Clin Exp Immunol 1998; 111:463–468
Cytokine release and its modulation by dexamethasone in whole blood
following exercise
H. H. SMITS, K. GRÜNBERG, R. H. DERIJK*, P. J. STERK & P. S. HIEMSTRA Department of Pulmonology,
Leiden University Medical Centre, and *Leiden/Amsterdam Centre for Drug Research (LACDR), Centre for Bio-Pharmaceutical
Sciences, Leiden University, Leiden, The Netherlands
(Accepted for publication 7 October 1997)
SUMMARY
Glucocorticoids (GC) play an important role in the treatment of inflammatory diseases like asthma.
However, in selected patients a relative resistance to GC has been reported. Recently, it has been
suggested that GC sensitivity of peripheral blood leucocytes may be regulated in a dynamic fashion
during exercise, in association with activation of the hypothalamic–pituitary–adrenal (HPA) axis. The
aim of the present study was to explore changes in the GC sensitivity of cytokine production by
leucocytes following strenuous exercise by well trained oarsmen. These changes were studied using
lipopolysaccharide (LPS)-induced and anti-CD2/anti-CD28 MoAb-stimulated cytokine release in
whole blood and its modulation by dexamethasone. Following exercise, significant decreases in LPSinduced release of IL-6, tumour necrosis factor-alpha (TNF-a) and IL-10 and anti-CD2/anti-CD28
MoAb-stimulated secretion of interferon-gamma (IFN-g) were observed. In addition, the inhibitory
effect of dexamethasone on both IL-6 and TNF-a secretion was significantly reduced following
exercise, whereas that on IL-10 and IFN-g release was not affected. These exercise-induced changes
were accompanied by activation of the HPA axis, as indicated by an increase in circulating
adrenocorticotropic hormone (ACTH) levels immediately following exercise. The results from the
present study suggest that GC sensitivity of whole blood cytokine release can be regulated in a dynamic
fashion and that this can be assessed using an ex vivo stimulation assay. Moreover, since dexamethasone
responsiveness of anti-CD2/anti-CD28 MoAb-induced IFN-g secretion in whole blood is not affected
by exercise, it may suggest that exercise differentially affects monocytes and lymphocytes. The
dynamic regulation of steroid responsiveness of leucocytes, as observed in the present study, could
have important consequences for the effectiveness of GC treatment in inflammatory diseases.
Keywords
cytokines
whole blood
exercise
INTRODUCTION
Glucocorticoid hormones display potent immunosuppressive and
anti-inflammatory activity [1], and therefore synthetic glucocorticoids (GC) have been found to be particularly useful in the
treatment of various inflammatory diseases, such as asthma,
rheumatoid arthritis or systemic lupus erythematosus [1]. The
anti-inflammatory actions of GC affect many cell types through a
receptor-mediated process, and include profound effects on leucocyte distribution that are mediated in part by inhibition of production and release of many proinflammatory mediators, including
cytokines [2]. In addition to these anti-inflammatory effects, GC
may also stimulate endogenous anti-inflammatory mechanisms by,
Correspondence: Dr P. S. Hiemstra, Department of Pulmonology,
Building 1, C3-P, PO Box 9600, 2300 RC Leiden, The Netherlands.
q 1998 Blackwell Science
dexamethasone
glucocorticoids
for example, stimulating the production of proteinase inhibitors [3]
and anti-inflammatory cytokines [4].
However, GC are not always successful in anti-inflammatory
therapy. In a minority of patients with chronic inflammatory or
immune diseases, an acquired form of GC resistance is reported
[5,6]. In addition to therapeutic unresponsiveness, in vitro effects
at the cellular level have been demonstrated. Peripheral blood
mononuclear cells (PBMC) of GC-resistant asthmatic patients
[7,8] or patients with Alzheimer’s disease [9] are less responsive
to dexamethasone inhibition of phytohaemagglutinin (PHA)induced proliferation. Moreover, the binding affinity of the GC
receptor on PBMC in steroid-resistant asthma patients was
reported to be markedly reduced [10].
