Cytokine response to acute running in recreationally

0 downloads 0 Views 431KB Size Report
Jonathan P. R. Scott • Craig Sale • Julie P. Greeves •. Anna Casey • John ..... 0.12 pg4mL-1 and an inter/intra assay CV of\14 % across the range 0.5–32.0 ...... Helge JW, Stallknecht B, Pedersen BK, Galbo H, Kiens B, Richter. EA (2003) The ...
Eur J Appl Physiol DOI 10.1007/s00421-013-2615-x

ORIGINAL ARTICLE

Cytokine response to acute running in recreationally-active and endurance-trained men Jonathan P. R. Scott • Craig Sale • Julie P. Greeves Anna Casey • John Dutton • William D. Fraser



Received: 17 October 2012 / Accepted: 12 February 2013 Ó Springer-Verlag Berlin Heidelberg 2013

Abstract To compare the cytokine response to exhaustive running in recreationally-active (RA) and endurancetrained (ET) men. Eleven RA men (VO2max 55 ± 7 mLmin-1kg-1) and 10 ET men (VO2max 68 ± 7 mLmin-1kg-1) followed a controlled diet and refrained from volitional exercise for 8 days. On the fourth day, participants completed 60 min of treadmill running (65 % VO2max), followed by intermittent running to exhaustion (70 % VO2max). Fasting blood was obtained at baseline, after 20, 40 and 60 min of exercise, at the end of intermittent exercise, during 2 h of recovery and on four follow-up days (FU1–FU4). Tumour necrosis factor-alpha (TNF-a), interleukin-1b (IL-1b), interleukin-6 (IL-6), interleukin-1 receptor antagonist (IL-1ra) and creatine Communicated by William J. Kraemer. J. P. R. Scott (&)  A. Casey Human Sciences, QinetiQ Ltd, Room 2022, Building A5, QinetiQ Ltd, Cody Technology Park, Ively Road, GU14 0LX, Farnborough, Hampshire, UK e-mail: [email protected] C. Sale Biomedical, Life and Health Sciences Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, UK J. P. Greeves Department of Occupational Medicine, HQ Army Recruiting and Training Division, Upavon, UK J. Dutton Department of Musculoskeletal Biology, University of Liverpool, Liverpool, UK W. D. Fraser Norwich Medical School, University of East Anglia, Norwich, UK

kinase (CK) were measured. Exercise increased the concentrations of all cytokines and CK, but there were no significant differences between groups. IL-1b increased (2.2–2.5-fold, P \ 0.001) during exercise, while TNF-a was increased (1.6–2.0-fold, P \ 0.001) during exercise and for 2 h post-exercise. IL-6 (71–84-fold, P \ 0.001) and IL-1ra (52–64-fold, P \ 0.001) were increased throughout exercise and up to FU1, peaking immediately after exercise and at 1.5–2 h post-exercise, respectively. CK concentrations were increased (P \ 0.001) throughout exercise and up to FU4, peaking at FU1, but were not associated with changes in any cytokines. Exhaustive running resulted in modest and transient increases in TNFa and IL-1b, and more marked and prolonged increases in IL-6 and IL-1ra, but improved training status did not affect this response. Increased CK might indicate either exerciseinduced muscle cell disruption or increased cell permeability, although neither appears to have contributed to the increased cytokine concentrations. Keywords Acute exercise  Training status  Tumour necrosis factor-a  Interleukin-1b  Interleukin-6  Interleukin-1 receptor antagonist Abbreviations BMI Body mass index CHO Carbohydrate CK Creatine kinase ET Endurance-trained FD Fixed duration FFM Fat-free mass IEE Intermittent, exhaustive exercise IL-1b Interleukin-1b IL-1ra Interleukin-1 receptor antagonist IL-6 Interleukin-6

123

Eur J Appl Physiol

LMM RA TNF-a VO2max

Linear mixed model Recreationally-active tumour Tumour necrosis factor-alpha Maximum oxygen uptake

