modulation of dendritic cells and toll-like receptors by marathon running
TRANSCRIPT
ORIGINAL ARTICLE
Modulation of dendritic cells and toll-like receptorsby marathon running
Thomas Nickel • I. Emslander • Z. Sisic • R. David •
C. Schmaderer • N. Marx • A. Schmidt-Trucksass •
E. Hoster • M. Halle • M. Weis • H. Hanssen
Received: 7 December 2010 / Accepted: 18 August 2011 / Published online: 1 September 2011
� Springer-Verlag 2011
Abstract The focus of this study was to assess exercise-
induced alterations of circulating dendritic cell (DC) sub-
populations and toll-like receptor (TLR) expression after
marathon running. Blood sampling was performed in 15
obese non-elite (ONE), 16 lean non-elite (LNE) and 16
lean elite (LE) marathon runners pre- and post-marathon as
well as 24 h after the race. Circulating DC-fractions were
measured by flow-cytometry analyzing myeloid DCs
(BDCA-1?) and plasmacytoid DCs (BDCA-2?). We
further analyzed the (TLR) -2/-4/-7 in peripheral blood
mononuclear cells (rt-PCR/Western Blot) and the cyto-
kines CRP, IL-6, IL-10, TNF-a and oxLDL by ELISA.
After the marathon, BDCA-1 increased significantly in all
groups [LE (pre/post): 0.35/0.47%; LNE: 0.26/0.50% and
ONE: 0.30/0.49%; all p \ 0.05]. In contrast, we found a
significant decrease for BDCA-2 directly after the mara-
thon (LE: 0.09/0.01%; LNE: 0.12/0.03% and ONE: 0.10/
0.02%; all p \ 0.05). Levels of TLR-7 mRNA decreased in
all groups post-marathon (LE 44%, LNE 67% and ONE
52%; all p \ 0.01), with a consecutive protein reduction
(LE 31%, LNE 52%, ONE 42%; all p \ 0.05) 24 h later.
IL-6 and IL-10 levels increased immediately after the run,
whereas increases of TNF-a and CRP-levels were seen
after 24 h. oxLDL levels remained unchanged post-mara-
thon. In our study population, we did not find any relevant
differences regarding training level or body weight. Pro-
longed endurance exercise induces both pro- and anti-
inflammatory cytokines. Anti-inflammatory cytokines,
such as IL-10, may help to prevent excessive oxidative
stress. Marathon running is associated with alterations of
DC subsets and TLR-expression independent of training
level or body weight. Myeloid and plasmacytoid DCs are
differently affected by the excessive physical stress.
Immunomodulatory mechanisms seem to play a key role in
the response and adaptation to acute excessive exercise.
Keywords Marathon running � Immunomodulation �Dendritic cells � Toll-like receptors � Inflammation
Introduction
Marathon running is a very popular sport. Worldwide more
than 700 annual events with up to 40,000 runners are being
Communicated by Susan A. Ward.
T. Nickel (&) � I. Emslander � Z. Sisic � R. David � M. Weis
Medizinische Klinik und Poliklinik 1, Campus Grosshadern,
Ludwig-Maximilians-Universitat Munchen, Marchioninistr.15,
81377 Munich, Germany
e-mail: [email protected]
C. Schmaderer
Department of Nephrology, Klinikum rechts der Isar,
Technische Universitat Munchen, Munich, Germany
N. Marx
Department of Cardiology, Medizinische Klinik I,
Rheinisch-Westfalische Technische Hochschule,
Aachen, Germany
A. Schmidt-Trucksass � H. Hanssen
Division of Sports Medicine, Institute of Exercise and Health
Sciences University Basel, Basel, Switzerland
E. Hoster
Institute for Medical Informatics Biometry and Epidemiology,
Ludwig-Maximilians-Universitat Munchen, Munich, Germany
M. Halle � H. Hanssen
Department of Prevention and Sports Medicine,
Klinikum rechts der Isar (MRI), Technische Universitat
Munchen, Munich, Germany
123
Eur J Appl Physiol (2012) 112:1699–1708
DOI 10.1007/s00421-011-2140-8
organized (Sanchez et al. 2009; Schiffer et al. 2010).
However, little is known about the effects of prolonged
excessive exercise on the immune system. Physical exer-
cise is known to have a dose-dependent effect on the
immune system. While regular moderate exercise has the
potential to reduce the incidence of cardiovascular disease
and enhance immune functions, excessive loads of pro-
longed exercise may increase myocardial burden and
induce immunosuppression (Neilan et al. 2006; Trivax
et al. 2010). Obesity is associated with an increased car-
diovascular burden (Yusuf et al., 2004) and inflammation,
which have been suggested to induce insulin resistance and
the metabolic syndrome (Bastard et al. 2006). Obesity is
associated with increased plasma levels of pro-inflamma-
tory cytokines such as TNF-a and IL-6 (Kasapis and
Thompson 2005). We have recently shown that regular
exercise training affects the immunomodulation depending
on training level and body composition (Nickel et al.
2011). Moreover, we demonstrated an exercise-induced
improvement of microvascular function in obese runners
(Hanssen et al. 2011).
Regular moderate exercise training is considered to
result in long-term anti-inflammatory effects (Kasapis and
Thompson 2005). In contrast, acute bouts of strenuous
exercise have been shown to induce rapid systemic cyto-
kine release and activation of neutrophils and monocytes
(Nieman et al. 2001; Pedersen et al. 2001a; Suzuki et al.
2003). This inflammatory response seems to be balanced
by an enhanced concomitant release of anti-inflammatory
defences such as IL-10 release. The balance of pro- and
anti-inflammatory cytokines has been argued to prevent
and restrict exercise-induced oxidative stress (Ostrowski
et al. 1999; Suzuki et al. 2003).
