The

Veterinary Journal
The Veterinary Journal 174 (2007) 113121
www.elsevier.com/locate/tvjl

Eects of exercise and oral antioxidant supplementation enriched

in (n  3) fatty acids on blood oxidant markers and
erythrocyte membrane uidity in horses
B. De Moarts a,1, K. Portier b,1, N. Kirschvink a,2, J. Coudert c, N. Fellmann c,
E. van Erck a, C. Letellier d, C. Motta e, J. Pincemail f, T. Art a, P. Lekeux a,*
a

Department for Functional Sciences B41, Faculty of Veterinary Medicine, University of Lie`ge, 4000 Lie`ge, Belgium
b
Equine Department, National Veterinary School of Lyon, France
c
Laboratory of Sport Physiology and Biology, Faculty of Medicine, University of Auvergne, France
d
CHU Hospital SUD, Laboratory of Biochemistry, Rennes, France
e
Faculty of Medicine, Laboratory of Biochemistry A, Rennes, France
f
Probiox S.A., CHU, University of Lie`ge, Belgium
Accepted 2 June 2006

Abstract
The aim of this study was to investigate in a placebo-controlled eld study the eect of a (n  3)-vitamin supplementation on erythrocyte membrane uidity (EMF), oxidant/antioxidant markers and plasmatic x3/x6 fatty acid ratio (FAR) in 12 eventing horses.
Venous blood was sampled at rest before (PRE) and after (POST) a three week treatment period with either the supplement (group
S, n = 6) or a placebo (group P, n = 6) as well as after 15 min (POST E15 0 ) and 24 h (POST E24h) after a standardised exercise test.
The following markers were analysed: EMF, plasma antioxidant capacity of water and lipid soluble components, ascorbic acid, uric acid
(UA), glutathione (reduced: GSH, oxidised: GSSG), vitamin E (Vit E), b-carotene, glutathione peroxidase (GPx) and superoxide dismutase (SOD) activity, selenium, copper (Cu), zinc (Zn), oxidised proteins (Protox), lipid peroxides (Pool) and FAR.
EMF did not dier between group S and P after treatment, but GPx remained unchanged in group S whereas it decreased in group P
and plasma Cu/Zn ratio remained unchanged whereas it increased in group P. FAR were signicantly increased in group S. Exercise
induced a signicant decrease of EMF (POST vs. E24h) in both groups, but which was signicantly lower at E15 0 in group S than in
group P. Exercise induced a signicant increase of UA and ACW (POST vs. E15 0 ) and Protox (POST vs. E24h) in both groups. An exercise-related decrease in GSH and Pool (POST vs. E15 0 ) was found in group P, whereas Vit E and FAR (POST vs. E24h) signicantly
decreased in both groups.
The study showed that exercise induced a decrease in EMF in horses associated with changes of blood oxidative balance. The (x-3)vitamin supplementation tested improved the oxidative balance poorly but delayed the exercise-induced decrease of EMF and increased
the FAR.
 2006 Elsevier Ltd. All rights reserved.
Keywords: Horses; Exercise; Antioxidant; Membrane uidity; (x-3)PUFA

1. Introduction
*

Corresponding author. Tel.: +32 4 366 40 30; fax: +32 4 366 29 35.
E-mail address: Pierre.lekeux@ulg.ac.be (P. Lekeux).
1
B. de Moarts and K. Portier contributed equally to this study.
2
Present address: Animal Physiology, Department for Veterinary
Medicine, University of Namur, Namur, Belgium.
1090-0233/$ - see front matter  2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tvjl.2006.06.001

Physical exercise is characterised by an increase in the

bodys oxygen consumption and is associated with a rise
in reactive oxygen species (ROS) generation (Cazzola
et al., 2003). The relationship between physical exercise
and oxidative tissue damage has been investigated in

114

B. De Moarts et al. / The Veterinary Journal 174 (2007) 113121

numerous studies in men and animals (Sen and Packer,

2000). Exercise has been shown to induce cellular and tissue damage by oxidation of cellular components, such as
membrane lipids, proteins, carbohydrates and (desoxy)ribonucleic acids (Clarkson and Thompson, 2000). Exercise-induced oxidative stress is believed to contribute to
accelerated muscle fatigue and muscle bre damage, leading to exercise intolerance and poor performance in dierent animal species (Sen and Packer, 2000), as well as to a
decreased immune defence of the organism (Nieman,
1997).
Biochemical assessment of oxidative stress is usually
performed using indirect markers such as antioxidant scavengers, antioxidant enzymes and vitamins. Indeed, the
quantication of ROS is dicult in practice because of
their very short half-life and requires highly sophisticated
techniques like spin trapping. ROS alter the structural
properties of erythrocyte cell membranes by changing the
composition and distribution of their lipids, and a reduction in red blood cell membrane uidity, i.e. a reduced
gliding capacity of the phospholipid layers (Hollan,
1996) in response to oxidants has been observed in humans
(Borst et al., 2000; Cazzola et al., 2003). Therefore, erythrocyte membrane uidity (EMF) might be considered as
an indirect marker (Keddad et al., 1996; Cazzola et al.,
2003, 2004) or a biosensor of oxidative stress (Benderitter
et al., 2003).
Recent researches has focused on the role of dietary fats,
which play an important role in animal and human health
by inuencing physiopathological processes (Schmidt et al.,
2005). Several studies have suggested that (x-3) long chain
poly-unsaturated fatty acids (PUFA) have benecial eects
on human health disorders, such as coronary heart disease,
diabetes, cancer and autoimmune disease (Calder, 2004).
The mechanisms of action have yet to be elucidated, but
a role of (x-3) fatty acids in membranes lipid composition
and thus in their function has been suggested (Ma et al.,
2004). However, the potential anti-inammatory properties
of (x-3) fatty acid could be counterbalanced by their proinammatory properties through their capacity to stimulate ROS production in neutrophils and macrophages
(Costabile et al., 2005). Accordingly, the anti-inammatory
eects of (x-3) fatty acids seem to be optimal if an appropriate oxidant/antioxidant balance is maintained using
antioxidants that prevent lipid peroxidation (DAquino
et al., 1991; Palozza et al., 1996; Song et al., 2000).
In horses, numerous experimental and eld studies have
demonstrated that exercise-induced oxidative changes
occur and support the rationale for using antioxidants in
competition horses (Mills et al., 1996; Deaton et al.,
2002; Hargreaves et al., 2002; Marlin et al., 2002; de Moarts et al., 2004, 2005). The eect of exercise on red blood
cell uidity has, to the best of the authors knowledge, not
yet been assessed in the equine species. Regarding PUFA,
some recent studies using orally administrated sh oil
described in vitro and in vivo anti-inammatory eects
(ONeill et al., 2002; Hall et al., 2004a,b) as well as

