for

chol

este

roli

na-

MC

I,m

d-M

CI,

AD

and

HC

grou

ps

HC

a-M

CI

md-

MC

IA

DA

NO

VAA

Dvs

.A

Dvs

.A

Dvs

.a-

MC

Ivs.

a-M

CIv

s.m

d-M

CIv

s.(n

=24

)(n

=24

)(n

=29

)(n

=37

)(d

f=3.

11)

a-M

CI

md-

MC

IH

Cm

d-M

CI

HC

HC

mea

n±S

Dm

ean±

SD

mea

n±S

Dm

ean±

SD

Fp

∗p

∗p

∗p

∗p

∗p

∗p

24S

-OH

C/c

hol

(ng/

mg)

28.9

2±12

.26

32.3

7±13

.77

30.4

4±12

.65

34.1

5±11

.70

0.98

60.

402

−−

−−

−−

27-O

HC

/cho

l(ng

/mg)

99.0

1±36

.18

90.9

1±43

.49

101.

85±

46.5

411

0.55±

45.7

41.

026

0.38

4−

−−

−−

−

7β-O

HC

/cho

l(ng

/mg)

3.44±

1.24

4.33±

2.81

4.46±

2.34

4.51±

3.90

0.78

40.

503

−−

−−

−−

7-K

C/c

hol(

ng/m

g)6.

73±

2.76

8.15±

6.76

7.36±

5.01

8.06±

5.94

0.40

00.

753

−−

−−

−−

(7β

-OH

C+

7-K

C)/

chol

(ng/

mg)

10.1

8±3.

8412

.49±

9.49

11.8

2±7.

2712

.58±

9.68

0.48

80.

691

−−

−−

−−

Vita

min

E/c

hol(µ

mol

/m

mol

)6.

60±

1.91

6.16±

1.49

5.81±

1.76

5.32±

1.80

2.80

80.

043

0.0

74

0.25

50.

006

0.49

40.

376

0.10

9

a-M

CI:

amne

stic

sing

le-d

omai

nm

ildco

gniti

veim

pairm

ent;

md-

MC

I:am

nest

icm

ulti-

dom

ain

mild

cogn

itive

impa

irmen

t;AD

:Alz

heim

er’s

dise

ase;

HC

:hea

lthy

cont

rols

;ch

ol:

chol

est

erol

;df

:de

gree

sof

free

dom

;SD

:sta

ndar

dde

viat

ion.

∗F

ishe

rpo

st-h

octe

st.

Sig

nific

antp

valu

es(

p<

0.05

)ar

ein

dica

ted

inbo

ld.

L. Iuliano et al. / Vitamin E and Oxysterols in MCI and AD 1389

Fig. 1. Values of plasmaα-tocopherol concentrations corrected forcholesterol in a-MCI, md-MCI, AD, and HC groups. a-MCI: amnes-tic single-domain mild cognitive impairment; md-MCI: amnesticmulti-domain mild cognitive impairment; AD: Alzheimer’s disease;HC: healthy controls. Fisher PLSD post-hoc test;∗HC vs. AD,p =

0.0064.

served between HC and patients. ANOVA results indi-cated that 7β-OHC and 7-KC and their sum did not dif-fer among groups even after correction for cholesterol(Table 4).

The vitamin E value corrected for cholesterol de-creased progressively as a function of categorical sever-ity of dementia. However,post-hocpairwise compar-isons showed that the vitamin E/cholesterol ratio dif-fered significantly only between the AD and HC groupsand that AD patients had lower scores (Table 4, Fig. 1).A trend toward significance (p < 0.1) was observed inthe difference between the AD and a-MCI groups. Nosignificant differences among groups emerged for theother oxysterols corrected for cholesterol.

Finally, statin use did not differ among the fourgroups. Vitamin E and oxysterols (both uncorrectedand cholesterol corrected values) did not differ statis-tically between subjects who did or did not use statindrugs. Furthermore, after removing the subgroup ofpatients taking statin drugs, differences in the vitaminE/cholesterol ratio level among the 4 groups remainedstatistically significant (F= 2.700; df= 3,95; p <

0.05).