Mediators of inflammation are closely linked to the
immune response and affect other organs such as the car-
diovascular system.
Dendritic cells (DCs) and toll-like receptors (TLRs) are
crucial mediators of adaptive and innate immune respon-
ses. DCs are omnipresent, highly potent antigen presenting
cells and are able to prime naive T-cells (Binder et al.
2004) (Svec et al. 2007). BDCA-1? DCs mainly trigger a
Th-1-immune response, whereas BDCA-2? DCs trigger a
Th-2-immune response (Kapsenberg 2003; Nickel et al.
2009). BDCA-1/-2 positive cells represent approximately
around 1% of the circulating leukocytes. Circulating
BDCA-1? DCs are reduced in cerebral and myocardial
infarction and are associated with an increased occurrence
of infections (Bastard et al. 2006; Nickel et al. 2010). In
contrast, plasmacytoid DCs (BDCA-2?) play a decisive
role in several autoimmune diseases such as Lupus ery-
thematodes and, in addition, detect viral RNA especially by
TLR-7 and induce a sufficient antiviral immune answer
(Di Domizio et al. 2009). The BDCA-1/-2 DC-ratio may
play a dominant role in virus elimination, for example,
form the respiratory tract (Smit et al. 2008). TLRs, which
are expressed on different types of immune cells, are cru-
cial mediators of the innate immune response. TLRs rec-
ognize so-called pathogen associated molecular patterns
(PAMP) thereby activating immune pathways (Carpenter
and O’Neill 2007).
To date, little is known about the effect of exhaustive
endurance exercise on DC subpopulations and TLRs. One
previous study examined the effect of an acute bout of
exercise on DC-differentiation (Suchanek et al. 2010).
After an intensive 60-min training session, both myeloid
and plasmacytoid DCs were found to be increased in ice-
hockey players. In a recent study, we were able to show
that a 10-week exercise training program increased BDCA-
1 and decreased BDCA-2 expression in obese participants.
Compared to pre-training, TLR-4 and -7 gene and protein
expressions were activated in both lean and obese runners
(Nickel et al. 2011). With respect to TLR expression,
monocyte TLR-2 and -4 expressions were decreased after
1.5 h of strenuous cycling exercise (Lancaster et al. 2005).
On the basis of the above observations, we hypothesize
that modulations of the innate immune system may be key
elements of the systemic response to acute strenuous
exercise. In our study, we examine the effect of marathon
running on key mediators of the immunomodulatory
response, focusing on parameters of the innate and adaptive
immune system. The primary objective was to assess
marathon-induced alterations of DC subsets and TLRs -2/-
4/-7. In addition, CRP as a common inflammatory marker,
pro-inflammatory Th1 cytokines (IL-6, TNF-a) and the
anti-inflammatory Th2 cytokine (IL-10) were assessed.
oxLDL serum levels were detected as a representative
marker for oxidative stress. As a secondary objective, the
study investigated whether the effects of acute bouts of
endurance exercise on the immunomodulation are affected
by training level and/or body composition.
Methods
Study design and screening
Healthy male amateur marathon runners were recruited by
a study appeal in a local newspaper and by written invi-
tations sent to local running clubs. The study was approved
by the hospital’s Ethics Committee. All athletes gave
written informed consent. Subjects were divided into age-
matched groups depending on the extensiveness of train-
ing: lean elite (LE) group included athletes who performed
regular intensive exercise throughout the year and were
scheduled for C55 km/week during the 10-week training
program. Lean non-elite (LNE) and obese non-elite (ONE)
1700 Eur J Appl Physiol (2012) 112:1699–1708
123
group were scheduled for B40 km/week with only seasonal
pre-marathon exercise training. Fasting blood samples
were taken 5–2 days before the marathon and immediately
after the marathon. Runners did not exercise during the
2 days prior to the baseline blood sampling.
Inclusion and exclusion criteria
300 runners replied to the initial study appeal. Male mar-
athon runners were eligible if aged 30–60 years with a
recent history of at least a half-marathon. Obese and
otherwise healthy subjects were defined by a BMI of
C30 kg/m2 and a waist circumference of C102 cm.
Exclusion criteria consisted of known coronary or struc-
tural heart disease, insulin-dependent diabetes mellitus,
drug treatment for type II diabetes or hypertension, renal
dysfunction, chronic inflammatory or musculoskeletal
disease and claustrophobia.
The criteria for the group assignment were body com-
position and ‘‘kilometres run/week’’, which are well
reflected by the individual anaerobic threshold of the dif-
ferent groups. The lean elite (LE) group included athletes
who were scheduled for C55 km/week, while the LNE and
ONE groups were scheduled for B40 km/week during the
10-week training program.
The marathon times of the LE group are in fact not
‘‘elite’’, but the term reflects both training extensiveness as
well as running experience.
Flow-cytometry analysis of leukocytes
100 ll whole blood was incubated with 5–20 lg/ml of the
antibody for 30–60 min at 4�C. Cells were washed in 2%
FCS in PBS. Cell-pellets were resuspended in 2 ml of lysis
buffer (Becton–Dickinson, USA) and incubated for 10 min
at room temperature. For phenotypic analysis, following
staining, the cells were washed and fixed with 1% para-
formaldehyde prior to analysis on a FACS Calibur
cytometer (Becton–Dickinson). For cell survival studies,
unfixed cells were analyzed, and propidium iodide (PI;
3 lg/ml; Sigma, Germany) was used to identify dead cells.