In addition to interindividual differences in GC responsiveness,
also intra-individual differences have been reported, suggesting a
463
464
H. H. Smits et al.
dynamic regulation of GC sensitivity. Recent in vitro [11] and in
vivo [12] data show that GC may down-regulate the expression of
their own receptor. In addition, it has been demonstrated that
strenuous exercise [13] or academical stress [14] induce a transient
decrease in GC sensitivity of leucocytes ex vivo. In both situations,
activation of the hypothalamic–pituitary–adrenal (HPA) axis, with
elevated levels of cortisol, has been demonstrated. These findings
suggest an association between stimulation of the HPA axis and
temporal decreases of leucocyte GC sensitivity. One could postulate that other conditions that are known to stimulate HPA activity,
such as mycobacterial [15] or viral infections [16], may also affect
GC sensitivity. Such decreased responsiveness could have therapeutic consequences when occurring in patients with inflammatory diseases.
It has recently been shown that the production of various
cytokines is differentially affected by GC, both in vivo and in vitro
[17,18]. Therefore, the aim of the present study was to investigate the
effect of strenuous exercise on the regulation of cytokine production
in whole blood ex vivo by dexamethasone. To that end we analysed
the production of various lipopolysaccharide (LPS)-induced and antiCD2/anti-CD28 MoAb-induced pro- and anti-inflammatory cytokines and their modulation by dexamethasone before and after a
strenuous exercise challenge in competitive oarsmen.
MATERIALS AND METHODS
Exercise challenge
Nine healthy, well trained competitive oarsmen (six males and
three females, 18–25 years old), free of asthmatic and endocrine
disorders, participated in the study that was approved by the
Medical Ethical Board of Leiden University Medical Centre.
Competitive oarsmen obeyed strict rules of their rowing club: no
alcohol, no drugs, 8 h of sleep every night. During their visit to our
laboratory the oarsmen performed a routine rowing ergometer test
(Concept II indoor rower; Morrisville, VT), which is a monthly
recurring part of their training programme. During the exercise
challenge, on-line measurements of exhaled O2 and CO2 were
made, in order to calculate O2 uptake (VO2) and CO2 production
(VCO2) (Oxycon Record; Jaeger, Würzburg, Germany). Moreover,
heart rate and cardiac function were monitored by electrocardiography. The test consisted of 3-min exercise periods at constant
power output, alternated with 0·5-min rest periods, until exhaustion
occurred. After every 0·5-min rest period, the power output was
increased stepwise by 40 W (males) or 30 W (females) from baseline (males 240 W; females 150 W). The total exercise challenge
lasted between 15 and 20 min, depending on personal fitness. The
VO2,max was expressed as percentage of predicted to assess the
extent of strenuous exercise per individual. An i.v. catheter was
placed in one forearm vein 40 min before the start of the exercise
challenge. Immediately, blood samples were drawn for the whole
blood stimulation assay, the assessment of basal adrenocorticotropic hormone (ACTH) and cortisol plasma levels and total
leucocyte count and differentiation. Blood was also collected
immediately after cessation of the exercise challenge for determination of ACTH and cortisol plasma levels. Finally, blood was
collected 20 min after cessation of the exercise challenge, for a
second whole blood stimulation assay and total leucocyte count
and differentiation.
Hormone assays
Plasma cortisol and ACTH levels were measured in blood that was
collected in iced ethylenediamine tetraacetate (EDTA) tubes
(Becton Dickinson, Rutherford, NJ) using a fluorometric assay
(cortisol; Abbott, TDx, Abbott Park, IL) and a radioimmunoassay
(ACTH; Nichols Diagnostics, San Diego, CA).
Leucocyte differentiation
Venous blood collected in EDTA tubes (Becton Dickinson) was
used to assess leucocyte counts and differential cell counts by
automated blood count analysis (Technicon H1; Technicon, Tarrytown, NY).
Whole blood stimulation assay and inhibition by dexamethasone
Venous blood was collected in heparinized tubes. Aliquots of
0·5 ml of dexamethasone-21-phosphate (Sigma Chemical Co., St
Louis, MO; final concentration 1, 10, 100 nM in pyrogen-free
saline) or 0·5 ml of pyrogen-free saline were added to 4·5 ml
whole blood, and incubated for 10 min at room temperature.