DISCUSSION

In this study, we used specific mass spectrometry-based assays to measure plasma 7β-OHC and 7-KC,two validated markers of lipid peroxidationin vivo[34],because they have been found elevated in several clin-ical conditions characterized by high systemic oxida-tive stress status and may be potential disease mark-

ers in vascular disorders [16,39,40]. 7β-OHC and 7-KC undergo mutual interconversionin vivo [36]. Un-like previous studies (including those based on iso-prostanes [15,41] which are considered accurate mark-ers of lipid peroxidation), we observed no increase in7β-OHC, 7-KC or 7β-OHC plus 7-KC levels in eitherthe crude or the cholesterol-corrected values and wefound no correlation between cholesterol and 7β-OHC,7-KC or 7β-OHC, plus 7-KC levels. This lack of cor-relation could limit the usefulness of these cholesteroloxidation markers in this specific disease.

In accordance with previous reports, we found nosignificant reduction of vitamin E in patients’ blood.This is in apparent conflict with the finding of increasedoxidative stress in AD patients, which should be as-sociated with the consumption of protective antioxi-dants. However, it is known that absolute concentra-tions ofα-tocopherol correlate with plasma lipids [42]and we actually found a positive correlation betweenα-tocopherol and cholesterol plasma levels. Conversely,α-tocopherol concentrations may be lower than controllevels if adjusted for lipid concentration, which is con-sidered the appropriate test for assessingα-tocopherolstatus in plasma [37]. In this study, after we adjustedα-tocopherol for plasma cholesterol we found a pro-gressive reduction of theα-tocopherol/cholesterol ra-tio from HC, throughout MCI, to AD patients. How-ever, we observed a statistically significant reductiononly in AD patients in comparison to HC. In fact, theAD population was the only one in which we observeda statistically significant reduction, in agreement withprevious reports [18–20]. The reduction of vitamin Ecould have been due to a nutritional deficit of this vi-tamin. However, the trend toward a decrease of theplasma vitamin E/cholesterol ratio across patient cate-gories could account for an accelerated degradation ofvitamin E. Consumption of vitamin E in the presence ofunaltered concentrations of plasma 7β-OHC and 7-KCis consistent with an increased flux of free radicals inthe brain of dementia patients with consumption of theantioxidant, which is reflected back into the plasmat-ic vitamin E pool. By contrast, the resulting increaseof cholesterol oxidation in the brain is not associatedwith the transfer of 7β-OHC and 7-KC to the bloodcompartment and might be explained as a hindranceto the blood-brain barrier. This hypothesis should beinvestigated in future studies.

Furthermore, in our study population we observedno statistically significant difference in 24S-OHC and27-OHC in either crude or cholesterol-corrected valuesand we found no correlation between cholesterol and

1390 L. Iuliano et al. / Vitamin E and Oxysterols in MCI and AD

24S-OHC or 27-OHC levels. In accordance with a pre-vious report [43], we found a high positive correlationbetween 24S-OHC and 27-OHC in the whole patientgroup, suggesting that the same mechanism contributesto the plasma level in the two oxysterols. 24S-OHCis an enzymatically oxidized product of cholesterol,which is synthesized by cytochrome CYP46A1 and isalmost exclusively located in the neuronal cells of thebrain [44]. It provides the major route for cholesterolelimination from the mammalian brain [14] by shuttlingacross the blood-brain barrier, a passage not allowed tocholesterol, to reach the systemic circulation. 27-OHC,which is synthesized by CYP27A1 (which is not brainspecific) prevalently in macrophages and is involved incholesterol elimination, is also found in the brain in anorder of magnitude lower than 24S-OHC [45]. A netblood-to-brain passage of 27-OHC has been demon-strated [46] and could serve as the link between hy-percholesterolemia and AD [47] because of the effectof oxysterols on amyloidogenesis [6]. It has also beenreported that, opposite to the flux of 24S-OHC, 27-OHC enters into the brain circulation to participate inthe formation of Aβ by antagonizing the protective ef-fect of 24S-OHC [48] and suppressing the formationof the Activity-Regulated Cytoskeleton-associated pro-tein [49].