Antibodies were matched with iso-type-controls (Mouse-
c2a-(FITC)/- c1(PE)-FastImmune; BD, USA). A total of
250.000 events were acquired and analyzed using Cellquest
(BD, Belgium). To determine subpopulations of DCs, we
analyzed BDCA-1/-2 expression (Miltenyi Biotec, Ger-
many) following our protocol (Nickel et al. 2011) and the
protocol of Narbutt et al. (2004).
Total RNA isolation and quantitative real-time PCR
TLR-2/-4 and -7 expressions were analyzed using qRT-
PCR. For isolation of mRNA from peripheral blood
mononuclear cells (PBMCs; isolated from 50 ml blood
samples by ficoll-gradient) the total mRNA isolation
RNeasy Mini Kit from Qiagen (Hilden, Germany) was
used according to the instructions provided by the manu-
facturer. 50 pg/tube mRNA (PBMC) was used. cDNA
synthesis and PCR were performed using Omniscript from
Qiagen (Hilden, Germany). The two-step quantitative real-
time PCR system was applied according to the manufac-
turers’ instructions. qRT-PCR was performed in the ABI
PRISMTM 7700 System (Applied Biosystems, Germany).
Data analysis was performed using the delta–delta ct
method (Livak and Petrie 2001; Livak and Schmittgen
2001). Primer sequences (MWG-Biotech AG, Germany)
for the analyzed receptors are:
GAPDH: 1 50-CGG AGT CAA CGG ATT TGG TCG
TAT-30; 2 50-AGC CTT CTC CAT GGT GGT GAA GAC-
30; TLR-2: 1 50-CCA CTT GCC AGG AAT GAA GT-30; 2
50-GAT GCC TAC TGG GTG GAG AA-30; TLR-4: 1 50-TCC ATA AAA GCC GAA AGG TG-30; 2 50-GAT ACC
AGC ACG ACT GCT CA-30; TLR-7: 1 50-TTA CCT GGA
TGG AAA CCA GCT ACT-30; 2 50-TCA AGG CTG AGA
AGC TGT AAG CTA-30.
Western blot
PBMCs were isolated by ficoll-gradient centrifugation and
the cells were lysed by RIPA-buffer. Protein extracts
(40 lg) were separated with 4–12% Bis–TRIS-gel 7.5%
SDS–PAGE (Invitrogen, USA), transferred to a nitrocel-
lulose membrane by electro transfer (200 V for
30–60 min), and blocked with 5% non-fat milk for 1 h at
room temperature. Western Blots were performed for 8
representative members of the LNE group.
The anti-TLR-2 (mouse) (IMGENEX, USA) was
diluted 1:500, the anti-TLR-4 (mouse) (IMGENEX,
USA) was diluted 1:150 and the anti-TLR-7 (rabbit)
(IMGENEX, USA) was diluted 1:150. All antibodies
were incubated at 4�C overnight. Beta actin (goat)
(1:500, Santa Cruz, USA) was used as the internal
standard to ensure that equal amounts of protein were
loaded. Furthermore we used a protein molecular weight
marker (MagicMark, Invitrogen, USA) to visualize the
protein standard bands.
The antigen–antibody complex was visualized using
anti-mouse HRP 1:2500 (Santa Cruz, USA) for TLR-2 and
TLR-4, anti-rabbit HRP 1:1000 (Santa Cruz, USA) for
TLR-7 and anti-goat HRP 1:5000 (Santa Cruz, USA) for
beta actin. An enhanced chemiluminescence detection
system (ECL-Pierce, Invitrogen; USA) was developed
using the X-Omat (Kodak; USA). Quantitative analysis of
Western Blots by densitometry was carried out using the
histogram function in Photoshop 7.0 software. All values
were normalised to the beta actin loading control.
Eur J Appl Physiol (2012) 112:1699–1708 1701
123
Serum concentration of CRP, IL-6, IL-10, TNF-aand oxLDL by ELISA
Serum tubes were centrifuged and the serum was frac-
tionated and frozen at -80�C. Samples were defrosted and
CRP, IL-6, IL-10, TNF-a and oxLDL were examined by
using cytokine-specific ELISA kit according to the manu-
facturer’s instructions (CRP: Biosource, USA; IL-6:
Bender Med-Systems, Austria; IL-10: R and D-systems,
USA; TNF-a: Biosource, USA; oxLDL: Immunoteck,
Germany).
Statistical analysis
Data are presented as ±standard errors of the mean (SEM)
and by boxplots representing the interquartile range (25th–
75th percentile) around the median (dark line in each box).
Our primary focus was to compare obese with lean runners.
In addition, we also compared LNE with LE runners. Our
primary goals were thus pre-specified two-group
comparisons.
The Kolmogorov–Smirnov test was used to determine
whether or not the data were normally distributed. Data
that were not normally distributed were analyzed using the
Wilcoxon signed rank test for paired samples. Unpaired,
not normally distributed samples were evaluated using the
Mann–Whitney U test. Differences between means were
considered significant with p \ 0.05 and highly significant
with p \ 0.01. SPSS (Version 16, IBM-USA) was used for
statistical analysis.
Results
Study group
From the originally recruited 20 participants per group, 15
in the ONE-group and 16 in the LNE- and LE-group
completed the 10-week training program. Drop out reasons
were viral-infections (four cases) and musculoskeletal
injuries (nine cases). During the marathon no participants
dropped out (Table 1).
Flow-cytometry analysis of leukocytes
In myeloid DCs (BDCA-1?), we found no differences
between the groups before marathon (LE 0.35%, min 0.08/
max 0.45; LNE 0.26%, min 0.05/max 0.34 and ONE
0.30%, min 0.17/max 0.45; p = ns).
