Biological Psychology 94 (2013) 527–533

Contents lists available at ScienceDirect

Biological Psychology
journal homepage: www.elsevier.com/locate/biopsycho

Oxidative stress is involved in fatigue induced by overnight deskwork

as assessed by increase in plasma tocopherylhydroqinone and
hydroxycholesterol
Mototada Shichiri a,∗,1 , Nobuyoshi Harada a,b,1 , Noriko Ishida a , Lilian Kaede Komaba a ,
Sunao Iwaki a , Yoshihisa Hagihara a , Etsuo Niki a , Yasukazu Yoshida a
a
Health Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), Osaka, Japan
b
Flicker Health Management Corp., Osaka, Japan

a r t i c l e i n f o a b s t r a c t

Article history: In this study, we examined the relationship between fatigue and plasma concentrations of antioxidants
Received 1 February 2013 and lipid peroxidation products. Fourteen healthy volunteers performed overnight desk work for 18 h
Received in revised form 24 June 2013 then took a nap for 4 h. Participants answered questionnaires of subjective symptoms of fatigue (QSSF)
Accepted 2 October 2013
and completed a self-assessment of fatigue using a visual analog scale (VAS). At each test time, they
Available online 10 October 2013
underwent a critical flicker frequency (CFF) test and blood samples were collected. Plasma levels of ␣-
tocopherol (␣T) decreased and ␣-tocopherylquinone (␣TQ), the oxidation product of ␣T, increased. The
Keywords:
ratio of 7␤-hydroxycholesterol (7␤-OHCh), the oxidation product of cholesterol, against total cholesterol
Lipid peroxidation
Fatigue
increased until the end of experiment. ␣TQ levels correlated with VAS and QSSF scores. The ratio of
Overnight deskwork 7␤-OHCh to total cholesterol and the value of CFF showed a significant correlation. From these results,
Critical flicker frequency plasma levels of ␣TQ and 7␤-OHCh are useful and objective indicators of fatigue induced by overnight
␣-Tocopherylquinone deskwork.
© 2013 Elsevier B.V. All rights reserved.

1. Introduction in the search for objective biological indices of fatigue. Another

It is considered that fatigue is one of the most important feel-
ings in daily life because it is closely related to the quality of life.
However, fatigue is usually assessed only subjectively and no gold
standard for its objective measurement has been established.
In previous reports, subjective evaluation of fatigue was usu-
ally assessed using visual analog scale (VAS) score (Kos, Nagels,
D’Hooghe, Duportail, & Kerckhofs, 2006) and self report ques-
tionnaires (Yoshihara, Yamanaka, & Kawakami, 2009). Subjective
evaluations of fatigue have many advantages as their simple oper-
ation and low costs but they have disadvantages such as difficulty
in standardization.
Heart function (Hartley, Arnold, Smythe, & Hansen, 1994) and
brain function (Kaseda, Jiang, Kurokawa, Mimori, & Nakamura,
1998; Suda et al., 2009) have been studied as correlates of fatigue
Abbreviations: CFF, critical flicker frequency; QSSF, questionnaires of sub-
jective symptoms of fatigue; VAS, visual analog scale; ␣T␣, -tocopherol; ␣TQ,
␣-tocopherylquinone; UQ10 , ubiquinone-10; UQ10 H2 , ubiquinol-10; CoQ10 , coen-
zyme Q10 ; 7␤-OHCh, 7␤-hydroxycholesterol; HODE, hydroxyoctadecadienoic acid.
∗ Corresponding author. Tel.: +81 72 751 8734; fax: +81 72 751 9517.
E-mail address: mototada-shichiri@aist.go.jp (M. Shichiri).
1
These authors are contributed equally to this work.
objective index has been proposed; the critical flicker frequency
(CFF) which is the level of individual sensitivity at the beginning
of light flickering, caused by changes in the frequency of light
flashes (Simonson & Brozek, 1952). The descending threshold is
measured as the highest frequency of flashing light when the flicker
is perceived. This is an accepted indicator of fatigue caused by
the workload in many different types of occupations (Luczak &
Sobolewski, 2005). Some reports used CFF as objective indicator
of mental fatigue (Marziale & Rozestraten, 1995).
Nozaki et al. reported fatigue-related biochemical alterations
after fatigue-inducing mental and physical sessions (Nozaki et al.,
2009). They evaluated 40 biomarkers. But, none of these proved a
key indicator for the evaluation of fatigue.
On the other hand, it is reported that strong exercise and var-
ious stresses that cause fatigue increase oxidative stress (Maes,
Kubera, Obuchowiczwa, Goehler, & Brzeszcz, 2011; Siktar et al.,
2011). We have proposed that the levels of hydroxyoctadecadienoic
acid (HODE) and 7␤-hydroxycholesterol (7␤-OHCh) are potential
biomarkers of the oxidative stress status (Kitano, Yoshida, Kawano,
Hibi, & Niki, 2007; Shichiri et al., 2011; Yoshida, Yoshikawa, Kinumi,
Imai, & Niki, 2009; Yoshida et al., 2010). HODE and 7␤-OHCh are
generated from linoleic acid and cholesterol, respectively.
The antioxidant ␣-tocopherol (␣T) and coenzyme Q10 (CoQ10 )
are able to prevent lipid peroxidation and we have previously