Blood was stimulated with LPS (final concentration 3 ng/ml;
LPS from Salmonella typhosa, Sigma) or a mixture of MoAbs to
CD2 and CD28 (anti-CD2 MoAbs CLB-T11.1/1 and CLB-T11.1/
2, final dilution 1:1000 of murine ascites; anti-CD28 MoAb CLBCD28/1, 1 mg/ml; CLB, Amsterdam, The Netherlands) dissolved in
sterile pyrogen-free saline. Twenty-five microlitres of LPS, antiCD2/anti-CD28 MoAbs or saline were added to 225 ml whole
blood containing dexamethasone in 96-well flat-bottomed plates
(Greiner, Alphen a/d Rijn, The Netherlands). After a 24-h incubation period in a humidified atmosphere at 378C in 5% CO2, the
plates were centrifuged for 10 min at 750 g. Plasma samples were
collected and stored at –808C until further analysis.
Cytokine assays
Levels of IL-6, tumour necrosis factor-alpha (TNF-a), interferongamma (IFN-g), IL-4 and IL-10 in plasma, harvested from the
whole blood assays, were measured by ELISA (CLB) according to
the manufacturer’s instructions. The lower limits of detection
were: 10 pg/ml (IL-6), 10 pg/ml (TNF-a), 20 pg/ml (IFN-g),
1·5 pg/ml (IL-4) and 5 pg/ml (IL-10). Briefly, 96-well plates were
coated with a murine MoAb to IL-6, TNF-a, IFN-g, IL-4 or IL-10
and subsequently incubated with the sample. Next, the plates were
incubated with a biotinylated MoAb to these cytokines. Finally,
peroxidase-conjugated streptavidin was added, and 3,3’,5,5’-tetramethylbenzidine (Sigma) was used as a substrate.
Statistic analysis
Cytokine production was expressed as the cytokine level per ml
plasma or as the amount of cytokine per monocyte or lymphocyte.
Dexamethasone sensitivity was expressed as percentage inhibition
of stimulated cytokine production. Differences in the time course
and in dexamethasone dose–response curves of LPS-induced or
anti-CD2/anti-CD28 MoAb-stimulated whole blood were determined by multiple variate analysis of variance (MANOVA). Significant MANOVA effects were explored using paired Student’s t-tests.
Likewise, changes in cell counts or plasma ACTH and cortisol
concentrations were calculated by Student’s paired t-tests. Nonparametric testing (Wilcoxon, matched pairs) was used for data
that were not normally distributed. Data are presented as mean and
s.e.m. P < 0·05 was considered to be statistically significant.
RESULTS
Exercise challenge
The exercise challenge in females had a duration of 15·5 6 0·0 min,
q 1998 Blackwell Science Ltd, Clinical and Experimental Immunology, 111:463–468
Exercise and cytokine release
*
***
25
100
4
0
After
Fig. 1. Effect of exercise on circulating adrenocorticotropic hormone
(ACTH) levels. Plasma ACTH levels were determined in blood samples
obtained 40 min before and immediately after exercise. *P < 0·01 versus
before exercise.
all reaching the maximal power output of 270 W. Then VO2,max was
138·4 6 12·3% predicted. The exercise challenge in males had a
duration of 16·0 6 0·92 min and the maximal power output was
between 360 W and 440 W, with VO2,max of 135·8 6 6·6% predicted.
Exercise-induced changes in endogenous hormone levels
Immediately after cessation of the exercise challenge, circulating
ACTH levels were significantly increased compared with preexercise levels, but cortisol levels were not affected at this time
(Fig. 1).
Changes in leucocyte numbers after exercise
Twenty minutes following cessation of the exercise challenge,
there was no change in the total number of circulating leucocytes,
neutrophils or basophils, whereas the number of circulating monocytes, lymphocytes and eosinophils was significantly decreased
(Table 1).
Effect of exercise on LPS-induced cytokine production in whole
blood
In pilot experiments, we selected a concentration of 3 ng/ml LPS to
Control
(c)
*
3
2
1
0
Control
LPS
4
***
(b)
3
2
1
0
LPS
30
IFN-γ (ng/ml)
IL-6 (ng/ml)
50
0
IL-10 (ng/ml)
ACTH (ng/ml)
200
(a)
TNF-α (ng/ml)
75
Before
465
***
Control
LPS
**
(d)
20
10
0
Control Anti-CD2/CD28
Fig. 2. Effect of exercise on lipopolysaccharide (LPS)-induced IL-6 (a),
tumour necrosis factor-alpha (TNF-a) (b) and IL-10 synthesis (c); and on
anti-CD2/anti-CD28-stimulated IFN-g release (d). Blood was obtained
40 min before (A) and 20 min after the ending of the exercise (B) and
stimulated with 3 ng/ml LPS or MoAbs against CD2 and CD28 or saline
during 24 h. Since the IFN-g levels were not normally distributed, nonparametric testing (Wilcoxon, matched pairs) was used to analyse statistical
significance. *P < 0·05; **P < 0·04; ***P < 0·01 versus before exercise.