Previous studies reported an increase [50] a de-crease [43,47,51–54], and no difference [55] in plasma24S-OHC in AD patients [56]. A bias in patient selec-tion or in the analytical methods used by different lab-oratories could account for the discrepancies in bloodoxysterol levels in these studies. Because these poten-tial biases limit the usefulness of oxysterols as molec-ular probes of disease, the standardization of analyti-cal methods is mandatory. Future studies on oxysterolmetabolism in neurodegenerative diseases should takeinto account additional steps in cholesterol metabolismin the brain. 24S-OHC is metabolized into 24-cholest-5-ene-3β, 7α, 24-triol by CYP39A1. It has been foundthat CYP39A1 expression increases progressively fromhealthy controls to AD [57]. A putative very efficientmechanism for eliminating 27-OHC from the brainis the metabolic transformation of 27-OHC into 7α-hydroxy-3-oxo-4-cholestanoic acid, which is able torapidly cross the blood-brain barrier [58].

In conclusion, we found that AD patients show re-duced levels of vitamin E corrected for cholesteroland unaltered oxysterols in their systemic circulation.Future studies on vitamin E in AD should take in-to account nutritional markers and additional oxida-tive stress markers to clarify why vitamin E levels

are altered in this disease. Further investigation ofoxysterols, including late cholesterol metabolic by-products, in neurodegenerative disease is necessary toidentify new markers for the diagnosis of AD.

ACKNOWLEDGMENTS

This work was supported by grants from the Minis-tero dell’ Universita, Ricerca Scientifica e Tecnologica(PRIN 2007L7BHK8) and from Sapienza Universityof Rome (Faculty Grant 2008) (to LI); and Ministerodella Salute (RC06-07-08-09/Aand RF 06-07) (to GS).

Authors’ disclosures available online (http://www.j-alz.com/disclosures/view.php?id=475).

REFERENCES

[1] Goedert M, Spillantini MG (2006) A century of Alzheimer’sdisease.Science314, 777-781.

[2] Bu G (2009) Apolipoprotein E and its receptors in Alzheimer’sdisease: pathways, pathogenesis and therapy.Nat Rev Neu-rosci 10, 333-344.

[3] Shobab LA, Hsiung GY, Feldman HH (2005) Cholesterol inAlzheimer’s disease.Lancet Neurol4, 841-852.

[4] Wolozin B (2001) A fluid connection: cholesterol and Abeta.Proc Natl Acad Sci U S A98, 5371-5373.

[5] Nelson TJ, Alkon DL (2005) Insulin and cholesterol path-ways in neuronal function, memory and neurodegeneration.Biochem Soc Trans33, 1033-1036.

[6] Bjorkhem I, Cedazo-Minguez A, Leoni V, Meaney S (2009)Oxysterols and neurodegenerative diseases.Mol Aspects Med30, 171-179.

[7] Marx J (2001) Alzheimer’s disease. Bad for the heart, badforthe mind?Science294, 508-509.

[8] Morris JC, Storandt M, Miller JP, McKeel DW, Price JL, Ru-bin EH, Berg L (2001) Mild cognitive impairment representsearly-stage Alzheimer disease.Arch Neurol58, 397-405.

[9] Markesbery WR (1999) The role of oxidative stress inAlzheimer disease.Arch Neurol56, 1449-1452.

[10] Nelson TJ, Alkon DL (2005) Oxidation of cholesterol by amy-loid precursor protein and beta-amyloid peptide.J Biol Chem280, 7377-7387.

[11] Mitomo H, Chen WH, Regen SL (2009) Oxysterol-inducedrearrangement of the liquid-ordered phase: a possible linktoAlzheimer’s disease?J Am Chem Soc131, 12354-12357.

[12] Kim DH, Frangos JA (2008) Effects of amyloid beta-peptideson the lysis tension of lipid bilayer vesicles containing oxys-terols.Biophys J95, 620-628.

[13] Zhang Q, Powers ET, Nieva J, Huff ME, Dendle MA, BieschkeJ, Glabe CG, Eschenmoser A, Wentworth P, Jr., Lerner RA,Kelly JW (2004) Metabolite-initiated protein misfolding maytrigger Alzheimer’s disease.Proc Natl Acad Sci U S A101,4752-4757.