improved exercise variables (i.e. lower heart rates were

observed during exercise in a sh oil treated horse group)
in horses (OConnor et al., 2004).
The aim of the present study was to investigate in a placebo-controlled eld investigation the eect of a (x-3)-vitamin supplementation on EMF, oxidant/antioxidant
markers and plasmatic x3/x6 fatty acid ratio. The study
was performed in healthy eventing horses, which were
investigated at rest and after exercise.
2. Materials and methods
2.1. Horses
Twelve clinically healthy and regularly trained eventing
horses (six mares and six geldings; mean standard deviation (SD): 10.3 3.5 years) were investigated in Spring
2004. During a run-in period of 6 weeks, all animals were
stabled on straw and were fed twice daily with hay (4 kg/
day) and a commercial cereal blend (23 kg/day) enriched
with vitamins and trace-elements (TWYDIL competition;
providing 30,000 IU Vitamin A, 1000 IU Vitamin E,
750 mg Vitamin C, 60 mg Cu, 150 mg Zn and 2.25 mg
Se per horse per day). The horses were trained daily
and were turned out to grass for 1 h per day. After the
run-in period, the same feeding and training regimen
was maintained for three further weeks but horses
received additionally either an oral x3/x6-vitamin supplement or a placebo.
2.2. Study design
After a 6 week run-in period, two homogenous groups
of six horses (three geldings, three mares, similar age distribution within each group) were formed. One group
received over 3 weeks an oral x3/x6-vitamin supplement
(S), whereas the other group received a placebo (P). The
rst blood sample was taken prior treatment when the
horses were at rest and had not been exercised (PRE
treatment).
Following 3 weeks of x3/x6-vitamin or placebo treatment, blood was sampled at rest (POST treatment),
15 min (E15 0 ) and 24 h (E24h) after a standardised exercise test (ET) performed on a race track. The following
marker analyses were performed: (i) assessment of EMF
by determining the relaxation contraction time (Tc),
which is inversely proportional to EMF (Hollan, 1996);
(ii) assessment of oxidant/antioxidant markers: antioxidant capacity of water soluble components in plasma
(ACW), ascorbic acid (AA), uric acid (UA), glutathione
(reduced: GSH, oxidised: GSSG), glutathione redox ratio:
(GSSG)/(GSH + GSSG), antioxidant capacity of lipid
soluble components in plasma (ACL), Vitamin E, b-carotene (b-car), activity of glutathione peroxidase activity
(GPx) and superoxide dismutase (SOD), selenium (Se),
copper (Cu), zinc (Zn), oxidised proteins (prot-ox), lipid
peroxides (Pool); (iii) plasma x3/x6 fatty acid ratio

B. De Moarts et al. / The Veterinary Journal 174 (2007) 113121

determination (based on the analysis of plasma C18.2x6,

C18.3x6, C18.3x3, C20.2x6, C20.3x6, C20.4x6,
C20.5x3, C22.5x3 and C22.6x3).

115

puncture for analysis of plasma lactate (LA) (Accusport,

Boehringer Mannheim).
2.6. Blood sample collection and processing

2.3. Supplement and placebo: composition and

administration
The x3/x6-vitamin supplement (100 mL/horse/day)
provided 50 mL of sh oil extract (guaranteed composition: 40% of C20.5x3 [eicosapentaenoic acid: EPA] and
20% of C22.6x3 [docosahexanoic acid: DHA]), 5000 mg
of D-a-tocopherol acetate and 5000 mg of copper orotate.
The placebo consisted of 100 mL of the supplements
matrix (water, xanthane gum, arachnid oil, sucran) but
did not provide EPA nor DHA, tocopherol or copper orotate. Supplement and placebo were provided by PAVESCO
AG (Switzerland). Both treatments were distributed once
daily with oats for three consecutive weeks.
A groom systematically veried the ingestion of supplement or placebo. Both treatments were well tolerated by
the horses and no adverse reactions were reported. Riders
and trainer were blinded to the treatment allocation, as
well as technicians performing blood sample analyses.
2.4. Field exercise test
A standardised ET with a nal step at maximal velocity
was performed on the race track of Ghlin (Mons, Belgium).
The ET was adapted to the eventing horses and comprised
a warm-up walk (5 min) and trot (3000 m at low velocity)
followed by three steps of gallop at determined velocity and
length (1000 m at 6.9 m/s, 1000 m at 8.3 m/s, 1000 m at
9.7 m/s). A nal step of 1000 m was performed by each
horse at its individual maximal velocity. A cool-down walk
of 5 min nished the ET. At the end of each gallop step, the
horses were shortly stopped and venous blood was sampled
for lactate determination.
2.5. Exercise monitoring
2.5.1. Measurement and control of speed
During the dierent exercise steps of horses, the rider
had to reach reference points placed along the racetrack
within a very tight time window. Each horse was further
equipped with the Equipilot GPS (Fidelak) system, which
allowed a posteriori and continuous calculation of mean
speed during each race step.
2.5.2. Heart rate monitoring
Heart rate (HR) was determined over 20 s at the end of
each exercise step using a telemetric electro-cardiogram
recorder (Life Scope 8, Nihon Kohden) that was placed
along the race track.
2.5.3. Venous blood lactate determination
The horses were shortly stopped after each gallop step
and venous blood (1 mL) was sampled by jugular vene-