study cytokine release in whole blood because at this concentration
a submaximal effect was observed. Following exercise, a significant decrease in LPS-induced IL-6, TNF-a and IL-10 release was
observed (Fig. 2a–c). The main source of IL-6, TNF-a and IL-10
in LPS-stimulated whole blood is the monocyte. Since the number
of monocytes was decreased following exercise, we calculated the
cytokine production per monocyte. This demonstrated that, also
when expressed per monocyte, the decrease in IL-6 and TNF-a
release was still significant (Table 2). In contrast, the amount of
IL-10 released per monocyte was not decreased following
exercise.
Effect of exercise on anti-CD2/anti-CD28 MoAb-induced cytokine
production in whole blood
In addition to LPS, we also employed a lymphocyte-specific
stimulus using MoAbs against CD2 and CD28 to study cytokine
Table 1. Numbers of circulating leucocytes before and after
exercise
Table 2. Cytokine production in whole blood calculated
on a per cell basis before and after exercise
Cytokine
Before
After
22 6 3
1·1 6 0·2
962
1·1 6 0·7
16 6 3*
0·4 6 0·1**
762
0·1 6 0·03*
Cell number (109/l)
Cell type
Eosinophils
Basophils
Neutrophils
Lymphocytes
Monocytes
Total leucocytes
Before exercise
After exercise
0·20 6 0·12
0·05 6 0·02
3·60 6 0·93
2·25 6 0·12
0·54 6 0·15
6·73 6 1·08
0·12 6 0·06*
0·04 6 0·02
4·50 6 2·62
1·61 6 0·19*
0·46 6 0·15*
6·81 6 2·60
Circulating leucocytes were determined 40 min before
exercise and 20 min after exercise. Data are presented as
mean 6 s.e.m. (n ¼ 7).
*P < 0·02 versus before exercise.
q 1998 Blackwell Science Ltd, Clinical and Experimental Immunology, 111:463–468
IL-6 (fg/monocyte)
TNF-a (fg/monocyte)
IL-10 (fg/monocyte)
IFN-g (fg/lymphocyte)
Whole blood collected 40 min before or 20 min after
exercise was stimulated with lipopolysaccharide (LPS;
3 ng/ml) to study IL-6, tumour necrosis factor-alpha
(TNF-a) and IL-10 production, or with anti-CD2/antiCD28 MoAbs to study IFN-g production. Data are
presented as mean amount of cytokine produced per
monocyte (for IL-6, TNF-a and IL-10) or lymphocyte
(for IFN-g) 6 s.e.m. (n ¼ 7).
*P < 0·04; **P < 0·01 versus before.
H. H. Smits et al.
(a)
75
50
*
25
0
Inhibition of IL-10 (%)
–25
0.1
100
1.0
10.0
Dex (nM)
100.0
Inhibition of IFN-γ (%)
Inhibition of IL-6 (%)
100
Inhibition of TNF-α (%)
466
(c)
75
50
25
0
–25
–50
0.1
1.0
10.0
Dex (nM)
100.0
100
***
(b)
75
***
50
**
25
0
–25
0.1
100
1.0
10.0
Dex (nM)
100.0
1.0
10.0
Dex (nM)
100.0
(d)
75
50
25
0
–25
–50
0.1
Fig. 3. Effect of exercise on dexamethasone sensitivity of lipopolysaccharide (LPS)-induced IL-6 (a), tumour necrosis factor-alpha (TNF-a)
(b) and IL-10 synthesis (c) and of anti-CD2/anti-CD28-stimulated IFN-g release (d). Blood was drawn 40 min before (W) and 20 min after the
ending of the exercise (X) and incubated for 24 h with 3 ng/ml LPS or MoAbs against CD2 and CD28 in the presence or absence of various
concentrations of dexamethasone (1, 10 or 100 nM). *P < 0·05; **P < 0·04; ***P < 0·02; ****P < 0·01 versus before exercise.
production in whole blood. In pilot experiments we observed that,
compared with LPS, anti-CD2/anti-CD28 stimulation resulted in
low levels of IL-6 and TNF-a release, but higher levels of IFN-g.