0301-0511/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.biopsycho.2013.10.002
528 M. Shichiri et al. / Biological Psychology 94 (2013) 527–533
Fig. 1. Investigation protocol.
reported that negative correlations were observed between the
level of total HODE (tHODE) and the concentration of ␣T and CoQ9
in mouse liver (Yoshida, Hayakawa, Habuchi, & Niki, 2006). Miwa
et al. reported that plasma ␣T level of chronic fatigue syndrome
patients was lower than control subjects (Miwa & Fujita, 2010).
Maes et al. reported that plasma CoQ10 was significantly lower
in chronic fatigue syndrome patients (Maes et al., 2009a) and in
depressed patients (Maes et al., 2009b). There is a possibility that
reduction of plasma levels of ␣T and CoQ10 increases oxidative
stress.
In the present study, we examined the relationship between
fatigue and oxidative stress. Healthy volunteers performed
overnight deskwork and the change of fatigue level was eval-
uated by VAS score, questionnaire of subjective symptoms of
fatigue (QSSF) and CFF. We measured the plasma levels of
HODE and 7␤-OHCh. We also measured plasma concentrations
of the reduced form and oxidized forms of antioxidant, includ-
ing ␣T; ␣-tocopherylquinone (␣TQ), the oxidation product of ␣T;
ubiquinone-10 (UQ10 ) the oxidized form of CoQ10 ; and ubiquinol-
10 (UQ10 H2 ), the reduced form of UQ10 . The correlation between
the above indicators of fatigue and plasma HODE, 7␤-OHCh, ␣T and
CoQ levels were analyzed.
2. Methods
2.1. Participants
A total of 14 males, aged 32.9 ± 10.8 years old were invited to participate in
the present study. Written informed consent was obtained from all subjects after
the procedures had been fully explained. All subjects were non-smokers, in good
physical health taking no medication, and had no history of psychiatric or somatic
diseases. The present protocol was approved by the committee for the ethics on
the experiments with human derivative samples and the committee for ergonomic
experiments of AIST, Japan.
2.2. Study design
In the present study, participants did deskwork by staying up all night. The
experiment started at 14:30, and deskwork was done until 8:30 of the following
day. From 8:30 to 12:30, they took naps up to 4-h in individual rooms. Most of their
time was spent in desk working and they engaged in no hard physical activities.
Supper was taken at 19:30 and breakfast was taken at 8:30.
Fig. 1 shows the schedule of this experiment. At 14:30, 20:30, 2:30, 8:30 and
12:30, participants were requested to answer a questionnaire concerning their sub-
jective feelings regarding fatigue and instructed to note their self-assessment of
fatigue using a visual analog scale. At the same time, they underwent a critical flicker
frequency (CFF) test, and blood samples were collected.
2.3. Measurements
To evaluate the subjective symptoms of fatigue, a Questionnaire of Subjective
Symptoms of Fatigue (QSSF) “Jikaku-sho shirabe” was used, based on the Industrial
Fatigue Research Committee of the Japan Occupational Health in 2002 (Takeyama
et al., 2009). This questionnaire consists of 25 subjective fatigue symptom items that
are categorized into 5 factors: feeling of drowsiness (Factor I), feeling of instability
(Factor II), feeling of uneasiness (Factor III), feeling of local pain or dullness (Factor
IV), and feeling of eyestrain (Factor V). For each item, respondents are requested
to estimate the intensity of the feelings as “Disagree completely,” “Agree scarcely,”
“Agree slightly,” “Agree considerably,” and “Agree strongly.” These five intensities
were assigned scores of 1–5 points, respectively.
Self-assessment of fatigue was assessed using a visual analog scale (VAS). VAS
scores was indicated from 0 to 10, with 0 = no fatigue and 10 = strong fatigue.
The critical flicker frequency (CFF) was successively measured five times in
descending order using a Roken Digital Flicker (Model RDF-1: Shibata Co., Ltd.,
Tokyo) and the mean value was obtained.
2.4. Materials
13-hydroxy-9Z,11E-octadecadienoic acid (13-(Z,E)-HODE),
9-hydroxy-10E,12Z-octadecadienoic acid (9-(E,Z)-HODE), and 13S-hydroxy-
10E,12Z-octadecadienoic-9,10,12,13-d4 acid (13-HODE-d4) were obtained from
Cayman Chemical (Ann Arbor, MI, USA). 9-Hydroxy-10E,12E-octadecadienoic
acid (9-(E,E)-HODE), 13-hydroxy-9E,11E-octadecadienoic acid (13-(E,E)-HODE),
10-hydroxy-8E,12Z-octadecadienoic acid (10-(Z,E) -HODE), and 12-hydroxy-
9Z,13E-octadecadienoic acid (12-(Z,E)-HODE) were obtained from Larodan Fine
Chemicals (Malmö, Sweden). 7␤-OHCh were obtained from Steroids (New Port, RI,
USA), and its isotope 7␤-hydroxycholesterol-25,26,26,26,27,27,27-d7 (7␤-OHCh-
d7) were obtained from Medical Isotopes (Pelham, NH, USA). The standard of
␣-tocopherol was kindly supplied by Eisai Co., Ltd. (Tokyo, Japan). The standard
of ␣-tocopherylquinone was prepared by oxidation of ␣-tocopherol with copper
sulfate. Other materials were of the highest grade available commercially.
2.5. Plasma sample processing
Plasma was obtained by centrifugation at 830 × g for 5 min at 4 ◦ C and imme-
diately subjected to analysis. The plasma samples were processed for analysis by
a reduction and saponification using the previously reported method (Yoshida,
Kodai, Takemura, Minamiyama, & Niki, 2008). The plasma (200 ␮l) was mixed with
300 ␮l of saline. Subsequently, 500 ␮l of methanol containing internal standards, 13-
HODE-d4 (50 ng), 7␤-OHCh-d7 (19 ng), 16-hydroxyhexadecanoic acid (70 ng) and
100 ␮M butylated hydroxytoluene was added to the samples. This was followed
by the reduction of hydroperoxides and ketones using 1 mM triphenylphosphine
(Sigma–Aldrich) at room temperature for 30 min. The reduced sample was then
mixed with 1 M KOH in methanol (500 ␮l) under nitrogen and incubated on a shaker
for 30 min in the dark at 40 ◦ C. The mixture was cooled on ice and acidified with 10%
(v/v) acetic acid in water (2 ml), then extracted with chloroform/ethyl acetate (4:1,
v/v, 5 ml). The sample was mixed using a vortex mixer for 1 min and centrifuged at
1500 × g for 5 min at 4 ◦ C. The chloroform/ethyl acetate layer was concentrated to
approximately 1 ml after the removal of the water layer and was divided equally into
two portions, a sample for liquid chromatography–mass spectrometry (LC–MS/MS)
and a sample for gas chromatography mass spectrometry (GC–MS).
2.6. Analysis of lipid peroxidation products by LC–MS/MS and GC–MS
The levels of t7-OHCh, total cholesterol (tCh), and total linoleic acid were mea-
sured by GC–MS and the levels of HODE were measured by LC–MS/MS using a
previously reported method (Yoshida, Kodai, Takemura, Minamiyama, & Niki, 2008).
The divided chloroform/ethyl acetate solution was evaporated to dryness under
nitrogen.
A silylating agent, N,O bis(trimethylsilyl)-trifluoroacetamide (30 ␮l), was added
to the dried residue. The solution was mixed vigorously by vortexing for 0.5 min and
incubated for 60 min at 60 ◦ C to obtain trimethylsilyl esters and ethers. An aliquot
of this sample was injected into a gas chromatograph (GC 6890N, Agilent Technolo-
gies, Palo Alto, CA, USA) that was equipped with a quadrupole mass spectrometer
(5973 Network, Agilent Technologies). A fused-silica capillary column (HP-5MS, 5%
phenyl methyl siloxane, 30 mm × 0.25 mm, Agilent Technologies) was used. Helium
was used as the carrier gas at a flow rate of 1.2 ml/min. Temperature programming
was carried out from 60 ◦ C to 280 ◦ C at 10 ◦ C/min. The injector temperature was set
at 250 ◦ C, and the temperatures of the transfer line to the mass detector and ion
source were 250 ◦ C and 230 ◦ C, respectively. Electron energy was set at 70 eV. 7␤-
OHCh, Ch, and linoleic acid were identified on the basis of their retention times and
mass patterns; ions having m/z = 456 for 7␤-OHCh, m/z = 458 for Ch, and m/z = 337
for linoleic acid were selected for the quantification. 7␤-OHCh and Ch were iden-
tified quantitatively using 7␤-OHCh-d7 as internal standard and linoleic acid was