Oxidant/antioxidant marker samples (ACW, AA, ACL,

Vit E, b-car, GPx, SOD, GSH, Hb, Se, Cu, Zn, cholesterol,
UA, packed cell volume, total protein) were processed and
analysed according to previously detailed procedures (de
Moarts et al., 2004, 2005). For determination of Protox
and Pool, EDTA tubes were spun (15 min, 900g) and
plasma was immediately frozen on dry ice and kept at
80 C. Fatty acid and EMF determination were performed on blood sampled in heparin tubes that were kept
at 4 C until analysis (within 36 h of collection).
2.7. Blood marker analysis
2.7.1. Erythrocyte membrane uidity
EMF was determined by electron spin resonance
(ESR) in order to determine the Tc. ESR experiments
were performed with a Bruker ECJ 106 spectrometer
equipped with a resonance cavity TMH 269 using similar
protocol than Keddad et al. (1996). After centrifugation
of the heparin tube (5 min, 900g) and the washing of
red blood cells, a radical marked probe (16 nitroxide stearic acid [C18]) was added to the erythrocyte preparation.
After incubation, study of positioning and rotational
movement of the probe, allowed to determine Tc (Keith
et al., 1973).
2.7.2. Oxidant antioxidant markers
Total antioxidant capacities. Plasma ACW and ACL
were determined by photochemiluminescence detection
using the Photochem assay (Photochem Analytic).
AA (AA). Plasma concentrations of AA were spectrophotometrically determined by using the reduction of 2,6dichloro-phenol-indophenol (Merck n 1.03028).
a-Tocopherol and b-carotene. Plasma concentrations of
Vit E and b-car were determined by high pressure liquid
chromatography (HPLC) on reversed phase columns
(C18 120, 100 45) with an isocratic elution (methanol/
water 98:2) and ultraviolet (UV) detection at 280 and
450 nm, respectively for Vit E and b-car.
Selenium, copper, zinc. Plasma concentrations of traceelements were determined by coupled mass spectrometry
(Agilent 7500a ICP-MS).
Superoxide dismutase (SOD) and glutathione peroxidase
(GPx). Whole blood was analysed spectrophotometrically
using a RANSOD and RANSEL kit, respectively (Randox
Laboratories).
Glutathione. Whole blood glutathione (GSH and GSSG)
was determined spectrophotometrically using the Biooxytech kit GSH/GSSG 412 Assay (Oxis).
Uric acid. Plasma concentration of UA was analysed
spectrophotometrically using an automatic analyser
(Roche Hitachi).

116

B. De Moarts et al. / The Veterinary Journal 174 (2007) 113121

Pool. Plasma concentration of lipid peroxides was analysed spectrophotometrically using OxyStat Kit (Biomedica) according to the method described by Hildebrandt
et al. (2002).
Protox. Plasma concentration of oxidised proteins was
analysed spectrophotometrically allowing determination
of carbonyl structures according to the method described
by Reznick and Packer (1994).
Standardisation of blood markers concentration. Vit E
was standardised by calculating the Vit E/cholesterol ratio,
GPx and SOD by expressing their activities per gram of
haemoglobin (Hb) and Protox by expressing per gram of
protein. Concentration of hydrophilic plasma markers
(AA and UA) and trace-elements was standardised by total
protein concentration whereas lipophilic plasma markers
(Vit E and b-car) were standardised by cholesterol. As
GSH and GSSG were determined in whole blood, their
concentrations were adjusted for variation in PCV. Adjustment was performed using the following equation (Hinchcli et al., 2000):
X adj X obs X obs  M R  M obs =M R 
where Xadj is the adjusted marker concentration, Xobs is the
measured marker concentration at dierent time points of
sampling, MR is the concentration of the standardising element determined at rest and Mobs is the concentration of
the standardising element determined at the time point of
sampling.
2.7.3. Fatty acids
Plasma fatty acid composition was determined, after
lipid extraction and preparation of methyl esters, by gasliquid chromatography (Chrompack CP 9001) coupled
with ame ionisation detectors. The technique was adapted
from Keddad et al. (1996).