IL-4 release was detected in only one subject, and therefore not
further analysed. Anti-CD2/anti-CD28 stimulation did not increase
IL-10 release above the levels observed in unstimulated blood
following exercise (Fig. 2c). A significant decrease in IFN-g
production in stimulated blood was observed after exercise (Fig.
2d). When analysing IFN-g production on a per cell basis, a
significant decrease per lymphocyte was found following exercise
(Table 2).
Effect of exercise on the modulation of LPS-induced cytokine
production in whole blood by dexamethasone
LPS-induced IL-6, TNF-a and IL-10 production was inhibited by
dexamethasone in a dose-dependent manner. Following exercise,
the effect of dexamethasone on LPS-induced IL-6 and TNF-a
release was significantly decreased (MANOVA, P < 0·04 and P < 0·01
versus before exercise; Fig. 3a,b). No changes were detected in the
effect of dexamethasone on IL-10 synthesis after exercise
(MANOVA, P > 0·28 versus before exercise; Fig. 3c). At 1 nM of
dexamethasone, a significant decrease in inhibition of IL-6 production was observed (Fig. 3a). At higher concentrations of
dexamethasone, there was no difference in the effect of dexamethasone on IL-6 release before and after exercise. In contrast,
the post-exercise dexamethasone sensitivity of LPS-induced
TNF-a production was significantly reduced at all three concentrations of dexamethasone tested (Fig. 3b).
Effect of exercise on the modulation of anti-CD2/anti-CD28induced cytokine production in whole blood by dexamethasone
IFN-g release in anti-CD2/anti-CD28 MoAb-stimulated blood was
inhibited by dexamethasone in a dose-dependent fashion. However, the inhibitory effect of dexamethasone on IFN-g release in
whole blood was not changed following exercise (MANOVA, P > 0·96
versus before exercise; Fig. 3d). In contrast to pre-exercise antiCD2/anti-CD28 MoAb-stimulated IL-10 levels, post-exercise stimulated IL-10 levels were not significantly increased from baseline. Therefore changes in IL-10 dexamethasone sensitivity were
not analysed.
DISCUSSION
The results from the present study show that strenuous exercise in
trained oarsmen results in a decreased release of IL-6, TNF-a, IL10 and IFN-g in whole blood stimulated ex vivo. These exerciseinduced changes were accompanied by significant decreases in the
number of circulating monocytes, lymphocytes and eosinophils.
However, this decrease in cell numbers appeared not to account
fully for the changes in IL-6, TNF-a or IFN-g production, but
might explain the decreased IL-10 production. Second, a postexercise decrease in the effect of dexamethasone on LPS-induced
IL-6 and TNF-a production was observed. In contrast, the sensitivity of LPS-induced IL-10 and anti-CD2/anti-CD28 MoAbinduced IFN-g production to inhibition by dexamethasone was
not affected following exercise. These exercise-induced changes
were accompanied by activation of the HPA axis, as indicated by
the increase in circulating ACTH levels immediately following
exercise. Previously, it has been shown that an increase in ACTH
levels immediately following exercise is associated with a maximal increase in cortisol 20 min after exercise [13]. Taken together,
these findings support the existence of a transient inducible form of
GC resistance.
In the present study we observed reduced cytokine production
q 1998 Blackwell Science Ltd, Clinical and Experimental Immunology, 111:463–468
Exercise and cytokine release
in whole blood stimulated either with LPS or with MoAbs against
CD2 and CD28 following exercise, accompanied by activation of
the HPA axis. A similar association between activation of the HPA
axis and decreased PHA-induced proliferation of PBMC or
reduced LPS- or PHA-stimulated cytokine release in whole
blood ex vivo has been reported during morning peaks of cortisol
circadian rhythm [19], academical stress [14], sepsis [20,21] and
small cell lung cancer [22].