After the marathon, we found a significant increase in all
three groups LE 0.47% (min 0.12/max 0.71) a 34%
increase compared to their baseline control, LNE 0.50%
(min 0.12/max 0.75) a 92% increase compared to their
baseline control and ONE 0.49% (min 0.04/max 0.85) a
63% increase compared to their baseline control (all
p \ 0.05). 24 h after the marathon, BDCA-1? decreased
significantly in LE (0.36%, min 0.00/max 0.55; p \ 0.05)
and LNE (0.30%, min 0.10/max 0.52; p \ 0.05) (Fig. 1a).
In plasmacytoid (BDCA-2?) DCs, no differences
between the groups were observed at baseline (LE 0.09%,
min 0.00/max 0.23; LNE 0.12%, min 0.04/max 0.19; ONE
0.10%, min 0.00/max 0.22; p = ns). In contrast to myeloid
DCs, the BDCA-2? population decreased significantly in
LE 0.01% (min 0.00/max 0.01) a -91% decrease com-
pared to their baseline control, LNE 0.03% (min 0.00/max
0.09) a -76% decrease compared to their baseline control
and ONE 0.02% (min 0.00/max 0.08) a -75% decrease
compared to their baseline control (all p \ 0.05) after the
marathon.
24 h post-marathon, BDCA-2? cells increased in all
groups compared to their respective baseline controls (LE
0.09%, min 0.00/max 0.14; LNE 0.05%, min 0.00/max 0.09;
ONE 0.08%, min 0.00/max 0.13; all p \ 0.05) (Fig. 1b).
BDCA-1/-2 ratio increased significantly (p \ 0.05) after
the marathon in LE and LNE (but not ONE) group, and
turned back to baseline values 24 h after the marathon
(Fig. 1c).
RT-PCR and Western Blot for TLRs
TLR-2 For TLR-2, we did not find any differences (p = ns)
neither between the groups in response to the marathon
race nor in the course of the next 24 h with respect to
mRNA levels and protein expression (Fig. 2a).
TLR-4 A significant down regulation was seen in the
mRNA expression in LNE after the marathon compared to
baseline (-46 ± 0.08%, p\0.01), which was not reflected
in the protein-expression. In LE and ONE, no significant
Table 1 Age, anthropometric parameters and fitness levels in study
participants at baseline (before marathon run)
LE age
40 ± 7 years
(n = 16)
LNE age
40 ± 6 years
(n = 16)
ONE age
40 ± 6 years
(n = 15)
Weight (kg) 74.4 ± 11 78.5 ± 9 97.6 ± 12.2
BMI (kgm2) 22 ± 1 24 ± 2 29 ± 2
WC (cm) 81 ± 7 86 ± 7 103 ± 7
Body fat (%) 11 ± 2 15 ± 4 24 ± 3
IAT (mmoI/I) 11.1 ± 1.0 12.1 ± 0.7 14.1 ± 1.0
Training distance per
week (km/week)
[55 (all
year)
B40
(seasonal)
B40
(seasonal)
Race results (min) 217 ± 28 235 ± 28 263 ± 32
BMI body mass index, WC waist circumference, IAT individual
anaerobic threshold
1702 Eur J Appl Physiol (2012) 112:1699–1708
123
changes (p = ns) directly after the marathon were found on
mRNA or protein levels.
However, 24 h after the marathon, a significant up
regulation of the mRNA expression was seen in all groups
compared to their respective baseline controls (LE
12 ± 4.54 fold increase; LNE 10.2 ± 2.88 fold increase;
p \ 0.01 for both; ONE 6.8 ± 2.13 fold increase;
p \ 0.05) (Fig. 2b). A concomitant increase in protein
expression during this period was not observed.
TLR-7 After the marathon, we found a down regulation
of mRNA expression for all three groups (p \ 0.05). In LE,
the expression decreased by -44 ± 0.10%, in LNE
-67 ± 0.06% and in ONE -52 ± 0.10% (all p\0.01). In
protein expression, we did not find any changes directly
after the marathon; but 24 h later, we registered a decrease
in LE by -31 ± 3.8%, in LNE by -52 ± 5.2% and in
ONE by -42 ± 4.7% (p \ 0.05).
In contrast 24 h after the run, mRNA expression was
increased compared to baseline (LE 11.7 ± 4.52 fold
increase, LNE 11.7 ± 6.22 fold increase, ONE 15 ± 9.59
fold increase, p \ 0.05) (Fig. 2c).
ELISA for CRP, IL-6, TNF-a, IL-10 and oxLDL
CRP Most frequently measured inflammatory marker.
Baseline CRP levels were within the normal range, did not
differ between the groups (LE: 0.23 mg/dl, min 0.04/max
0.53; LNE: 0.26 mg/dl, min 0.20/max 0.52; ONE:
0.23 mg/dl, min 0.20/max 0.52; p = ns) and did not
change immediately after the marathon. However, a sig-
nificant delayed increase was seen in all three groups 24 h
after the marathon (LE: 1.13 mg/dl, min 0.20/max 2.84;
LNE: 1.51 mg/dl, min 0.67/max 2.57; ONE: 2.19 mg/dl,
min 1.01/max 4.27; p \ 0.01 for all; Fig. 3a).
IL-6 Secreted by T-cells and macrophages. It stimu-
lates the immune response and is released by skeletal
muscles in response to endurance exercise. By inducing
hepatic glucose generation and lipolysis, IL-6 has been
shown to link skeletal muscles contraction with exercise-
induced metabolic needs (Pedersen et al. 2001b). At
baseline, the lowest levels were seen in LE (0.44 pg/ml,
min 0.00/max 1.65), followed by LNE (1.52 pg/ml, min
0.14/max 10.55) and ONE (1.72 pg/ml, min 0.01/max
5.22) (inter group comparison between LE and LNE
p \ 0.05, respectively, LE and ONE p \ 0.01). IL-6
increased significantly immediately after the marathon
(LE: 12.99 pg/ml, min 11.20/max 13.35; LNE: 13.23
pg/ml, min 12.87/max 14.49 and ONE: 13.22 pg/ml, min
12.63/max 14.23; p \ 0.01 for all).