[14] Lutjohann D, Breuer O, Ahlborg G, Nennesmo I, Siden A,Diczfalusy U, Bjorkhem I (1996) Cholesterol homeostasis inhuman brain: evidence for an age-dependent flux of 24S-hydroxycholesterol from the brain into the circulation.ProcNatl Acad Sci U S A93, 9799-9804.

L. Iuliano et al. / Vitamin E and Oxysterols in MCI and AD 1391

[15] Pratico D (2008) Oxidative stress hypothesis in Alzheimer’sdisease: a reappraisal.Trends Pharmacol Sci29, 609-615.

[16] Micheletta F, Iuliano L (2006) Free radical attack on choles-terol: oxysterols as markers of oxidative stress and as bioac-tive molecules.Immun Endoc Metab Agents in Med Chem6,305-316.

[17] de la Torre JC (2002) Alzheimer disease as a vascular disorder:nosological evidence.Stroke33, 1152-1162.

[18] Guidi I, Galimberti D, Lonati S, Novembrino C, BamontiF, Tiriticco M, Fenoglio C, Venturelli E, Baron P, BresolinN, Scarpini E (2006) Oxidative imbalance in patients withmild cognitive impairment and Alzheimer’s disease.NeurobiolAging27, 262-269.

[19] Ricciarelli R, Argellati F, Pronzato MA, Domenicotti C(2007)Vitamin E and neurodegenerative diseases.Mol Aspects Med28, 591-606.

[20] Grundman M (2000) Vitamin E and Alzheimer disease: thebasis for additional clinical trials.Am J Clin Nutr71, 630S-636S.

[21] Wengreen HJ, Munger RG, Corcoran CD, Zandi P, HaydenKM, Fotuhi M, Skoog I, Norton MC, Tschanz J, Breitner JC,Welsh-Bohmer KA (2007) Antioxidant intake and cognitivefunction of elderly men and women: the Cache County Study.J Nutr Health Aging11, 230-237.

[22] Isaac MG, Quinn R, Tabet N (2008) Vitamin E for Alzheimer’sdisease and mild cognitive impairment.Cochrane DatabaseSyst Rev, CD002854.

[23] Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG,Kokmen E (1999) Mild cognitive impairment: clinical char-acterization and outcome.Arch Neurol56, 303-308.

[24] Morris JC (1993) The Clinical Dementia Rating (CDR): cur-rent version and scoring rules.Neurology43, 2412-2414.

[25] Winblad B, Palmer K, Kivipelto M, Jelic V, Fratiglioni L,Wahlund LO, Nordberg A, Backman L, Albert M, AlmkvistO, Arai H, Basun H, Blennow K, de Leon M, DeCarli C, Erk-injuntti T, Giacobini E, Graff C, Hardy J, Jack C, Jorm A,Ritchie K, van Duijn C, Visser P, Petersen RC (2004) Mildcognitive impairment–beyond controversies, towards a con-sensus: report of the International Working Group on MildCognitive Impairment.J Intern Med256, 240-246.

[26] McKhann G, Drachman D, Folstein M, Katzman R, Price D,Stadlan EM (1984) Clinical diagnosis of Alzheimer’s disease:report of the NINCDS-ADRDA Work Group under the aus-pices of Department of Health and Human Services Task Forceon Alzheimer’s Disease.Neurology34, 939-944.

[27] Arca M, Natoli S, Micheletta F, Riggi S, Di AngelantonioE, Montali A, Antonini TM, Antonini R, Diczfalusy U, Iu-liano L (2007) Increased plasma levels of oxysterols,in vivomarkers of oxidative stress, in patients with familial combinedhyperlipidemia: reduction during atorvastatin and fenofibratetherapy.Free Radic Biol Med42, 698-705.

[28] Folstein MF, Folstein SE, McHugh PR (1975) “Mini-mentalstate”. A practical method for grading the cognitive state ofpatients for the clinician.J Psychiatr Res12, 189-198.

[29] Carlesimo GA, Caltagirone C, Gainotti G (1996) The MentalDeterioration Battery: normative data, diagnostic reliabilityand qualitative analyses of cognitive impairment. The Groupfor the Standardization of the Mental Deterioration Battery.Eur Neurol36, 378-384.