not dier between groups (Fig. 1). Regarding oxidant/antioxidant markers shown in Table 1, the following result
description refers to statistical dierences detected for
either group (S or P) eects or time-exercise-related eects
(PRE, POST, E15 0 , E24h) concerning both groups. Withingroup and between-group related dierences at each time
point are indicated in Table 1 by superscripts.
Activity of GPx remained unchanged in group S whereas
it decreased in group P (Table 1). No supplement-related differences were detected for AA, UA, GSH, GSSG, GRR,
ACL, Pool, Protox and SOD (Table 1). In group S, the
plasma Cu/Zn ratio remained unchanged (1.7 0.12 vs.
1.97 0.13; n.s.), whereas it increased in group P
(1.9 0.18 vs. 2.47 0.3; P < 0.05). No signicant dierences between groups were found for Vit E (PRE: Placebo
vs. Supplement: 2.77 0.44 vs. 3.2 0.36 and POST:
2.99 0.41 vs. 3.71 0.49 mg/g chol), b-car (PRE:
0.46 0.27 vs. 0.35 0.33 and POST: 0.56 0.24 vs.
0.41 0.33 lg dL1), Se (PRE: 187 10 vs. 209 12.8
and POST: 179 8.6 vs. 194 11 lg L1), Cu (PRE:
1.05 0.05 vs. 0.94 0.038 and POST: 1.29 0.118 vs.
1.08 0.053 mg L1) and Zn (PRE: 0.56 0.056 vs. 0.56
0.046 and POST: 0.57 0.028 vs. 0.64 0.046 mg L1).
At POST treatment, plasma C20.5x3 and C22.6x3 concentrations and x3/x6 fatty acid ratio were signicantly
increased in group S (Fig. 2ac).
A signicant correlation was observed between Tc and
SOD. Indeed, at PRE treatment, the correlation indicated

Tc (ns)

2.8. Statistical analysis

5

Data are shown as means with their standard error of

mean (SEM). In order to test the treatment eect at rest,
relative changes between PRE and POST (delta) were analysed by analysis of variance (ANOVA). Markers determined at PRE, POST, E15 0 and E24h were analysed by a
mixed linear model for repeated measures (SAS) in order
to assess eects of sampling time, supplement and timesupplement interaction. Correlation analyses were performed separately at T0 and at T1 by linear regression
between x3/x6 ratios and oxidant markers and between
Tc and oxidant markers. Dierences were considered to
be signicant when P < 0.05.
3. Results
3.1. Supplement vs. placebo at rest
At PRE treatment, no signicant dierences between
group S and P were found. At POST treatment, Tc did

*
4

3
0
PRE

POST

E15

E24h

Fig. 1. The relaxationcontraction time (Tc) (n.s.) was determined in

jugular venous blood of 12 healthy trained event horses at rest before
(PRE) and after a supplement period (POST) as well as 15 min (E15 0 ) and
24 h (E24h) after a exercise test, horses were treated during 3 weeks with
an oral placebo (h, n = 6) or an antioxidant supplement enriched in
(n  3) fatty acids (r, n = 6). Data are shown as means SEM. Exercisetime eect is signicant as well as interaction between exercise and
supplement eect, with P < 0.05: (w) signicantly dierent from POST
and (d) signicantly dierent from placebo at the same time, with
P < 0.05.

B. De Moarts et al. / The Veterinary Journal 174 (2007) 113121

117

Table 1
Markers determined in jugular venous blood of 12 healthy trained event horses at rest before (PRE) and after a supplement period (POST) as well as
15 min (E15 0 ) and 24 h (E24h) after an exercise test; horses were treated over 3 weeks with an oral placebo (P, n = 6) or an antioxidant supplement
enriched in (x-3) fatty acids (S, n = 6)
Marker (unit)

Part A
ACW (gmol eq AA mL1)
AA (lmol L1)
UA (lg L1)
GSH (lmol L1)
GSSG (lmol L1)
GRR (%)
Part B
ACL (gmol eq TROLOX mL1)
Pool (lmol L1)
Protox (lmol gprot1)
SOD (IU gHb1)
GPx (IU gHb1)

Group

Time
0

Group eect

Exercise-time eect

PRE (rest)

POST (rest)

E15

P
S
P
S
P
S
P
S
P
S
P
S

50.8 6.7ac
63.7 8.5ac
20.9 4.2ac
27.5 2ac
9 2.9ac
9 3.2ac
839 50ac
801 45ac
92.6 32.6ac
67.2 23.3ac
9.9 3.4ac
7.36 2.1ac

45.7 7.17ac
61.8 6.87ac
15.3 0.9ad
24.4 2.8bc
9 3.7ac
9 2.8ac
1177 71ad
1030 69ad
108 19.5ac
88 25.2ac
7.51 1.06ac
6.69 1.35ac

110 35.7ad
86.5 13.5ad
16.7 0.9ac
21.7 1.9ac
48 23.7ad
21 3.6ad
971 106ae
997 72ad
95 16.1ac
108 12.8ac
8.42 1.32ac
8.8 0.64ac

54.7 8.7ac
67.3 12.6ac
17.5 2.7ac
24.2 3.3bc
9 3.1ac
9 3.6ac
1113 35ad
1137 73ad
86 19.0ac
76 18.0ac
6.64 1.33ac
5.58 1.02ac

n.s.

n.s.

n.s.

n.s.

n.s.

n.s.

n.s.

n.s.

P
S
P
S
P
S
P
S
P
S

3.53 0.71ac
6.7 2ac
61.2 19ac
32.4 10ac
0.052 0.003ac
0.049 0.003ac
1360 128ac
1275 68ac
242 13ac
244 9.9ac

3.72 0.26ac
4.3 1.49ac
83 21ac
45 10ac
0.04 0.003ad
0.038 0.004ad
1377 122ac
1289 64ac
224 13ad
253 11ac

2.71 0.77ac
5.7 2.18ac
66 16ad
43 11ac
0.043 0.004ad
0.036 0.004ad

4.4 0.29ac
6.6 2.12ac
84 19ac
58.8 17ac
0.051 0.002ac
0.056 0.005ac
1284 101ac
1399 89ac
259 8ac
262 16ac

n.s.

n.s.

n.s.

n.s.

n.s.

n.s.

n.s.

E24h

Data are shown as means SEM.