quantified using 16-hydroxyhexadecanoic acid as an internal standard.
The other portion of the chloroform/ethyl acetate solution was also evaporated
to dryness under nitrogen. The sample was reconstituted with methanol/water
(70:30, v/v, 200 ␮l), and a portion of the sample (10 ␮l) was subjected to the LC-
MS/MS analysis. LC was carried out on an octadecylsilane (ODS) column (Hypersil
Gold, 3.0 ␮m, 100 × 2.1 mm, Thermo Fisher Scientific, Fremont, MA, USA) in a column
oven (CTO-20A, Shimadzu, Kyoto, Japan) set at 40 ◦ C. The LC apparatus consisted
of an autosampler (SIL-20AC, Shimadzu, Kyoto, Japan), pump and mixer (LC-20AB,
Shimadzu, Kyoto, Japan). The eluent condition was a gradient comprising solvent A
(2 mM ammonium acetate in water) and solvent B (methanol/acetonitrile, 5:95) at a
flow rate of 0.2 ml/min. The initial composition of the gradient was 80% A and 20% B.
It was holded for 2 min, and the composition was changed to 50% A and 50% B after
45 min. MS was carried out using a Thermo Finnigan TSQ Quantum Discovery Max,
a triple quadrupole mass spectrometer (Thermo Fisher Scientific, MA, USA) fitted
with an electrospray ionization (ESI) source. ESI was carried out at a needle voltage
of 4.2 kV. Nitrogen was used as the sheath gas (17 psi) and auxiliary gas (12 units).
The capillary was heated to 280 ◦ C, and the mass spectrometers were optimized to
obtain the maximum sensitivity. A specific precursor-to-product ion transition was
 M. Shichiri et al. / Biological Psychology 94 (2013) 527–533
(a) (b)
VAS
6 80
p<0.001
5
60 p<0.05
4 39
p<0.001
3 40
2 36
20
1
0 0
14:30 20:30 2:30 8:30 12:30 14:30 20:30
Fig. 2. Changes of visual analog scale scores, the score of questionnaire of subjective symptoms of fatigue and critical flicker frequency throughout the work period.
(a) Visual analog scale (VAS) scores.
(b) Questionnaire of subjective symptoms of fatigue (QSSF) scores.
(c) The critical flicker frequency (CFF) was measured five times in descending order and the mean value was obtained.
All data are expressed as mean ±S.E. (bars). Statistical analysis was carried out using repeated measure ANOVA.
carried out using selected reaction monitoring after collision-induced dissociation in
the negative mode. Argon was used as the collision gas, and the collision cell pressure
was 1.5 mTorr. The precursor, product ions, and collision energy were determined
after the optimization of MS/MS as follows: m/z = 295.0 and 194.6–195.6 at 21 eV
for both 13-(Z,E)-HODE and 13-(E,E)-HODE, m/z = 295.0 and 170.5–171.5 at 24 eV
for both 9-(E,Z)-HODE and 9-(E,E)-HODE, m/z = 295.0 and 182.6–183.6 at 22 eV for
both 10-(Z,E)-HODE and 12-(Z,E)-HODE, and m/z = 299.0 and 197.6–198.6 at 26 eV
for 13-HODE-d4.
2.7. Antioxidant analysis with high performance liquid chromatography
Chloroform/methanol (2/1, v/v, 150 ␮l) containing 100 ␮M butylated hydroxy-
toluene was added to the plasma sample (50 ␮l). Lipids and vitamin E were extracted
by centrifugation (15,000 rpm, at 4 ◦ C for 5 min) after mixing vigorously with a
vortex mixer. ␣-Tocopherol and ␣-tocopherylquinone were measured using high
performance liquid chromatography with an amperometric electrochemical detec-
tor (Nanospace SI-2, Shiseido, Tokyo, Japan) set at 700 mV, with an ODS column
(Wakosil-II 5C18 RS, 5 ␮m, 250 × 4.6 mm, Wako, Japan) followed by a reducing
column (RC-10, 15 × 4 mm, Shiseido) with methanol containing 50 mM sodium per-
chlorate as eluent at 0.7 ml/min.
Plasma UQ10 and UQ10 H2 were also measured by using HPLC with the amper-
ometric electrochemical detector (Nanospace SI-1), set at 700 mV. The samples
were passed through a reverse phase column (LC-8, 5 ␮m, 250 mm × 4.6 mm,
Sigma–Aldrich Japan, Tokyo, Japan), followed by a reducing column (RC-10,
30 mm × 4 mm, Shiseido) and methanol/tert-butyl alcohol (85/15, v/v) containing
50 mM sodium perchlorate as an eluent at a rate of 1 ml/min. Plasma was diluted
with methanol and hexane (1/5/10, v/v/v) and mixed vigorously by a vortex mixer
for 1 min followed by centrifugation (20,400 × g at 4 ◦ C for 10 min) and immediately
an aliquot of the upper layer was injected into the HPLC column.
2.8. Statistics
Statistical analyses were performed using repeated measure ANOVA by using
SPSS version 21.0, and a p-value of less than 0.05 was considered significant. Corre-
lations were also analyzed using the Pearson’s test and p-values less than 0.05 were
considered significant. Data are presented as mean ± S.E.
3. Results
3.1. Changes in the visual analog scale scores, the scores of
questionnaire of subjective symptoms of fatigue and critical flicker
frequency
The VAS score (Fig. 2a) and the score of QSSF (Fig. 2b) increased
time-dependently after the start of the examination. Both of these
scores increased significantly at 2:30 and 8:30 compared with the
initial scores. At 12:30, after taking a nap, the scores decreased
to levels close to their initial scores. Contrary to these changes,
the frequency of CFF test decreased time-dependently and at 8:30
decreased significantly compared with starting data. At 12:30, the
frequencies recovered to almost the same levels as the initial values.
There was a strong positive correlation between the VAS score
and the score of QSSF (r = 0.726, p < 0.001). Significant negative
529
(c)
QSSF CFF
42
p<0.01 p<0.05
(Hz)
33
2:30 8:30 12:30 14:30 20:30 2:30 8:30 12:30
correlations between CFF and either the VAS score (r = −0.387,
p < 0.001) or the score of QSSF (r = −0.208, p < 0.001) were observed.
3.2. Changes of ˛-tocopherol, ˛-tocopherylquinone and CoQ10
The plasma level of ␣-tocopherol (␣T) decreased to 95.3% of the
initial level (Fig. 3a and d). The level of ␣-tocopherylquinone (␣TQ),
an oxidation product of ␣T, increased to 121.2% of its initial level
by 2:30 but it did not change after that (Fig. 3b and e). The ratio of
␣TQ to ␣T (␣TQ/␣T) showed a similar pattern to ␣TQ (Fig. 3c and
f).
On the other hand, the plasma concentrations of ubiquinol-10
(UQ10 H2 ), ubiquinone-10 (UQ10 ), reduced and oxidized forms of
coenzyme Q10 , and their sum (total CoQ10 ) all decreased slightly
until 12:30. The final concentrations of UQ10 H2 and total CoQ10
decreased to 97.9% and 95.4% of each of their initial concentra-
tions (Fig. 3g). The ratio of UQ10 to total CoQ10 (UQ10 /total CoQ10 )
decreased until 8:30 and it increased to 116.0% of initial ratio at
12:30 (Fig. 3h).
We analyzed the correlation between biomarkers on antioxi-
dants and indicators of fatigue (Table 1). The level of ␣T negatively
correlated with the score of VAS and QSSF. %␣T, relative ratio to ini-
tial levels of each ␣T, correlated with VAS score. ␣TQ/␣T positively
correlated with VAS score and %␣TQ/␣T, relative ratio to initial lev-
els of each ␣TQ/␣T, positively correlated with the score of QSSF.
Not only subjective indicators but also %CFF, relative ratio of CFF
against initial frequency, negatively correlated with ␣TQ (Fig. 3i).
UQ10 H2 negatively correlated with VAS score (Table 1). How-
ever, UQ10 and UQ10 /total CoQ10 negatively correlated with both
subjective indicators.
3.3. Changes of lipid peroxidation product, 7ˇ-OHCh and HODE
To reduce the influence of meals, we measured the ratio of 7␤-
OHCh to total Cholesterol (7␤-OHCh/tCh) and the ratio of HODE to
linoleic acid (HODE/LA). 7␤-OHCh/tCh did not change until 2:30,
but it then increased until the end of the experimental period
(Fig. 4a and b).
The changes in plasma levels of HODE/LA (Fig. 4c) and mean
relative ratio to the initial levels of each HODE/LA (%HODE/LA) are
shown in Fig. 4d. The levels of HODE/LA did not change from the
start of the experiment to the end. %tHODE/LA increased by 26%
until 2:30, decreased once at 8:30, and increased again at 12:30.
The same trends of changes were observed in %ZE-HODE/LA. In %EE-
HODE/LA level increased until 2:30 and fell back by 8:30.
We analyzed the correlation between lipid peroxidation prod-
ucts and antioxidants (Table 2). A significant correlation between
530 M. Shichiri et al. / Biological Psychology 94 (2013) 527–533