24 h after the marathon, IL-6 levels decreased without
yet reaching baseline levels (LE: 2.52 pg/ml, min 0.15/max
6.72; LNE: 3.62 pg/ml, min 0.63/max 7.87; ONE 3.85
pg/ml, min 1.05/max 9.74; p \ 0.01 compared to baseline
and to immediately after marathon; Fig. 3b).
TNF-a A cytokine and mediator of systemic inflamma-
tion. It is involved in acute phase reactions. At baseline, we
observed differences in systemic TNF-a serum levels
between the groups, which failed significance (LE:
0.22 pg/ml, min 0.00/max 3.47; LNE: 0.77 pg/ml, min
0.00/max 7.26; ONE: 1.20 pg/ml, min 0.00/max 8.85;
p = ns). Directly after the run, changes remained marginal.
However, 24 h post-marathon, TNF-a levels increased in
all three groups (LE: 3.87 pg/ml, min 2.05/max 8.74; LNE:
3.64 pg/ml, min 1.56/max 8.31; ONE: 3.82 pg/ml, min
140
100
120
marathon
baseline
8024h post marathon
40
60
BD
CA
-1/B
DC
A-2
Rat
io
0
20
LE LNE ONE
1.0
0.8
baseline
0.6
marathon
24h post marathon
BD
CA
-1 p
ositi
ve c
ells
in %
of
leuk
ocyt
es
0.2
0.4
LE LNE ONE
0
0.20
0.25
0.15
marathon
24h post marathon
baseline
0.05
0.10
LE LNE ONEBD
CA
-2 p
ositi
ve c
ells
in %
of
leuk
ocyt
es
0
a
b
c
Fig. 1 a–c Flow-cytometer analysis of BDCA-1/-2 expression in
leukocytes Flow-cytometer analysis of BDCA-1/-2 positive cells (in %)
of leukocytes and their ratio at baseline (white bars), immediately after
the marathon (striped bars) and 24 h after the marathon (dotted bars)
Eur J Appl Physiol (2012) 112:1699–1708 1703
123
1.40/max 7.76; p \ 0.05 for all; inter group analysis found
no significant differences at any time point; Fig. 3c).
IL-10 This anti-inflammatory cytokine is mainly
expressed in type 2 T helper cells (Th2) and monocytes. At
baseline, a small difference between the LE and LNE
groups and ONE was detected (LE: 1.48 pg/ml, min 0.41/
max 3.43 and LNE: 1.47 pg/ml, min 0.44/max 3.64 com-
pared to ONE: 0.38 pg/ml, min 0.14/max 1.74; p \ 0.05).
After marathon, IL-10 increased significantly in all three
groups compared to their respective baseline controls (LE:
100
* ** *
1
10
ΔΔ
CT-
fold
incr
ease
s
RT-PCRfor all groups
0.1 LNE ONE
Log
-
** ****LE
TLR-7
marathon
baseline
Western Blot for LNE 121kDa
before marathon marathon 24h post marathon
Actin
100 *ns
24h post marathon
40
60
80
0
20
marathon 24h post marathonbefore marathondens
itom
etry
ana
lysi
s
10
nsRT-PCRfor all groups
1
LE LNE ONE
ΔΔC
T-fo
ld in
crea
ses
LE LNE ONE
TLR-2
Actin
marathon
baseline
Western Blot for LNE 90kDa
before marathon marathon
marathon
24h post marathon
24h post marathon
100
24h post marathon
40
60
80
0
20
before marathon
dens
itom
etry
ana
lysi
s
10
100
****
*RT-PCR
1
ΔΔC
T-fo
ld in
crea
ses
for all groups
0.1LE LNE ONE
**
TLR-4
Actin
Western Blot for LNE
marathon
baseline
95kDa
before marathon marathon 24h post marathon
100
24h post marathon
40
60
80
0
20
marathon 24h post marathonbefore marathon
dens
itom
etry
ana
lysi
s
a b
c
Fig. 2 a–c qRT-PCR of TLR-2/-4/-7 from all groups and Western Blot results using the example of LNE qRT-PCR (top divided into groups)
and Western Blot (bottom 8 representative members of the LNE group) results of TLR-2/-4/-7-expression
1704 Eur J Appl Physiol (2012) 112:1699–1708
123
16.9 pg/ml, min 1.29/max 66.48; LNE: 16.6 pg/ml, min
0.92/max 75.55; ONE: 9.3 pg/ml, min 3.51/max 47.01;
p \ 0.01 for all). IL-10 levels returned to baseline values
24 h after the marathon (LE: 2.2 pg/ml, min 0.67/max
3.49; LNE: 1.9 pg/ml, min 0.78/max 3.51; ONE 1.2 pg/ml,
min 0.01/max 2.62; p \ 0.01 compared to levels immedi-
ately after marathon; Fig. 3d).
In the intergroup analysis, no significant changes
(p = ns) were found for the measurements immediately
after the marathon and 24 h post-marathon.
oxLDL oxLDL is a principal form of cholesterol that
accumulates in atherosclerotic lesions or plaques. oxLDL
levels are generally considered to be associated with
increased oxidative stress. No changes in oxLDL levels
were found between the groups, nor were the oxLDL levels
affected by the marathon running (data not shown; all
p = ns).