We considered several mechanisms that may explain decreased
cytokine release as observed in the present study. First, a decrease
in cytokine production might be the direct result of increased levels
of endogenous cortisol. Second, changes in specific leucocyte
subsets following exercise may explain our observations, since in
the present study we observed a decrease in the number of
circulating monocytes and lymphocytes post-exercise. However,
after correction for the number of monocytes (for IL-6, TNF-a and
IL-10 release [23,24]) or lymphocytes (for IFN-g release [25]), the
release of IL-6, TNF-a and IFN-g was still significantly decreased
following exercise, whereas IL-10 release was not. Third, in septic
patients decreased TNF-a and IL-6 production was accompanied
by decreased stability of mRNA encoding TNF-a and IL-6 resulting in lower mRNA levels [20]. It is possible that exercise also
results in decreased stability of mRNA encoding selected cytokines in circulating leucocytes, and subsequent lower production of
these cytokines. Another possibility is that exercise increases the
production of compounds such as IL-4, IL-10 or prostaglandin E2
that may suppress cytokine production [26], but this possibility is
not supported by our own or other studies. Finally, decreased
TNF-a secretion following exercise might be explained by exercise-induced inhibition of processing of membrane-bound precursor TNF-a to the secreted mature protein [27,28].
In addition to effects on stimulated cytokine production in
whole blood, we also observed marked effects of exercise on the
sensitivity of cytokine production to dexamethasone. Similar
results regarding steroid responsiveness of LPS-induced IL-6
production in whole blood have been reported by DeRijk et al.
using treadmill exercise in moderately trained subjects [13]. In
addition, Sauer et al. reported a decreased cortisol sensitivity in
vitro of PHA-induced PBMC proliferation in subjects undergoing
academical stress [14]. We extended these observations by demonstrating not only a decreased sensitivity for dexamethasone of LPSinduced IL-6 secretion, but also of TNF-a secretion, whereas the
sensitivity of LPS-induced IL-10 secretion was not affected following strenuous exercise in well trained oarsmen. In addition, we
also used the monocyte-independent T cell-specific stimulus
anti-CD2/anti-CD28. Using this stimulus, IFN-g secretion was
markedly reduced following exercise, whereas its sensitivity to
dexamethasone was not affected. This suggests that exercise
differentially affects monocytes and T cells. Alternatively, the
CD2/CD28 stimulatory route of IFN-g production itself may be
resistant to exercise-induced changes in dexamethasone sensitivity
[29]. Since we have not used alternative routes for T cell stimulation, we can not discriminate between these alternative
explanations.
Several mechanisms may explain transient changes in GC
sensitivity as observed in the present study. First, GC may
influence the expression of the glucocorticoid receptor (GR) as
demonstrated both in vitro [11] and in vivo [12]. Endogenous GC
may thus contribute to GC sensitivity, since low circulating cortisol levels in neonates are associated with increased dexamethasone sensitivity of PBMC proliferation ex vivo [30]. Conversely,
467
increased levels of endogenous GC may have contributed to the
decreased GC sensitivity following exercise, as observed in our
study. A second mechanism that may be involved in regulation of
GC sensitivity is the modulatory activity of cytokines such as IL-2
and IL-4, that in vitro have been shown to reduce GR-binding
activity and steroid responsiveness of PBMC [31]. Finally, recent
studies suggest that changes in GC sensitivity may be the direct
result of differential expression of a- and b-isoforms of the GR.
Since the b-isoform is able to inhibit hormone-induced, GRamediated stimulation of gene expression, it may function as a
dominant negative inhibitor of GRa activity [32,33]. Whether such
mechanisms underlie the changes in GC sensitivity following
exercise remains to be determined.
We observed very rapid changes in GR-sensitivity of monocytes that are probably not explained by altered GR expression,
since this would require de novo mRNA and protein synthesis.
Rapid changes in GC sensitivity may be the result of an interaction
of the activated GR with transcription factors such as AP-1 and
NF-kB [17,33]. An alternative explanation for the rapid changes is
that exercise results in differential migration of subsets of monocytes from the circulating pool, resulting in depletion of the subset
most sensitive to GC from the circulation. To our knowledge,
subsets of monocytes that differ in their sensitivity to GC have not
been described.
Our results may have clinical implications, since a postexercise decrease in the number of circulating monocytes and
lymphocytes, together with a decreased cytokine release, are
indicative of the existence of temporal immunosuppression. This
is in accordance with observations on decreased immune responses
in overtrained athletes accompanied by an increased susceptibility
to infections [34,35]. Our findings indicate that there are no clear
changes in monocyte-derived IL-10 production and GC sensitivity
of IL-10 release. In contrast, the sensitivity to GC of the proinflammatory cytokines IL-6 and TNF-a is decreased following
exercise. Whether this implies that GC treatment in vivo differentially affects the balance between pro- and anti-inflammatory
cytokines before and after exercise (or other conditions that
result in HPA axis activation) remains to be determined. In view
of the important role of GC in the treatment of inflammatory
disorders, mechanisms leading to a change in GC sensitivity as
described in the present study may contribute to effectiveness of
GC treatment.