(a) (b) (c)

αT αTQ αTQ/αT
18 0.14 0.01

17 0.12

(mol/mol)
0.008
16

μM)
μM)
0.10

(μ
(μ
15
0.006
0.08
14

13 0.06 0.004
14:30 20:30 2:30 8:30 12:30 14:30 20:30 2:30 8:30 12:30 14:30 20:30 2:30 8:30 12:30

(d) (e) (f)

%α
αT %α
αTQ %α
αTQ/αT
110% 160% 200%
180%
140%
160%
100%
120% 140%
120%
100% 100%
90%
80%
80%
60%
80% 60% 40%
14:30 20:30 2:30 8:30 12:30 14:30 20:30 2:30 8:30 12:30 14:30 20:30 2:30 8:30 12:30

(g) (h) UQ10 / total CoQ10 (i)

UQ10H2 & UQ10
140 16% 0.4
120 n=53
r=-0.313
100 14% 0.3
p=0.011

μM)
(nM)

(%)

80

αTQ (μ
total CoQ10 12% 0.2
60
UQ10H2
40 10%
UQ10 0.1
20
0 8% 0
14:30 20:30 2:30 8:30 12:30 14:30 20:30 2:30 8:30 12:30 70% 80% 90% 100% 110%
%CFF

Fig. 3. Changes of plasma levels of ␣-tocopherol, ␣-tocopherylquinone, ubiquinol-10 and ubiquinone-10.