Discussion
The effects of acute exhaustive exercise on human DCs
and TLRs have only rudimentarily been studied to date
(Oliveira and Gleeson 2010; Simpson et al. 2009;
Suchanek et al. 2010). This is the first study examin-
ing the effects of marathon running on DC subsets and
TLR.
After the marathon, BDCA-1-positive DCs were
increased in leukocytes whereas BDCA-2-positive DCs
were found to be decreased. Furthermore, we found a
significant decrease of TLR-7 mRNA expression immedi-
ately after the marathon, which was associated with a
decrease of TLR-7 protein expression 24 h post-marathon.
One day after marathon, TLR-7 and TLR-4 mRNA
expression was significantly upregulated compared to
baseline.
30
20
10marathon
24h post marathon
baseline
IL-1
0 co
ncen
trat
ion
in p
g/m
l
0
ONELE LNE
4
baseline
3
marathon
24h post marathon
2
CR
P c
once
ntra
tion
in m
g/dl
1
LE LNE ONE
0
14
12
8
10
marathon
baseline
4
6
24h post marathon
0
2IL-6
con
cent
ratio
n in
pg/
ml
LE LNE ONE
8
6
marathon
24h post marathon
baseline4
αco
ncen
trat
ion
in p
g/m
l
2
TN
F-
LE LNE ONE
0
a b
c d
Fig. 3 a–d Serum levels of CRP, IL-6, IL-10, TNF-a, Serum levels of CRP, IL-6, IL-10, TNF-a, at baseline, immediately after the marathon and
24 h after marathon
Eur J Appl Physiol (2012) 112:1699–1708 1705
123
In contrast to the effects of regular exercise training on
the immunomodulation in obese subjects (Nickel et al.
2011), we found that body weight and fitness levels did not
affect the immunomodulatory stress reaction of the innate
immune system after acute bouts of endurance exercise. In
our study, obese subjects showed significantly increased
IL-6 and decreased IL-10 levels, although differences for
both parameters were in a range of questionable relevance.
The reason for the missing disparity is probably due to the
fact that the obese athletes in our study setting were pre-
trained and not sedentary obese participants.
The inflammatory potential of marathon running was
reflected by the change of specific circulating cytokines.
The inflammatory marker CRP and the Th1 cytokines IL-6
and TNF-a showed a sevenfold increase after marathon.
IL-6 increased immediately post-marathon, whereas the
increase in CRP and TNF-a was delayed for 24 h. IL-10
was significantly upregulated in all groups after the race.
Interestingly, oxLDL levels remained stable throughout the
visits. This suggests that the overall oxidative stress may
have been buffered by a balance of pro-and anti-inflam-
matory cytokines (Rehman et al. 1997). It has previously
been demonstrated that IL-6 release from contracting
muscles can account for the exercise-induced increase in
arterial IL-6 levels (Steensberg et al. 2000). IL-6 is an
important mediator of metabolic adaptations and regulator
of energy status during exercise and has both pro-inflam-
matory and anti-inflammatory properties (Steensberg et al.
2000). It is known, for example, to directly inhibit TNF-aand IL-1. Therefore, the increase of IL-6 post-marathon
seems to underline the balanced cytokine release after
excessive exercise.
On the immunomodulatory level, the reduction of
BDCA-2 positive DCs after the race is of high interest.
These so called plasmacytoid DCs are capable of produc-
ing interferon-alpha (IFN). They play a major role in the
anti-viral immune defense. The essential role of DCs and
TLR-7 in viral infection and host defense has been dem-
onstrated in animal models. Depletion of plasmacytoid
DCs results in a decreased viral clearance and increased
inflammation of the respiratory tract in mice (Smit et al.
2006). Furthermore, TLR-7 deficient mice also demon-
strate a reduced response to the influenza virus (Lund et al.
2004). The mechanisms of the TLR-7 reduction after
marathon are likely to be multifactorial. Next to the
decrease in BDCA-2? cells, the increased levels of cyto-
kines may further suppress TLR expression, which has
previously been shown in vivo in human disease conditions
(Seibl et al. 2003).
In our study, the decrease of BDCA-2? cells was
associated with an increase of BDCA-1? cells. Smit et al.
(2008) found that a reduction of BDCA-2? DCs and
expansion of only BDCA-1 ?DCs significantly enhanced
Th2 type responses to respiratory syncytial virus (RSV). In
contrast, expansion of both BDCA-2? DCs and BDCA-
1?DCs resulted in a decrease in the type Th2 cell answer
and an increase in the Th1 cell response, resulting in a
generally lower immunopathology after virus infection
(Smit et al. 2008).
In accordance, we observed a decrease of BDCA-2 cells
post-marathon and a recovery 24 h later, which was
accompanied by a decrease of TLR-7 levels immediately
after the race and a significant increase 24 h thereafter.
These alterations may implicate an increased susceptibility
to viral infections in the immediate hours after the mara-
thon race. Several studies reported a higher incidence of
viral infection after marathon running (Flegg 1988;
Nieman 1997). However, it remains unclear whether or not
changes in few immunological parameters can result in an
increased susceptibility to infections after prolonged exer-
cise (Ilback et al. 1991; Malm 2006). Based on an animal
model investigating the coherence of acute exercise and the
immune system, it is also possible that subclinical infec-
tions before the race may be responsible for post-race
exacerbations of viral infections (Ilback et al. 1984). None
of the participants in our study showed any signs of
infection pre-race.