ACW: antioxidant capacity of water soluble components in plasma, AA: ascorbic acid, UA: uric acid, GSH: reduced glutathione, GSSG: oxidised
glutathione, GRR: glutathione redox ratio (GSSG/[GSH + GSSG]), ACL: antioxidant capacity of lipid soluble components in plasma, Protox: oxidised
protein, Pool: lipid peroxide, SOD: superoxide dismutase and GPx: glutathione peroxidase.
Indicates that variation between PRE and POST (delta) value was signicantly dierent between groups. For general comparison, P-values of group
supplement or time eects are indicated in separate columns. Dierences between the placebo group and the supplement group at each sampling point are
signicant (P < 0.05) if the rst superscript is dierent (a or b for with-in column comparisons), whereas dierences between sampling times are signicant
(P < 0.05) if the second superscript is dierent (c, d or e for within-line comparisons); n.s.: non-signicant.

a positive and signicant relation between the both markers (r = 0.68 and P < 0.01) (Fig. 3). At POST treatment,
the same analysis (Tc vs. SOD) was performed separately
within each group. Linear regression remained positive
and signicant in placebo-treated horses (r = 0.86 and
P = 0.029), whereas no more signicant regression was
found in group S (r = 0.086, n.s.).
At PRE, the linear regression indicated a positive and
signicant correlation between the x3/x6 ratio and Pool
(r = 0.74 and P = 0.015). At POST, the analysis was made
separately within groups. The regression remained positive
and signicant in placebo-treated horses (r = 0.76 and
P < 0.05) whereas it became non-signicant in supplement-treated horses.
3.2. Supplement vs. placebo after exercise
Exercise induced a signicant increase of Tc in both
groups at E24h, whereas only group P showed a signicant increase at E15 0 (Fig. 1). Mean value (SEM) of
blood markers assessed at rest (POST value), E15 0 and
E24h are shown in Table 1. Exercise induced a signicant

increase of UA (POST vs. E15 0 ), ACW (POST vs. E15 0 )

and Protox (POST vs. E24h) in both groups. GPx significantly increased (POST vs. E24h) in the placebo group,
but not in the supplemented group. An exercise-related
decrease of GSH (POST vs. E15 0 ) and Pool (POST vs.
E15 0 ) was observed in group P, whereas values for group
S tended to decrease in a non-signicant manner. A signicant exercise-related decrease of Vit E occurred in both
groups (P: Post vs. E24h: 2.99 0.41 vs. 2.73 0.39 mg/g
chol; S: Post vs. E24h: 3.71 0.49vs. 3.45 0.41 mg/g
chol). Markers known to be unaected in short term by
a single exercise (Se, Cu, Zn, b-carotene) were not
assessed. In group S, C20.5x3, C22.6x3 and x3/x6 ratio
signicantly decreased after exercise, whereas only the x3/
x6 ratio decreased in group P (POST vs. E24h, Fig. 2a
c).
3.3. Physiological variables during exercise test
The determination of velocity, HR and lactate at each
step allowed the determination of the speed at which
plasma lactate of 4 mmol L1 (VLA4), a HR of 180 beats/

118

B. De Moarts et al. / The Veterinary Journal 174 (2007) 113121

C22.6 3 (%)
2

SOD (IU/gHb)
y = 472x - 504
R = 0. 68

1.5

1
0.5
0

P < 0.01

2000

1600

1200

PRE

POST

E24h

E15

C20.5 3 (%)

800

0
3

Tc (ns)

Fig. 3. Correlation between superoxide dismutase (SOD) (IU/gHb)

determined in venous jugular blood and corresponding relaxation
contraction time (Tc) (n.s.) at the beginning of the protocol (PRE) in 12
healthy eventing horses. Equation, coecient of correlation and P-value
are indicated.

PRE

POST

E15

E24h

3/6 (% )

20

15
*

and between x3/x6 fatty acid ratio and Pool were found
in plasma. To our knowledge this is the rst report using
EMF as functional marker of oxidative stress in exercising
horses. Exercise-induced changes of EMF could partially
be modulated by oral administration of a (x-3)-vitamin
supplement.

10

4.1. Supplement vs. placebo at rest

5
*

0
PRE

POST

E15

E24h

Fig. 2. (ac) The plasmatic(x-3)/(x-6) fatty acid ratio (x3/x6) (%),

C20.5x3 fatty acid and C22.6x3 fatty acid relative fraction (%) were
determined in jugular venous blood of 12 healthy trained event horses at
rest before (PRE) and after a supplement period (POST) as well as 15 min
(E15 0 ) and 24 h (E24h) after a exercise test, horses were treated during
3 weeks with an oral placebo (h, n = 6) or an antioxidant supplement
enriched in (n  3) fatty acids (r, n = 6). Data are shown as
means SEM: () signicantly dierent from PRE; (w) signicantly
dierent from POST; and (d) signicantly dierent from placebo at the
same time, with P < 0.05.

min (V180) or of 200 beats/min (V200) was reached. No signicant dierences between group P and S were found for
the following variables: VLA4 (P: 8.53 0.32 vs. S:
8.75 0.28 m/s); V180 (P: 8.9 0.44 vs. S: 8.86 0.36 m/
s); V200 (P: 10.4 0.59 vs. S: 10.5 0.39 m/s), maximal
velocity during the last step (P: 11.3 0.42 vs. S:
11 0.29 m/s), maximal lactate concentration after the last
step (P: 13.9 2.63 vs. S: 13.1 1.81 mmol L1).
4. Discussion
This study shows that intense exercise induced signicant modications of EMF and oxidant/antioxidant markers in healthy eventing horses. Signicant and positive
correlations between EMF and erythrocyte SOD activity