The changes of plasma levels of ␣-tocopherol (␣T) (a), ␣-tocopherylquinone (␣TQ) (b) and the ratio of ␣TQ against ␣T (c) were analyzed. Fig. 3(d)–(f) show the changes of
relative ratio against initial levels of each ␣T, ␣TQ and ␣TQ/␣T. The changes of plasma levels of ubiquinol-10 (UQ10 H2 ), ubiquinone-10 (UQ10 ) and sum of them (total CoQ10 )
were shown in (g). The ratio of UQ10 against total CoQ10 was shown in (h). All data are expressed as mean ±S.E. (bars).
Correlation between the value of %CFF and ␣TQ (i). Correlation was analyzed using the Pearson’s test, a P-value of less than 0.05 was considered significant.

7␤-OHCh/tCh and the reduction of %␣T was observed. All types of
HODE/LA were significantly correlated with ␣TQ/␣T and %␣TQ/␣T.
As well as HODE/LA, %7␤-OHCh/tCh significantly correlated with
%␣TQ/␣T.
Finally, we analyzed the correlation between lipid peroxida-
tion products and indicators of fatigue. No significant correlation
was observed between HODEs and indicators of fatigue (data
not shown). The correlation between the 7␤-OHCh/tCh of four
points from 20:30–12:30 and the value of indicators of fatigue
of four points from 14:30–8:30 were analyzed. Significant neg-
ative correlation was observed between %7␤-OHCh/tCh and CFF
(Fig. 4e).

Table 1
Correlation between antioxidants and indicators of fatigue.

Subjective fatigue feeling scores Objective indicator

VAS QSSF CFF

␣T −0.271 (0.013) −0.329 (0.003) 0.132 (0.141)

%␣T −0.221 (0.035) −0.127 (0.150) −0.007 (0.477)
␣TQ/␣T 0.206 (0.047) 0.076 (0.269) −0.071 (0.283)
%␣TQ/␣T 0.122 (0.162) 0.223 (0.035) 0.007 (0.478)
UQ10 H2 −0.222 (0.034) −0.199 (0.052) 0.164 (0.090)
%UQ10 H2 −0.037 (0.384) −0.069 (0.287) −0.030 (0.403)
UQ10 −0.229 (0.031) −0.225 (0.034) 0.064 (0.303)
%UQ10 −0.050 (0.344) −0.175 (0.079) 0.094 (0.224)
total CoQ10 −0.172 (0.083) −0.210 (0.044) 0.189 (0.063)
% total CoQ10 −0.082 (0.256) −0.054 (0.331) −0.089 (0.237)
UQ10 H2 /total CoQ10 −0.063 (0.305) 0.107 (0.193) −0.119 (0.169)
%UQ10 H2 /total CoQ10 0.097 (0.218) −0.069 (0.290) 0.160 (0.098)
UQ10 /total CoQ10 −0.238 (0.026) −0.165 (0.092) −0.063 (0.308)
%UQ10 /total CoQ10 −0.041 (0.372) −0.213 (0.042) 0.198 (0.053)

Pearson correlation coefficient (p-value).

 M. Shichiri et al. / Biological Psychology 94 (2013) 527–533 531

(a) 7β
β-OHCh/tCh
(b)
%7β
β-OHCh/tCh
110 120%

μmol/mol)
100 110%

90 100%

(μ
80 90%

70 80%
14:30 20:30 2:30 8:30 12:30 14:30 20:30 2:30 8:30 12:30

(c) (d)
140 160%
140%
120 120% %tHODE/LA
100%
100 80%
μmol/mol)

tHODE/LA 160%
80 140%
ZE-HODE/LA 120% %ZE-HODE/LA
(μ

60 100%
EE-HODE/LA 80%
40 220%
180%
20 140% %EE-HODE/LA
100%
0 60%
14:30 20:30 2:30 8:30 12:30 14:30 20:30 2:30 8:30 12:30

(e)
200%
n=56
180%
r=0.31273
β-OHCh / tCh

160% P=0.01894
140%
120%
100%
% 7β

80%
60%
40%
30 35 40 45
CFF

Fig. 4. Changes of plasma levels of 7␤-hydroxycholesterol and hydroxyoctadecadienoic acid.

(a) The changes in the ratio of 7␤-OHCh to total cholesterol (7␤-OHCh/tCh).
(b) The changes in the ratios of 7␤-OHCh/tCh as a percentage of the initial ratio (%7␤-OHCh/tCh).
(c) The changes of the ratios of tHODE, ZE-HODE and EE-HODE to linoleic acid.
(d) The percentage changes in the ratios to the initial levels of tHODE/LA (%tHODE/LA), ZE-HODE/LA (%ZE-HODE/LA) and EE-HODE/LA (%EE-HODE/LA).
All data are expressed as mean ±S.E. (bars).
(e) Correlation between the value of CFF of four points from 14:30–8:30 and %7␤-OHCh/tCh of four points from 20:30–12:30. Correlation was analyzed using the Pearson’s
test, a P-value of less than 0.05 was considered significant.

4. Discussion deprivation would mimic the load above. It was carried out to
confirm whether plasma concentrations of antioxidants and lipid-
Those who work overnight, such as doctors and nurses, are peroxidation markers are correlated with indicators of fatigue.
bound not only to work but also to sleep deprivation. In this study, It is reported that CFF estimated the level of alertness (Niepel
we considered that overnight deskwork accompanied by sleep et al., 2013), and thus we consider that CFF can be a good index

Table 2
Correlation between antioxidants and lipid peroxidation products.

␣T %␣T ␣TQ/␣T %␣TQ/␣T

%tHODE/LA −0.021 (0.433) 0.156 (0.102) 0.401 (<0.001) 0.348 (0.002)

%ZE-HODE/LA 0.008 (0.474) 0.150 (0.111) 0.376 (0.001) 0.294 (0.008)
%9ZE-HODE/LA 0.012 (0.462) 0.171 (0.082) 0.371 (0.001) 0.290 (0.009)
%13ZE-HODE/LA 0.002 (0.493) 0.127 (0.152) 0.369 (0.001) 0.288 (0.009)
%EE-HODE/LA −0.103 (0.203) 0.104 (0.199) 0.371 (0.001) 0.429 (<0.001)
%9EE-HODE/LA −0.115 (0.175) 0.110 (0.185) 0.341 (0.002) 0.402 (<0.001)
%13EE-HODE/LA −0.094 (0.223) 0.085 (0.246) 0.401 (<0.001) 0.403 (<0.001)
7␤-OHCh/tCh −0.010 (0.467) −0.348 (0.002) −0.166 (0.090) −0.129 (0.150)
%7␤-OHCh/tCh −0.149 (0.112) −0.007 (0.479) −0.045 (0.360) 0.254 (0.019)

Pearson correlation coefficient (p-value).