The increase of BDCA-1-positive cells may be a result
of cell mobilization from different tissues or derived from
the monocyte pool in the blood. Monocyte subsets have the
potential to differentiate into inflammatory dendritic cells
under inflammatory conditions. One may speculate that the
marathon-related increase of cytokines such as TNF-a may
induce the generation or, more likely, the extravasation of
monocyte-derived dendritic cells, especially the BDCA-1?
type. However, in the complex context of exercise-induced
immunological cascades, this may only be one possible
explanation. It is also plausible to assume that other
mononuclear cell populations, such as lymphocytes and
monocytes, affect inflammatory and immunological pro-
cesses such as the TLR-7 release. More research is needed
to clarify the modulation of the immune system by den-
dritic subsets and TLRs in response to acute strenuous
exercise. To examine the susceptibility to viral infections
after exercise, interferon (IFN) may prove to be of future
interest.
Further research is warranted to analyze whether the
observed exercise-induced alterations of the immune sys-
tem may result in an impairment of vascular structure and
function.
Conclusions
The excessive physical stress is accompanied by differen-
tiated modulations of DC subpopulations and TLRs
1706 Eur J Appl Physiol (2012) 112:1699–1708
123
independent of training level or body composition. After
the marathon, BDCA-1-positive DCs were increased in
leukocytes whereas BDCA-2-positive DCs were found to
be decreased. Furthermore, we found a significant decrease
of TLR-7 mRNA expression immediately after the mara-
thon, which was associated with a decrease of TLR-7
protein expression 24 h post-marathon.
Marathon running seems to induce a balance of pro- and
anti-inflammatory cytokines. IL-6 and TNF-a showed a
sevenfold increase after marathon and IL-10 was upregu-
lated 10- to 20-fold post-marathon. Immunomodulatory
mechanisms seem to play a key role in the response and
adaptation to acute excessive exercise. Whether the mara-
thon-induced immunomodulatory alterations represent a
balanced physiological response to excessive exercise, or
pathophysiological mechanisms that increase the suscepti-
bility to infections remains to be elucidated.
Acknowledgments This work was supported in part by the Heinrich
and Lotte Muehlfenzl Foundation, which provide educational grants
for young scientists.
Conflict of interest The authors have no conflicts of interest to
disclose.
References
Bastard JP, Maachi M, Lagathu C, Kim MJ, Caron M, Vidal H,
Capeau J, Feve B (2006) Recent advances in the relationship
between obesity, inflammation, and insulin resistance. Eur
Cytokine Netw 17:4–12
Binder CJ, Hartvigsen K, Chang MK, Miller M, Broide D, Palinski
W, Curtiss LK, Corr M, Witztum JL (2004) IL-5 links adaptive
and natural immunity specific for epitopes of oxidized LDL and
protects from atherosclerosis. J Clin Invest 114:427–437
Carpenter S, O’Neill LA (2007) How important are Toll-like
receptors for antimicrobial responses? Cell Microbiol
9:1891–1901
Di Domizio J, Blum A, Gallagher-Gambarelli M, Molens JP, Chaperot
L, Plumas J (2009) TLR7 stimulation in human plasmacytoid
dendritic cells leads to the induction of early IFN-inducible genes
in the absence of type I IFN. Blood 114:1794–1802
Flegg PJ (1988) Risk of upper respiratory tract infection and malaria
prophylaxis. Br Med J (Clin Res Ed) 296:1329
Hanssen H, Nickel T, Drexel V, Hertel G, Emslander I, Sisic Z,
Lorang D, Schuster T, Kotliar KE, Pressler A, Schmidt-
Trucksass A, Weis M, Halle M (2011) Exercise-induced
alterations of retinal vessel diameters and cardiovascular risk
reduction in obesity. Atherosclerosis 216:433–439
Ilback NG, Friman G, Beisel WR, Johnson AJ, Berendt RF (1984)
Modifying effects of exercise on clinical course and biochemical
response of the myocardium in influenza and tularemia in mice.
Infect Immun 45:498–504
Ilback NG, Crawford DJ, Neufeld HA, Friman G (1991) Does
exercise stress alter susceptibility to bacterial infections? Ups J
Med Sci 96:63–68
Kapsenberg ML (2003) Dendritic-cell control of pathogen-driven
T-cell polarization. Nat Rev Immunol 3:984–993
Kasapis C, Thompson PD (2005) The effects of physical activity on
serum C-reactive protein and inflammatory markers: a system-
atic review. J Am Coll Cardiol 45:1563–1569
Lancaster GI, Khan Q, Drysdale P, Wallace F, Jeukendrup AE,
Drayson MT, Gleeson M (2005) The physiological regulation of
toll-like receptor expression and function in humans. J Physiol
563:945–955
Livak F, Petrie HT (2001) Somatic generation of antigen-receptor
diversity: a reprise. Trends Immunol 22:608–612
Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression
data using real-time quantitative PCR and the 2(-Delta Delta
C(T)) method. Methods 25:402–408
Lund JM, Alexopoulou L, Sato A, Karow M, Adams NC, Gale NW,
Iwasaki A, Flavell RA (2004) Recognition of single-stranded
RNA viruses by Toll-like receptor 7. Proc Natl Acad Sci USA
101:5598–5603
Malm C (2006) Susceptibility to infections in elite athletes: the
S-curve. Scand J Med Sci Sports 16:4–6
Narbutt J, Lesiak A, Zak-Prelich M, Wozniacka A, Sysa-
Jedrzejowska A, Tybura M, Robak T, Smolewski P (2004) The
distribution of peripheral blood dendritic cells assayed by a new
panel of anti-BDCA monoclonal antibodies in healthy represen-
tatives of the polish population. Cell Mol Biol Lett 9:497–
509
Neilan TG, Januzzi JL, Lee-Lewandrowski E, Ton-Nu TT, Yoerger
DM, Jassal DS, Lewandrowski KB, Siegel AJ, Marshall JE,
Douglas PS, Lawlor D, Picard MH, Wood MJ (2006) Myocardial
injury and ventricular dysfunction related to training levels
among nonelite participants in the Boston marathon. Circulation
114:2325–2333
Nickel T, Schmauss D, Hanssen H, Sicic Z, Krebs B, Jankl S, Summo
C, Fraunberger P, Walli AK, Pfeiler S, Weis M (2009) oxLDL
uptake by dendritic cells induces upregulation of scavenger-
receptors, maturation and differentiation. Atherosclerosis
205:442–450
Nickel T, Hanssen H, Sisic Z, Pfeiler S, Summo C, Schmauss D,
Hoster E, Weis M (2010) Immunoregulatory effects of the
flavonol quercetin in vitro and in vivo. Eur J Nutr 50:163–172
Nickel T, Hanssen H, Emslander I, Drexel V, Hertel G, Schmidt-
Trucksass A, Summo C, Sisic Z, Lambert M, Hoster E, Halle M,
Weis M (2011) Immunomodulatory effects of aerobic training in
obesity. Mediators Inflamm 2011:308965
Nieman DC (1997) Risk of upper respiratory tract infection in
athletes: an epidemiologic and immunologic perspective. J Athl
Train 32:344–349
Nieman DC, Henson DA, Smith LL, Utter AC, Vinci DM, Davis JM,
Kaminsky DE, Shute M (2001) Cytokine changes after a
marathon race. J Appl Physiol 91:109–114
Oliveira M, Gleeson M (2010) The influence of prolonged cycling on
monocyte Toll-like receptor 2 and 4 expression in healthy men.