A 3 week period of (x-3)-vitamin supplementation did

not signicantly aect EMF in resting horses. In other species, an increase in EMF after antioxidant complementation (Drieu et al., 2000; Ochoa et al., 2003) or x-3-fatty
acid administration (Mueller and Talbert, 1988) has been
described. Besides species- and supplement-related dierences contributing to this lack of eect, the fact that the
horses were trained might further have reduced the treatment eect. In fact, regular exercise has been reported to
increase EMF in man (Cazzola et al., 2003). Given that
the training regimen was unchanged throughout the runin period as well as during the supplementation period, a
training-related increase of EMF might have occurred
before the study was started.
Although other studies investigating the eect of sh oil
supplements in equines also used a supplementation period
of similar duration (34 weeks) (OConnor et al., 2004;
Hall et al., 2004a,b), there is no scientic evidence that this
supplementation period allowed reaching an optimal eect.
Similarly, the dose regimen of sh oil and Vit E chosen for
this study was based on earlier eld investigations (de
Moarts et al., 2005) and human requirements that were
adapted to the metabolic weight of a medium sized warm
blood horse as used in this study.
The (x-3)-vitamin supplement had a signicant eect
on the x3/x6 plasma ratio, which was higher in group
S at POST treatment. Long chain x-3 fatty acid (EPA
and DHA) plasma concentration increased in group S.

B. De Moarts et al. / The Veterinary Journal 174 (2007) 113121

These results indicate that feeding EPA and DHA-containing oil is sucient to change plasma lipid content to
high amounts of (x-3) fatty acids. The slight and non-signicant decrease of the x3/x6 ratio in group P at POST
was probably due to the supplement itself, which mainly
contained arachnid oil.
Despite the provision of Vit E in the (x-3)-vitamin supplement, no signicant changes of plasma Vit E was
observed in group S. Given that (x-3) supplementation
increases tissue membranes susceptibility to peroxidation,
Vit E consumption for membrane protection might have
been increased (Song et al., 2000). Increased PUFA intake
has been associated with decreased Vit E concentration in
rat tissue (Leibovitz et al., 1990). As the placebo treatment
did not provide EPA, DHA and/or Vit E, the results
obtained in group P cannot support or reject this
hypothesis.
Although no signicant eect on plasma Cu and Zn was
observed in group S, the Cu/Zn ratio, which plays an
important role for SOD activity, remained unchanged,
whereas an increase was observed in group P. An increase
of the Cu/Zn ratio might have pro-inammatory eects
(Ojuawo and Keith, 2002) and this observation is therefore
interesting. The activity of GPx remained unchanged in
group S after treatment, whereas a signicant decrease
was noted in group P. GPx activity and/or expression have
been found to be increased after DHA supplementation,
possibly because the implication of GPx in PUFA-peroxides (Venkatraman et al., 1994; Crosby et al., 1996; Lematre et al., 1997).
Recent studies realised in horses indicate that PUFA
might have anti-inammatory properties. Indeed, axseed
supplementation has been associated with a reduction of
skin test reactivity in horses with Culicoides hypersensitivity (ONeill et al., 2002). The comparison of two feeding
regimens enriched in sh or corn oil has shown that the sh
oil regimen inhibited the LPS-induced increase of PGE2 in
bronchoalveolar macrophages, but did not reduce their
phagocytic capacity and tumor necrosis factor-a excretion
(Hall et al., 2004a,b).
4.2. Supplement vs. placebo after exercise
The horses were subjected to a standardised ET whose
monitoring did not show any dierence between groups.
Exercise signicantly decreased EMF in group P and group
S. This decrease was observed as soon as 15 min after the
test and lasted for at least 24 h. An exercise-induced alteration of rheological properties of red blood cells has been
associated with oxidative processes in human athletes (Senturk et al., 2005). As several oxidant/antioxidant markers
were signicantly aected at E15 0 and E24h in our study,
a relation between oxidative stress and decreased EMF
was likely.
Although the EMF was decreased to a similar extent in
group P at E24h, there was no signicant change at E15 0 ,
suggesting that the onset of red blood cell rigidity was

119

delayed. This nding might further explain why sh oil

treated horses were found to have lower heart rates during
a standardised treadmill exercise (OConnor et al., 2004).
Blood viscosity and vascular resistance might be modulated by (x-3) fatty acids and could therefore account for
lower heart rate during exercise. In our study, no treatment-related eects on exercise variables were found, but
it should be remembered that the horses were tested under
eld conditions that are less standardised than during
treadmill tests.
Regarding x3/x6 plasma ratio, a signicant decrease
was observed in both groups between E15 0 and E24h.
The main reason for this decrease was probably the fact
that supplement and placebo administration were stopped
the day of the exercise test and that plasma PUFA concentrations decreased within 24 h.
The positive and signicant correlation that was found
between Tc and SOD further suggests that oxidative processes negatively aect EMF. Cazzola et al. (2003) consider
EMF as a functional marker of oxidative stress, which
might be aected negatively by a single and intense exercise
but could be positively modulated by training.
Among the blood oxidant/antioxidant markers, exercise-related changes were similar to those reported in earlier studies performed by our group (de Moarts et al.,
2004, 2005) but did not seem to be aected by the (x-3)vitamin treatment.
It was observed that fatty acid (c20.5x3 and c22.6x3)
and x3/x6 plasma ratio decreased signicantly at E24h
in both groups. In group S, this decrease was mostly due
to the fact that the treatment was administrated for the last
time prior ET. As plasma PUFA concentration is directly
depending on PUFA ingestion, c20.5x3 and c22.6x3 were
likely to be decreased 24 h after the last administration.
Interestingly, a decrease also occurred in group P, possibly
because exercise-related decrease or redistribution of
PUFA occurs (OConnor et al., 2004).
5. Conclusion
This study has shown that exercise induced a decrease
in EMF in healthy eventing horses, which was associated
with changes of the blood oxidant/antioxidant balance.
In comparison to the placebo group, the administration
of a (x-3)-vitamin supplement did not aect the oxidant/
antioxidant balance but delayed the exercise-induced
decrease of EMF and increased the plasma x3/x6 ratio.
Acknowledgements
We wish to thank the family Caulier for providing
horses and collaboration during the study and M. Leblond
for preparing the manuscript. This study was partially supported by TWYDIL-PROBIOX and the Hippodrome de
Wallonie. B. de Moarts was supported by a grant from
the Fonds pour la Recherche en Industrie et en Agriculture (FRIA), Belgium.