532 M. Shichiri et al. / Biological Psychology 94 (2013) 527–533
of effects induced by our overnight work model. CFF has been
used to investigate human fatigue, including mental fatigue of
nurses in a hospital with shift schedules (Marziale & Rozestraten,
1995) and fatigue of ambulance paramedics on a night shift system
(Takeyama et al., 2009). In this study, we showed significant corre-
lations between CFF and subjective indicators of fatigue. It indicates
that CFF could evaluate objectively the level of fatigue induced by
overnight deskwork.
We showed the following: (i) plasma levels of ␣T and CoQ10
decreased and ␣TQ increased trends; (ii) ␣T was negatively corre-
lated with the subjective fatigue feeling score, VAS and QSSF; (iii)
␣TQ negatively correlated with the objective indicator of fatigue,
CFF; (iv) ␣TQ/␣T positively correlated with VAS and QSSF; (v)
7␤-OHCh/tCh increased until an experiment was completed; (vi)
7␤-OHCh/tCh was negatively correlated with CFF. Interestingly,
␣TQ and 7␤-OHCh/tCh were correlated not only with the subjective
fatigue feeling score but also with the objective indicator of fatigue.
These results suggest that plasma levels of ␣TQ and 7␤-OHCh/tCh
are useful as objective indicators for fatigue induced by overnight
deskwork.
The mechanisms of increase of the lipid peroxidation prod-
ucts by overnight deskwork are unclear, however a mechanism
as described below can be considered. Prolonged cognitive work
load accelerates neuronal activity and causes activation of various
neurotransmitter receptors, and rise of neuronal metabolism. The
depolarization of the cell membrane accompanying activation of
an AMPA/Kainic acid type glutamic acid receptor activates the cal-
cium channel on a cell membrane, and induces inflow Ca2+ into the
cell. Elevated intracellular Ca2+ concentration alters a mitochondria
function directly (Bindokas, Jordán, Lee, & Miller, 1996), and makes
the amount of generation of reactive oxygen species increase. Many
of generated reactive oxygen species are scavenged by antioxi-
dants, and oxidation-reduction balance is maintained. However, if
neuronal activity rises over a long period of time by various factors,
proteins, lipids, and nucleic acids will receive oxidization by the
reactive oxygen species.
There are reports that oxidative stress is an important conse-
quence of sleep deprivation. Reimund et al. theorized that sleep
increases the efficacy of antioxidant mechanisms (Reimund, 1994).
Süer et al. reported that sleep deprived rats displayed elevated
malondialdehyde levels and decreased superoxide dismutase and
glutathione peroxidase activities in hippocampus tissue (Süer et al.,
2011). It is suggested that continuous deskwork without sleep
induces oxidative stress and accumulation of lipid peroxidation
products.
In present study, the individual rate of change of ␣TQ (%␣TQ)
tends to increase during daytime (14:30–20:30). Since the data of
%␣TQ during daytime is not influenced by sleep deprivation, the
increase of %␣TQ during daytime is assumed to be influenced only
by continuous deskwork. On the other hand, although 7␤-OHCh
and %7␤-OHCh did not change during 14:30–2:30, they increased
during 2:30–8:30. The increases of 7␤-OHCh and %7␤-OHCh dur-
ing night-time were influenced by the both of prolonged deskwork
and fatigue by sleep deprivation. The influences of these two fac-
tors cannot be discerned. Studies to evaluate each effect are needed
to address this point, however, we speculate from the results in
previous animal experiments (Süer et al., 2011) that the lipid per-
oxidation products were accumulated by sleep deprivation.
This is the first study to demonstrate that plasma ␣TQ level
increased with the accumulation of fatigue. ␣TQ, an oxidation prod-
uct of ␣T, is reported as a useful indicator of oxidative stress.
Ravaglia et al. reported that plasma concentration of ␣TQ was asso-
ciated with cognitive impairment in elderly people (Ravaglia et al.,
2008). Kanazawa et al. reported that the ratios of ␣TQ/␣T were
increased in several tissue types by hyperoxia (Kanazawa, Miyata,
Nagata, Urano, & Matsushima, 2000). Murphy et al. reported that
the concentration of ␣TQ increased in plasma samples taken from
patients who had undergone an aortic cross-clamping operation
(Murphy, Kolvenbach, Aleksis, Hansen, & Sies, 1992). Previously,
we reported that ␣TQ was significantly higher in the plasma and
hippocampus of Ts65Dn mouse model of Down syndrome (Shichiri
et al., 2011). In the present study, increase in ␣TQ was correlated
with fatigue measured by CFF and VAS. As well as 7␤-OHCh, ␣TQ
may indicate the accumulation of oxidation product induced by
continuous deskwork without sleep.
It has been reported previously that ␣TQ can exist in a reduced
form, ␣-tocopheryl hydroquinone (␣TQH2 ) (Moore & Ingold, 1997).
␣TQH2 has been shown to be capable of acting as a radical-
scavenging antioxidant (Niki, 2007; Shi, Noguchi, & Niki, 1999;
Valente & Cavallini, 1985). This may be ascribed to a higher reduc-
ing capacity of ␣TQH2 than UQ10 H2 . It was found previously that
␣TQH2 reduces UQ10 to UQ10 H2 (Neuzil, Witting, & Stocker, 1997).
However, ␣TQH2 was not detected in plasma in this study. It should
be noted that ␣TQH2 could be detected in mouse brain tissue, but
not in plasma (Shichiri et al., 2011).
Maes et al. reported that plasma CoQ10 was significantly lower
in chronic fatigue syndrome patients (Maes et al., 2009a) and in