[30] Rey A (1941) L’examen psychologique dan les casd’encephalopathie traumatique (les problemes).Arch Psy-chologique28, 286-340.

[31] Osterrieth P (1944) le test de copie d’une figure complexe.Arch Psychologique30, 206-356.

[32] Scoop J (1935) Studies of interference in serial verbalreac-tions.J Experim Psycol18, 643-662.

[33] Dzeletovic S, Breuer O, Lund E, Diczfalusy U (1995) De-termination of cholesterol oxidation products in human plas-ma by isotope dilution-mass spectrometry.Anal Biochem225,73-80.

[34] Iuliano L, Micheletta F, Natoli S, Ginanni Corradini S,IappelliM, Elisei W, Giovannelli L, Violi F, Diczfalusy U (2003)Measurement of oxysterols and alpha-tocopherol in plasmaand tissue samples as indices of oxidant stress status.AnalBiochem312, 217-223.

[35] Liebler DC, Burr JA, Philips L, Ham AJ (1996) Gaschromatography-mass spectrometry analysis of vitamin E andits oxidation products.Anal Biochem236, 27-34.

[36] Larsson H, Bottiger Y, Iuliano L, Diczfalusy U (2007)In vivo interconversion of 7beta-hydroxycholesterol and 7-ketocholesterol, potential surrogate markers for oxidativestress.Free Radic Biol Med43, 695-701.

[37] Traber MG, Jialal I (2000) Measurement of lipid-soluble vita-mins – further adjustment needed?Lancet355, 2013-2014.

[38] Iuliano L, Monticolo R, Straface G, Zullo S, Galli F, Boaz M,Quattrucci S (2009) Association of cholesterol oxidation andabnormalities in fatty acid metabolism in cystic fibrosis.Am JClin Nutr 90, 477-484.

[39] Zieden B, Kaminskas A, Kristenson M, Kucinskiene Z, VessbyB, Olsson AG, Diczfalusy U (1999) Increased plasma 7 beta-hydroxycholesterol concentrations in a population with a highrisk for cardiovascular disease.Arterioscler Thromb Vasc Biol19, 967-971.

[40] Alkazemi D, Egeland G, Vaya J, Meltzer S, Kubow S (2008)Oxysterol as a marker of atherogenic dyslipidemia in adoles-cence.J Clin Endocrinol Metab93, 4282-4289.

[41] de Leon MJ, DeSanti S, Zinkowski R, Mehta PD, Pratico D,Segal S, Rusinek H, Li J, Tsui W, Saint Louis LA, Clark CM,Tarshish C, Li Y, Lair L, Javier E, Rich K, Lesbre P, MosconiL, Reisberg B, Sadowski M, DeBernadis JF, Kerkman DJ,Hampel H, Wahlund LO, Davies P (2006) Longitudinal CSFand MRI biomarkers improve the diagnosis of mild cognitiveimpairment.Neurobiol Aging27, 394-401.

[42] Thurnham DI, Davies JA, Crump BJ, Situnayake RD, Davis M(1986) The use of different lipids to express serum tocopherol:lipid ratios for the measurement of vitamin E status.Ann ClinBiochem23(Pt 5), 514-520.

[43] Kolsch H, Heun R, Kerksiek A, Bergmann KV, Maier W,Lutjohann D (2004) Altered levels of plasma 24S- and 27-hydroxycholesterol in demented patients.Neurosci Lett368,303-308.

[44] Lund EG, Guileyardo JM, Russell DW (1999) cDNA cloningof cholesterol 24-hydroxylase, a mediator of cholesterol home-ostasis in the brain.Proc Natl Acad Sci U S A96, 7238-7243.

[45] Leoni V, Masterman T, Patel P, Meaney S, Diczfalusy U,Bjorkhem I (2003) Side chain oxidized oxysterols in cere-brospinal fluid and the integrity of blood-brain and blood-cerebrospinal fluid barriers.J Lipid Res44, 793-799.

[46] Heverin M, Meaney S, Lutjohann D, Diczfalusy U, WahrenJ, Bjorkhem I (2005) Crossing the barrier: net flux of 27-hydroxycholesterol into the human brain.J Lipid Res46, 1047-1052.