120

B. De Moarts et al. / The Veterinary Journal 174 (2007) 113121

References
Benderitter, M., Vincent-Genod, L., Pouget, J.P., Voisin, P., 2003. The cell
membrane as a biosensor of oxidative stress induced by radiation
exposure: a multiparameter investigation. Radiation Research 159,
471483.
Borst, J.W., Visser, N.V., Kouptsova, O., Visser, A.J., 2000. Oxidation of
unsaturated phospholipids in membrane bilayer mixtures is accompanied by membrane uidity changes. Biochimica et Biophysica Acta
1487, 6173.
Calder, P.C., 2004. n  3 Fatty acids, inammation, and immunity
relevance to post surgical and critically ill patients. Lipids 39, 11471161.
Cazzola, R., Russo-Volpe, S., Cervato, G., Cestaro, B., 2003. Biochemical
assessments of oxidative stress, erythrocyte membrane uidity and
antioxidant status in professional soccer players and sedentary
controls. European Journal of Clinical Investigation 33, 924930.
Cazzola, R., Rondanelli, M., Russo-Volpe, S., Ferrari, E., Cestaro, B.,
2004. Decreased membrane uidity and altered susceptibility to
peroxidation and lipid composition in overweight and obese female
erythrocytes. Journal of Lipid Research 45, 18461851.
Clarkson, P.M., Thompson, H.S., 2000. Antioxidants: what role do they
play in physical activity and health? American Journal of Clinical
Nutrition 72 (Suppl. 2), 637S646S.
Costabile, M., Hii, C.S., Melino, M., Easton, C., Ferrante, A., 2005. The
immunomodulatory eects of novel beta-oxa, beta-thia, and gammathia polyunsaturated fatty acids on human T lymphocyte proliferation,
cytokine production, and activation of protein kinase C and MAPKs.
Journal of Immunology 174, 233243.
Crosby, A.J., Wahle, K.W.J., Duthie, G.G., 1996. Modulation of
glutathione peroxidase activity in human vascular endothelial cells
by fatty acids and the cytokine interleukin-1 beta. Biochimica et
Biophysica Acta 1303, 187192.
DAquino, M., Benedetti, P.C., Di Felice, M., Gentili, V., Tomassi, G.,
Maiorino, M., Ursini, F., 1991. Eect of sh oil and coconut oil on
antioxidant defence system and lipid peroxidation in rat liver. Free
Radical Biology and Medicine 1, 147152.
de Moarts, B., Kirschvink, N., Art, T., Pincemail, J., Michaux, C.,
Cayeux, K., Defraigne, J.O., Lekeux, P., 2004. Impact of training
and exercise intensity on blood antioxidant markers in healthy
Standardbred horses. Equine and Comparative Exercise Physiology
1, 211220.
de Moarts, B., Kirschvink, N., Art, T., Pincemail, J., Lekeux, P., 2005.
Eect of oral antioxidant supplementation on blood antioxidant
status in trained thoroughbred horses. The Veterinary Journal 169,
6574.
Deaton, C.M., Marlin, D.J., Roberts, C.A., Smith, N., Harris, P.A., Kelly,
F.J., Schroter, R.C., 2002. Antioxidant supplementation and pulmonary function at rest and exercise. Equine Veterinary Journal Supplements 34, 5865.
Drieu, K., Vranckx, R., Benassayad, C., Haourigi, M., Hassid, J., Yoa,
R.G., Rapin, J.R., Nunez, E.A., 2000. Eect of the extract of Ginkgo
biloba (EGb 761) on the circulating and cellular proles of polyunsaturated fatty acids: correlation with the anti-oxidant properties of
the extract. Prostaglandins Leukotrienes and Essential Fatty Acids 63,
293300.
Hall, J.A., Van Saun, R.J., Wander, R.C., 2004a. Dietary (n  3) fatty
acids from menhaden sh oil alter plasma fatty acids and leukotriene B
synthesis in healthy horses. Journal of Veterinary Internal Medicine
18, 871879.
Hall, J.A., Van Saun, R.J., Tornquist, S.J., Gradin, J.L., Pearson, E.G.,
Wander, R.C., 2004b. Eect of type of dietary polyunsaturated fatty
acid supplement (corn oil or sh oil) on immune responses in healthy
horses. Journal of Veterinary Internal Medicine 18, 880886.
Hargreaves, B.J., Kronfeld, D.S., Waldron, J.N., Lopes, M.A., Gay, L.S.,
Saker, K.E., Cooper, W.L., Sklan, D.J., Harris, P.A., 2002. Antioxidant status and muscle cell leakage during endurance exercise. Equine
Veterinary Journal Supplements 34, 116121.