depressed patients (Maes et al., 2009b). They showed a signifi-
cant negative correlation between CoQ10 and the score of fatigue
scale and they suggested that CoQ10 play a role in the pathophy-
siology of chronic fatigue syndrome (Maes et al., 2009a). Mizuno
et al. reported that oral administration of CoQ10 improved subjec-
tive fatigue during fatigue-inducing workload trials on a bicycle
ergometer (Mizuno et al., 2008). These reports indicated that
plasma CoQ10 level is effective indicator for fatigue. CoQ is not only
a member of the mitochondrial respiratory chain but also is a potent
antioxidant (Matsura, Yamada, & Kawasaki, 1992). In the present
study, we showed that the concentration of UQ10 H2 , a reduced
form of CoQ10 , and total CoQ10 decreased slightly until the end of
the examination. It is possible that the reduction of CoQ10 level is
related to continuous deskwork without sleep.
The increasing trend of 7␤-OHCh/tCh was delayed com-
pared to that of ␣TQ. In addition, 7␤-OHCh/tCh over the period
20:30–12:30 negatively correlated with CFF over the 14:30–8:30
period. Although no significant correlation was observed between
plasma CoQ10 level and indicators of fatigue, CoQ10 may increase
further, later than the plasma levels of 7␤-OHCh/tCh. The mech-
anism underlying the increase in plasma HODE levels before the
increase of fatigue will be explored in future studies.
Although significant statistical differences were not observed
due to the larger differences between individuals, this study indi-
cated an enhancement of lipid peroxidation during overnight
deskwork. This suggests that lipid peroxidation products are poten-
tial objective markers of fatigue that can substantiate the subjective
fatigue feeling scores. In order to verify the specificity of ␣TQ and
7␤-OHCh as fatigue markers, it is necessary to do further compar-
ative studies, as an example; research sampling from doctors and
nurses who are actually doing overnight work. It is possible that
nutritional supplements of antioxidants, ␣T and CoQ10 , can reduce
lipid peroxidation and fatigue. These experiments suggest that the
overnight deskwork loading and measurements of CFF, VAS and
levels of ␣TQ and 7␤-OHCh/tCh could be effective for the evaluation
of anti-fatigue foods and nutritional supplements.
Funding
This study was supported in part by Grants-in-Aid for Young Sci-
entists (A) No. 22680051 and Scientific Research (B) 22300242 from
the Ministry of Education, Culture, Sports, Science, and Technology
of Japan, the research grants from Mitsui Sumitomo Insurance Wel-
fare Foundation and Suzuki Foundation. These grants had no further
 M. Shichiri et al. / Biological Psychology 94 (2013) 527–533
role in the study design; in the collection, analysis and interpreta-
tion data; in the writing of the report; and in the decision to submit
the paper for publication.
Conflict of interest
The authors declare that are no conflicts of interest.
References
Bindokas, V. P., Jordán, J., Lee, C. C., & Miller, R. J. (1996). Superoxide production in
rat hippocampal neurons: Selective imaging with hydroethidine. The Journal of
Neuroscience, 16(4), 1324–1336.
Hartley, L. R., Arnold, P. K., Smythe, G., & Hansen, J. (1994). Indicators of fatigue in
truck drivers. Applied Ergonomics, 25(3), 143–156.
Kanazawa, H., Miyata, C., Nagata, Y., Urano, S., & Matsushima, Y. (2000). Determina-
tion of alpha-tocopherol and alpha-tocopherylquinone in rat tissues and plasma
by high-performance liquid chromatography with electrochemical detection.
Chemical and Pharmaceutical Bulletin, 48(10), 1462–1466.
Kaseda, Y., Jiang, C., Kurokawa, K., Mimori, Y., & Nakamura, S. (1998). Objective eval-
uation of fatigue by event-related potentials. Journal of the Neurological Sciences,
158(1), 96–100.
Kitano, S., Yoshida, Y., Kawano, K., Hibi, N., & Niki, E. (2007). Oxidative status of
human low density lipoprotein isolated by anion-exchange high-performance
liquid chromatography – Assessment by total hydroxyoctadecadienoic acid, 7-
hydroxycholesterol, and 8-iso-prostaglandin F(2alpha). Analytica Chimica Acta,
585(1), 86–93.
Kos, D., Nagels, G., D’Hooghe, M. B., Duportail, M., & Kerckhofs, E. (2006). A rapid
screening tool for fatigue impact in multiple sclerosis. BMC Neurology, 6, 27.
Luczak, A., & Sobolewski, A. (2005). Longitudinal changes in critical flicker fusion
frequency: An indicator of human workload. Ergonomics, 48(15), 1770–1792.
Maes, M., Mihaylova, I., Kubera, M., Uytterhoeven, M., Vrydags, N., & Bosmans, E.
(2009a). Coenzyme Q10 deficiency in myalgic encephalomyelitis/chronic fatigue
syndrome (ME/CFS) is related to fatigue, autonomic and neurocognitive symp-
toms and is another risk factor explaining the early mortality in ME/CFS due to
cardiovascular disorder. Neuro Endocrinology Letters, 30(4), 470–476.
Maes, M., Mihaylova, I., Kubera, M., Uytterhoeven, M., Vrydags, N., & Bosmans, E.
(2009b). Lower plasma Coenzyme Q10 in depression: A marker for treatment
resistance and chronic fatigue in depression and a risk factor to cardiovascular
disorder in that illness. Neuro Endocrinology Letters, 30(4), 462–469.
Maes, M., Kubera, M., Obuchowiczwa, E., Goehler, L., & Brzeszcz, J. (2011). Depres-
sion’s multiple comorbidities explained by (neuro)inflammatory and oxidative
& nitrosative stress pathways. Neuro Endocrinology Letters, 32(1), 7–24.
Marziale, M. H., & Rozestraten, R. J. (1995). Alternating shifts: Mental fatigue in