Eur J Appl Physiol 109:251–257
Ostrowski K, Rohde T, Asp S, Schjerling P, Pedersen BK (1999) Pro-
and anti-inflammatory cytokine balance in strenuous exercise in
humans. J Physiol 515(Pt 1):287–291
Pedersen BK, Steensberg A, Fischer C, Keller C, Ostrowski K,
Schjerling P (2001a) Exercise and cytokines with particular
focus on muscle-derived IL-6. Exerc Immunol Rev 7:18–31
Pedersen BK, Steensberg A, Schjerling P (2001b) Muscle-derived
interleukin-6: possible biological effects. J Physiol 536:329–337
Rehman J, Mills PJ, Carter SM, Chou J, Thomas J, Maisel AS (1997)
Dynamic exercise leads to an increase in circulating ICAM-1:
further evidence for adrenergic modulation of cell adhesion.
Brain Behav Immun 11:343–351
Sanchez LD, Pereira J, Berkoff DJ (2009) The evaluation of cardiac
complaints in marathon runners. J Emerg Med 36:369–376
Eur J Appl Physiol (2012) 112:1699–1708 1707
123
Schiffer T, Montiel G, Hildebrandt U, Predel HG, Knackstedt C
(2010) Der Marathonlauf als gesundheitliches Risiko? Klinikarzt
39:288–291
Seibl R, Birchler T, Loeliger S, Hossle JP, Gay RE, Saurenmann T,
Michel BA, Seger RA, Gay S, Lauener RP (2003) Expression
and regulation of Toll-like receptor 2 in rheumatoid arthritis
synovium. Am J Pathol 162:1221–1227
Simpson RJ, McFarlin BK, McSporran C, Spielmann G, o Hartaigh B,
Guy K (2009) Toll-like receptor expression on classic and pro-
inflammatory blood monocytes after acute exercise in humans.
Brain Behav Immun 23:232–239
Smit JJ, Rudd BD, Lukacs NW (2006) Plasmacytoid dendritic cells
inhibit pulmonary immunopathology and promote clearance of
respirory syncytial virus. J Exp Med 203:1153–1159
Smit JJ, Lindell DM, Boon L, Kool M, Lambrecht BN, Lukacs NW
(2008) The balance between plasmacytoid DC versus conven-
tional DC determines pulmonary immunity to virus infections.
PLoS One 3:e1720
Steensberg A, van Hall G, Osada T, Sacchetti M, Saltin B, Klarlund
Pedersen B (2000) Production of interleukin-6 in contracting
human skeletal muscles can account for the exercise-induced
increase in plasma interleukin-6. J Physiol 529(Pt 1):237–242
Suchanek O, Podrazil M, Fischerova B, Bocinska H, Budinsky V,
Stejskal D, Spisek R, Bartunkova J, Kolar P (2010) Intensive
physical activity increases peripheral blood dendritic cells. Cell
Immunol 266:40–45
Suzuki K, Nakaji S, Yamada M, Liu Q, Kurakake S, Okamura N,
Kumae T, Umeda T, Sugawara K (2003) Impact of a competitive
marathon race on systemic cytokine and neutrophil responses.
Med Sci Sports Exerc 35:348–355
Svec P, Vasarhelyi B, Paszthy B, Korner A, Kovacs L, Tulassay T,
Treszl A (2007) Do regulatory T cells contribute to Th1
skewness in obesity? Exp Clin Endocrinol Diabetes 115:439–
443
Trivax JE, Franklin BA, Goldstein JA, Chinnaiyan KM, Gallagher
MJ, de Jong AT, Colar JM, Haines DE, McCullough PA (2010)
Acute cardiac effects of marathon running. J Appl Physiol
108:1148–1153
Yusuf S, Hawken S, Ounpuu S, Dans T, Avezum A, Lanas F,
McQueen M, Pais P, Varigos J, Liseng L (2004) Effect of
potentially modifiable risk factors associated with myocardial
infarction in 52 countries (the INTERHEART study): case-
control study. Lancet 364:937–952
1708 Eur J Appl Physiol (2012) 112:1699–1708
123