ACKNOWLEDGMENTS
The authors thank J. J. Brahim for assistance during the exercise challenges.
This study was supported in part by the Netherlands Asthma Foundation
(grant 93.17).
REFERENCES
1 Cohen JJ, Scheimer HN, Claman HN, Oronsky AL, eds Anti-inflammatory steroid action. Basic and clinical aspects. San Diego: Academic
Press, 1989:111–31.
2 Schwiebert LA, Beck LA, Stellato C, Bickel CA, Bochner BS,
Schleimer RP. Glucocorticosteroid inhibition of cytokine production:
relevance to antiallergic actions. J Allergy Clin Immunol 1996;
97:143–52.
3 Abbinante-Nissen JM, Simpson LG, Leikauf GD. Corticosteroids
increase secretory leukocyte protease inhibitor transcript levels in
airway epithelial cells. Am J Physiol 1995; 12:L601–L606.
4 van Furth AM, Seijmonsbergen EM, Langermans JA, van der Meide
PH, van Furth R. Effect of xanthine derivates and dexamethasone on
Streptococcus pneumoniae-stimulated production of tumor-necrosis
q 1998 Blackwell Science Ltd, Clinical and Experimental Immunology, 111:463–468
468
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
H. H. Smits et al.
factor alpha, interleukin-1 beta (IL-1 beta), and IL-10 by human
leukocytes. Clin Diagn Lab Immunol 1995; 2:689–92.
Cypcar D, Busse WW. Steroid-resistant asthma. J Allergy Clin
Immunol 1993; 92:362–72.
Kirkham B, Corkill MM, Davison SC, Panayi GS. Response to
glucocorticoid treatment in rheumatoid arthritis: in vitro cell-mediated
immune assay predicts in vivo response. J Rheumatol 1991; 18:1130.
Corrigan CJ, Brown PH, Barnes NC et al. Glucocorticoid resistance in
chronic asthma: glucocorticoid pharmacokinetics, glucocorticoid
receptor characteristics, and inhibition of peripheral blood T cell
proliferation by glucocorticoids in vitro. Am Rev Res Dis 1991;
144:101–25.
Spahn JD, Landwehr LP, Nimmagadda S, Susrs W, Leung DYM,
Szefler SJ. Effects of glucocorticoids on lymphocyte activation in
patients with steroid-sensitive and steroid-resistant asthma. J Allergy
Clin Immunol 1996; 98:1073–9.
Nijhuis E, Hinloopen B, Van Duijn C, Hofman A, Rozing J,
Nagelkerken L. Decreased sensitivity to dexamethasone in lymphocytes
from patients with Alzheimer’s disease. Clin Immunol Immunopathol
1994; 73:45–52.
Adcock IM, Lane SJ, Brown CR, Peters MJ, Lee TH, Barnes PJ.
Differences in binding of glucocorticoid receptor to DNA in steroidresistant asthma. J Immunol 1995; 154:3500–5.
Rosewicz S, McDonald AR, Maddux BA, Goldfine ID, Miesfeld RL,
Logsdon CD. Mechanism of glucocorticoid receptor down-regulation
by glucocorticoids. J Biol Chem 1988; 263:2581–4.
Korn SH, Wouters EFM, Wesseling G, Arends J, Thunnissen BJM. In
vitro and in vivo modulation of a- and b-glucocorticoidreceptor mRNA
in human bronchial epithelium. Am J Resp Crit Care Med 1997;
155:1117–22.
DeRijk RH, Petrides J, Deuster P, Gold PW, Sternberg EM. Changes in
corticosteroid sensitivity of peripheral blood lymphocytes after strenuous exercise in humans. J Clin Endo Metabol 1996; 81:228–35.
Sauer J, Polack E, Wikinski S, Holsboer F, Stalla GK, Arzt E. The
glucocorticoid sensitivity of lymphocytes changes according to the
activity of the hypothalamic–pituitary–adrenocortical system. Psychoneuroendocrinol 1995; 20:269–80.