Hildebrandt, W., Alexander, S., Bartsch, P., Droge, W., 2002. Eect of Nacetyl-cysteine on the hypoxic ventilatory response and erythropoietin
production: linkage between plasma thiol redox state and O(2)
chemosensitivity. Blood 99, 15521555.
Hinchcli, K.W., Reinhart, G.A., DiSilvestro, R., Reynolds, A., BlosteinFujii, A., Swenson, R.A., 2000. Oxidant stress in sled dogs subjected to
repetitive endurance exercise. American Journal of Veterinary
Research 61, 512517.
Hollan, S., 1996. Membrane uidity of blood cells. Haematologia
(Budapest) 27, 109127.
Keddad, K., Therond, P., Motta, C., Baussan, C., Legrand, A., 1996.
Alterations in erythrocyte membrane uidity and fatty acid composition in glycogen storage disease. Biochimica et Biophysica Acta 1315,
6165.
Keith, A.D., Sharno, M., Cohn, G.E., 1973. A summary and evaluation
of spin labels used as probes for biological membrane structure.
Biochimica et Biophysica Acta 300, 379419.
Leibovitz, B.E., Hu, M.L., Tappel, A.L., 1990. Lipid peroxidation in rat
tissue slices: eect of dietary Vit E, corn oil-lard and menhaden oil.
Lipids 25, 125129.
Lematre, D., Vericel, E., Polette, A., Lagarde, M., 1997. Eects of fatty
acids on human platelet glutathione peroxidase: possible role of
oxidative stress. Biochemical Pharmacology 53, 479486.
Ma, D.W., Seo, J., Switzer, K.C., Fan, Y.Y., McMurray, D.N., Lupton,
J.R., Chapkin, R.S., 2004. n  3 PUFA and membrane microdomains:
a new frontier in bioactive lipid research. Journal of Nutritional
Biochemistry 15, 700706.
Marlin, D.J., Fenn, K., Smith, N., Deaton, C.D., Roberts, C.A., Harris,
P.A., Dunster, C., Kelly, F.J., 2002. Changes in circulatory antioxidant
status in horses during prolonged exercise. Journal of Nutrition 132
(Suppl. 2), 1622S1627S.
Mills, P.C., Smith, N.C., Casas, I., Harris, P., Harris, R.C., Marlin, D.J.,
1996. Eects of exercise intensity and environmental stress on indices
of oxidative stress and iron homeostasis during exercise in the horse.
European Journal of Applied Physiology and Occupational Physiology
74, 6066.
Mueller, B.A., Talbert, R.L., 1988. Biological mechanisms and cardiovascular eects of omega3 fatty acids. Clinical Pharmacology 7, 795
807.
Nieman, D.C., 1997. Immune response to heavy exertion. Journal of
Applied Physiology 82, 13851394.
OConnor, C.I., Lawrence, L.M., Lawrence, A.C., Janicki, K.M.,
Warren, L.K., Hayes, S., 2004. The eect of dietary sh oil
supplementation on exercising horses. Journal of Animal Sciences
82, 29782984.
Ochoa, J.J., Vilchez, M.J., Palacios, M.A., Garcia, J.J., Reiter, R.J., MunozHoyos, A., 2003. Melatonin protects against lipid peroxidation and
membrane rigidity in erythrocytes from patients undergoing cardiopulmonary bypass surgery. Journal of Pineal Research 35, 104108.
Ojuawo, A., Keith, L., 2002. The serum concentrations of zinc, copper and
selenium in children with inammatory bowel disease. Cent Africa
Journal of Medicine 48, 116119.
ONeill, W., McKee, S., Clarke, A.F., 2002. Flaxseed (Linum usitatissimum) supplementation associated with reduced skin test lesional area
in horses with Culicoides hypersensitivity. Canadian Journal of
Veterinary Research 66, 272277.
Palozza, P., Sgarlata, E., Luberto, C., Piccioni, E., Anti, M., Marra, G.,
Armelao, F., Franceschelli, P., Bartoli, G.M., 1996. N  3 fatty acids
induce oxidative modications in human erythrocytes depending on
dose and duration of dietary supplementation. American Journal of
Clinical Nutrition 64, 297304.
Reznick, A.Z., Packer, L., 1994. Oxidative damage to proteins: spectrophotometric method for carbonyl assay. Methods in Enzymology 233,
357363.
Schmidt, E.B., Arnesen, H., de Caterina, R., Rasmussen, L.H., Kristensen, S.D., 2005. Marine n  3 polyunsaturated fatty acid and coronary
disease. Thrombosis Research 115, 163170.

B. De Moarts et al. / The Veterinary Journal 174 (2007) 113121

Sen, C.K., Packer, L., 2000. Thiol homeostasis and supplements in
physical exercise. American Journal of Clinical Nutrition 72 (Suppl. 2),
653S669S.
Senturk, U.K., Yalcin, O., Gunduz, F., Kuru, O., Meiselman, H.J.,
Baskurt, O.K., 2005. Eect of antioxidant vitamin supplement on the
time course of hematological and hemorheological alterations after an
exhausting exercise episode in human subjects. Journal of Applied
Physiology 98, 12721279.

121

Song, J.H., Fujimoto, K., Miyazawa, T., 2000. Polyunsaturated (n  3)

fatty acids susceptible to peroxidation are increased in plasma and
tissue lipids of rats fed docosahexaenoic acid-containing oils. Journal
of Nutrition 130, 30283033.
Venkatraman, J.T., Chandrasekar, B., Kim, J.D., Fernandes, G., 1994.
Eects of n  3 and n  6 fatty acids on the activities and expression of
hepatic antioxidant enzymes in autoimmune-prone NZB NZW F1
mice. Lipids 29, 561568.