[47] Bjorkhem I, Heverin M, Leoni V, Meaney S, Diczfalusy U(2006) Oxysterols and Alzheimer’s disease.Acta Neurol ScandSuppl185, 43-49.

[48] Famer D, Meaney S, Mousavi M, Nordberg A, Bjorkhem I,Crisby M (2007) Regulation of alpha- and beta-secretase ac-tivity by oxysterols: cerebrosterol stimulates processing of

1392 L. Iuliano et al. / Vitamin E and Oxysterols in MCI and AD

APP via the alpha-secretase pathway.Biochem Biophys ResCommun359, 46-50.

[49] Mateos L, Akterin S, Gil-Bea FJ, Spulber S, Rahman A,Bjorkhem I, Schultzberg M, Flores-Morales A, Cedazo-Minguez A (2009) Activity-regulated cytoskeleton-associatedprotein in rodent brain is down-regulated by high fat dietinvivo and by 27-hydroxycholesterol in vitro.Brain Pathol19,69-80.

[50] Lutjohann D, Papassotiropoulos A, Bjorkhem I, Locatelli S,Bagli M, Oehring RD, Schlegel U, Jessen F, Rao ML, vonBergmann K, Heun R (2000) Plasma 24S-hydroxycholesterol(cerebrosterol) is increased in Alzheimer and vascular dement-ed patients.J Lipid Res41, 195-198.

[51] Heverin M, Bogdanovic N, Lutjohann D, Bayer T, PikulevaI, Bretillon L, Diczfalusy U, Winblad B, Bjorkhem I (2004)Changes in the levels of cerebral and extracerebral sterolsinthe brain of patients with Alzheimer’s disease.J Lipid Res45,186-193.

[52] Leoni V, Masterman T, Mousavi FS, Wretlind B, Wahlund LO,Diczfalusy U, Hillert J, Bjorkhem I (2004) Diagnostic use ofcerebral and extracerebral oxysterols.Clin Chem Lab Med42,186-191.

[53] Bretillon L, Siden A, Wahlund LO, Lutjohann D, MinthonL, Crisby M, Hillert J, Groth CG, Diczfalusy U, BjorkhemI (2000) Plasma levels of 24S-hydroxycholesterol in patientswith neurological diseases.Neurosci Lett293, 87-90.

[54] Solomon A, Leoni V, Kivipelto M, Besga A, Oksengard AR,

Julin P, Svensson L, Wahlund LO, Andreasen N, WinbladB, Soininen H, Bjorkhem I (2009) Plasma levels of 24S-hydroxycholesterol reflect brain volumes in patients with-out objective cognitive impairment but not in those withAlzheimer’s disease.Neurosci Lett462, 89-93.

[55] Papassotiropoulos A, Lutjohann D, Bagli M, Locatelli S,Jessen F, Buschfort R, Ptok U, Bjorkhem I, von Bergmann K,Heun R (2002) 24S-hydroxycholesterol in cerebrospinal fluidis elevated in early stages of dementia.J Psychiatr Res36,27-32.

[56] Papassotiropoulos A, Lutjohann D, Bagli M, Locatelli S,Jessen F, Rao ML, Maier W, Bjorkhem I, von Bergmann K,Heun R (2000) Plasma 24S-hydroxycholesterol: a peripheralindicator of neuronal degeneration and potential state markerfor Alzheimer’s disease.Neuroreport11, 1959-1962.

[57] Blalock EM, Geddes JW, Chen KC, Porter NM, MarkesberyWR, Landfield PW (2004) Incipient Alzheimer’s disease: mi-croarray correlation analyses reveal major transcriptional andtumor suppressor responses.Proc Natl Acad Sci U S A101,2173-2178.

[58] Meaney S, Heverin M, Panzenboeck U, Ekstrom L, Axels-son M, Andersson U, Diczfalusy U, Pikuleva I, Wahren J,Sattler W, Bjorkhem I (2007) Novel route for elimination ofbrain oxysterols across the blood-brain barrier: conversion in-to 7alpha-hydroxy-3-oxo-4-cholestenoic acid.J Lipid Res48,944-951.