nurses. Revista LatinoAmericana de Enfermagem, 3(1), 59–78.
Matsura, T., Yamada, K., & Kawasaki, T. (1992). Difference in antioxidant activity
between reduced coenzyme Q9 and reduced coenzyme Q10 in the cell: Studies
with isolated rat and guinea pig hepatocytes treated with a water-soluble radical
initiator. Biochimica Biophysica Acta, 1123(3), 309–315.
Miwa, K., & Fujita, M. (2010). Fluctuation of serum vitamin E (alpha-tocopherol) con-
centrations during exacerbation and remission phases in patients with chronic
fatigue syndrome. Heart and Vessels, 25(4), 319–323.
Mizuno, K., Tanaka, M., Nozaki, S., Mizuma, H., Ataka, S., Tahara, T., Sugino, T., Shirai,
T., Kajimoto, Y., Kuratsune, H., Kajimoto, O., & Watanabe, Y. (2008). Antifatigue
effects of coenzyme Q10 during physical fatigue. Nutrition, 24(4), 293–299.
Moore, A. N., & Ingold, K. U. (1997). Alpha-Tocopheryl quinone is converted into
vitamin E in man. Free Radical Biology and Medicine, 22(5), 931–934.
Murphy, M. E., Kolvenbach, R., Aleksis, M., Hansen, R., & Sies, H. (1992). Antioxidant
depletion in aortic crossclamping ischemia: Increase of the plasma alpha-
tocopheryl quinone/alpha-tocopherol ratio. Free Radical Biology and Medicine,
13(2), 95–100.
533
Neuzil, J., Witting, P. K., & Stocker, R. (1997). Alpha-tocopheryl hydroquinone is an
efficient multifunctional inhibitor of radical-initiated oxidation of low density
lipoprotein lipids. Proceedings of National Academy of Sciences of the United States
of America, 94(15), 7885–7890.
Niepel, G., Bibani, R. H., Vilisaar, J., Langley, R. W., Bradshaw, C. M., Szabadi, E., &
Constantinescu, C. S. (2013). Association of a deficit of arousal with fatigue in
multiple sclerosis: Effect of modafinil. Neuropharmacology, 64, 380–388.
Niki, E. (2007). Tocopherylquinone and tocopherylhydroquinone. Redox Report,
12(5), 204–210.
Nozaki, S., Sugino, T., Kajimoto, Y., Kuratsune, H., Kajimoto, O., & Watanabe, Y. (2009).
Mental and physical fatigue-related biochemical alterations. Nutrition, 25(1),
51–57.
Ravaglia, G., Forti, P., Lucicesare, A., Pisacane, N., Rietti, E., Mangialasche, F., Cecchetti,
R., Patterson, C., & Mecocci, P. (2008). Plasma tocopherols and risk of cognitive
impairment in an elderly Italian cohort. The American Journal of Clinical Nutrition,
87(5), 1306–1313.
Reimund, E. (1994). The free radical flux theory of sleep. Medical Hypotheses, 43(4),
231–233.
Shi, H., Noguchi, N., & Niki, E. (1999). Comparative study on dynamics of
antioxidative action of alpha-tocopheryl hydroquinone, ubiquinol, and alpha-
tocopherol against lipid peroxidation. Free Radical Biology and Medicine, 27(3–4),
334–346.
Shichiri, M., Yoshida, Y., Ishida, N., Hagihara, Y., Tamai, H., & Niki, E. (2011).
␣-Tocopherol suppresses lipid peroxidation and behavioral and cognitive
impairments in the Ts65Dn mouse model of Down syndrome. Free Radical Biol-
ogy and Medicine, 50(12), 1801–1811.
Siktar, E., Ekinci, D., Siktar, E., Beydemir, S., Gülçin, I., & Günay, M. (2011). Protec-
tive role of l-carnitine supplementation against exhaustive exercise-induced
oxidative stress in rats. European Journal of Pharmacology, 668(3), 407–413.
Simonson, E., & Brozek, J. (1952). Flicker fusion frequency; background and applica-
tions. Physiological Reviews, 32(3), 349–378.
Suda, M., Fukuda, M., Sato, T., Iwata, S., Song, M., Kameyama, M., & Mikuni, M. (2009).
Subjective feeling of psychological fatigue is related to decreased reactivity in
ventrolateral prefrontal cortex. Brain Research, 1252, 152–160.
Süer, C., Dolu, N., Artis, A. S., Sahin, L., Yilmaz, A., & Cetin, A. (2011). The effects of
long-term sleep deprivation on the long-term potentiation in the dentate gyrus
and brain oxidation status in rats. Neuroscience Research, 70(1), 71–77.
Takeyama, H., Itani, T., Tachi, N., Sakamura, O., Murata, K., Inoue, T., Takanishi, T.,
Suzumura, H., & Niwa, S. (2009). Effects of a modified ambulance night shift
system on fatigue and physiological function among ambulance paramedics.
Journal of Occupational Health, 51(3), 204–209.
Valente, M., & Cavallini, L. (1985). Inhibition of lipid peroxidation by alpha-
tocopherolquinone and alpha-tocopherolhydroquinone. Biochemistry Interna-
tional, 10(5), 753–761.
Yoshida, Y., Hayakawa, M., Habuchi, Y., & Niki, E. (2006). Evaluation of the dietary
effects of coenzyme Q in vivo by the oxidative stress marker, hydroxyoctadeca-
dienoic acid and its stereoisomer ratio. Biochimica Biophysica Acta, 1760(10),
1558–1568.
Yoshida, Y., Kodai, S., Takemura, S., Minamiyama, Y., & Niki, E. (2008). Simultaneous
measurement of F2-isoprostane, hydroxyoctadecadienoic acid, hydroxye-
icosatetraenoic acid, and hydroxycholesterols from physiological samples.
Analytical Biochemistry, 379(1), 105–115.
Yoshida, Y., Yoshikawa, A., Kinumi, T., Imai, Y., & Niki, E. (2009). Hydroxyoctadeca-
dienoic acid and oxidatively modified peroxiredoxins in the blood of Alzheimer’s
disease patients and their potential as biomarkers. Neurobiology Aging, 30(2),
174–185.
Yoshida, Y., Imai, Y., Sawai, Y., Saito, Y., Fukuda, K., & Niki, E. (2010). Hydrox-
yoctadecadienoic acid as a potential biomarker for oxidative stress in
patients with chronic hepatitis C. Journal of Gastroenterology Hepatology, 25(1),
107–115.
Yoshihara, A., Yamanaka, K., & Kawakami, M. (2009). Effects of polyphenol on visual
fatigue caused by VDT work. International Journal of Occupational Safety and
Ergonomics, 15(3), 339–343.