Brown DH, Sheridan JF, Pearl D, Zwillig BS. Regulation of mycobacterial growth by the hypothalamus–pituitary–adrenal axis: differential
responses of Mycobacterium bovis BCC-resistant and -susceptible
mice. Infect Immun 1993; 61:4793–800.
Dunn AJ, Vickers SL. Neurochemical and neuroendocrine responses to
Newcastle disease virus administration in mice. Brain Res 1994;
645:103–12.
Barnes P. Mechanisms of action of glucocorticoids in asthma. Am J
Resp Crit Care Med 1996; 154:S21–S27.
Brattsand R, Linden M. Cytokine modulation by glucocorticoids: mechanisms and actions in cellular studies. Aliment Pharmacol 1996; 10:81–90.
DeRijk R, Michelson D, Karp B et al. Exercise and circadian rhythminduced variations in plasma cortisol differentially regulate interleukin-1b
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
(IL-1b), IL-6 and tumor necrosis factor-a (TNFa) productions in
humans: high sensitivity of TNFa and resistance of IL-6. J Clin Endo
Metabol 1997; 82:2182–91.
Ertel W, Kremer J, Kennedy J et al. Downregulation of proinflammatory cytokine release in whole blood from septic patients. Blood 1995;
85:1341–7.
Kremer J, Jarrar D, Steckholzer U, Ertel W. Interleukin-1, -6 and tumor
necrosis factor-a release is down-regulated in whole blood from septic
patients. Acta Haematol 1996; 95:268–73.
Fischer JR, Schindel M, Stein N et al. Selective suppression of cytokine
secretion in patients with small-cell lung cancer. Anal Oncol 1995;
6:921–6.
Heinrich PC, Castell JV, Andus T. Interleukin-6 and the acute phase
response. Biochem J 1990; 265:621–36.
Dinarello CA. Interleukin-1 and interleukin-1 antagonism. Blood 1991;
77:1627–52.
Van der Pouw-Kraan T, Van Kooten C, Rensink I, Aarden L. Interleukin (IL)-4 production by human T cells: differential regulation of IL4 vs. IL-2 production. Eur J Immunol 1992; 22:1237–41.
Grbic JT, Mannick JA, Gough DB, Rodrick ML. The role of prostaglandin E2 in immune suppression following injury. Ann Surg 1991;
214:253–63.
Gearing AJH, Beckett P, Christodoulou M et al. Processing of tumour
necrosis factor-a precursor by metalloproteinases. Nature 1994;
370:555–7.
McGeehan GM, Becherer JD, Bast RC Jr et al. Regulation of tumour
necrosis factor-a processing by metalloproteinase inhibitor. Nature
1994; 370:558–61.
Nijhuis EWP, Hinloopen B, Odding J, Nagelkerken L. Abrogation of
the suppressive effects of dexamethasone by PKC activation or CD28
triggering. Cell Immunol 1994; 156:438–47.
Kavelaars A, Zijlstra J, Bakker JM et al. Increased dexamethasone
sensitivity of neonatal leukocytes: different mechanisms of glucocorticoid inhibition of T cell proliferation in adult and neonatal cells. Eur J
Immunol 1995; 25:1346–51.
Kam JC, Szefler SJ, Surs W, Sher ER, Leung DYM. Combination IL-2
and IL-4 reduces glucocorticoid receptor-binding affinity and T cell
response to glucocorticoids. J Immunol 1993; 151:346–6.
Bamberger CM, Bamberger A, de Castro M, Chrousos GP. Glucocorticoid receptor b, a potential endogenous inhibitor of glucocorticoid
action in humans. J Clin Invest 1995; 95:2435–41.
Bamberger CM, Schulte HM, Chrousos GP. Molecular determinants of
glucocorticoid receptor function and tissue sensitivity to
glucocorticoids. Endoc Rev 1996; 17:245–61.
Espersen GT, Elbaek A, Ernst E et al. Effect of physical exercise on
cytokines and lymphocyte subpopulations in human peripheral blood.
APMIS 1990; 98:395–400.
Field CJ, Gougeon R, Marliss EB. Circulating mononuclear cell
numbers and function during intense exercise and recovery. J Appl
Physiol 1991; 71:1089–97.
q 1998 Blackwell Science Ltd, Clinical and Experimental Immunology, 111:463–468