Introduction Prolonged running has been shown to result in an increase in the circulating concentrations of cytokines (Ostrowski et al. 1998a, 1999; Starkie et al. 2001). These changes occur in a cascade-like manner, with modest increases in tumour necrosis factor-alpha (TNF-a) and interleukin-1b (IL-1b) and a marked increase in interleukin-6 (IL-6), which peak at the end of exercise, and precede a marked increase in interleukin-1 receptor antagonist (IL-1ra), which peaks 1–2 h later (Ostrowski et al. 1998a, 1999). Increases in TNF-a and IL-1b during running are 1- to 2-fold (Ostrowski et al. 1998a, 1999; Starkie et al. 2001; Bernecker et al. 2011) and the response might be specific to this mode of exercise, since no changes in cytokine concentrations have been reported with dynamic, knee-extensor exercise (Steensberg et al. 2002) or cycling (Starkie et al. 2005). The mechanism for this is not clear, but might include an immune response to local damage in working muscle (Nieman et al. 1998) or the development of systemic endotoxemia (Steensberg et al. 2002). IL-6 concentrations increase between 1- and 100-fold during endurance exercise of different types, with the most marked increases being shown with running (Fischer 2006). The change in IL-6 is not linear over time, increasing exponentially during longer (up to 2.5 h) duration exercise (Ostrowski et al. 1999; Steensberg et al. 2002). IL-6 is released from contracting skeletal muscle during exercise (Steensberg et al. 2002) and is increased by reduced muscle glycogen (Chan et al. 2004) and higher exercise intensity (Helge et al. 2003; Scott et al. 2011). IL-6 released during exercise might have numerous endocrine effects including increased hepatic glucose output, increased lipolysis and the production of C-reactive protein, IL-1ra and IL-10 (Fischer, 2006). Increased IL-6 may still be evident during the days (1–5) following prolonged strenuous exercise (Starkie et al. 2001; Neubauer et al. 2008), although this might also reflect the activity of cytokine-releasing macrophages involved in muscle repair (Fielding et al. 1993). The cytokine response to acute exercise can also be altered with exercise training (Gokhale et al. 2007; Croft et al. 2009). Fischer et al. (2004) report no difference in the IL-6 response to acute, knee-extensor exercise following 10 wks of training, despite an increase the absolute exercise intensity, suggesting an attenuation of the acute IL-6 response. In line with this, Croft et al. (2009) reported an

123

attenuation of the IL-6 response to a bout of shuttle running following 6 weeks of high-intensity interval running, using the same absolute workload for both tests. Only one study has measured the cytokine response to endurance running in participants with different training status (Gokhale et al. 2007). This study reports increases in mean IL-6 concentrations and decreases in TNF-a, with less marked changes in athletes. However, IL-6 increased by only 7 % in nonathletes and concentrations decreased in about one-third of all participants, which is in contrast to the majority of previous studies, as is the reported decrease in TNF-a. Relatively short and intermittent exercise, low and poorly defined exercise intensities and premature exhaustion in 30 % of non-athletes might account for some of this variability. The aim of the present study was to examine changes in pro- and anti-inflammatory cytokines in response to an acute bout of running in recreationally-active and endurance-trained young men. In order to promote the typical cascade-like increase in circulating cytokines typical of acute endurance running, we chose an exercise protocol that was both longer, and undertaken at a higher exercise intensity, than used previously (Gokhale et al. 2007).

Methods Participants Eleven recreationally-active (RA) and 10 endurancetrained (ET) males were recruited to the study, which was approved by the QinetiQ Research Ethics Committee (Table 1). Participants provided written informed consent. Participants were included if they were non-smokers, were free from musculoskeletal injury and were not taking any medication or suffering from any condition known to affect inflammation. Compliance with these inclusion criteria was confirmed from a medical screening questionnaire and a medical examination. Participants were considered RA if they completed two to three exercise sessions per week including at least one run, but performed not more than 2 h of exercise per week and were not involved in an organised training program (Table 1). Participants in ET were required to have been running for a minimum of 2 years without a significant break and to have completed a measured 10 km distance in less than 40 min within the previous 4 week. Experimental design Participants completed two preliminary visits for medical screening, habituation with trial procedures and measurement of maximal rate of oxygen uptake (VO2max). Participants then

Eur J Appl Physiol

completed one 8 day experimental condition during which they refrained from non-prescribed physical activity and ate a personalised controlled diet (Fig. 1). Table 1 Participant characteristics of the recreationally-active (RA) and endurance-trained (ET) groups

n

RA

ET

11

10

Reference range

Age (years)

30 ± 3

31 ± 3

Height (m) Body mass (kg)

1.84 ± 0.06 84.4 ± 8.4

1.73 ± 3*** 69.8 ± 6.6***

BMI (kgm-2)

25.0 ± 2.9

23.3 ± 1.6

VO2max (mLmin-1kg-1)

55.5 ± 6.8

67.9 ± 6.1***

Physical activity (hweek-1)

1.4 ± 0.4

4.5 ± 1.3***

Haemoglobin (gL-1)

149 ± 5

145 ± 9

White blood cells (9109L-1)

6.19 ± 2.44

6.31 ± 1.51

On Days 1–3, participants refrained from all physical activity and followed their individually prescribed diet. On Day 4, participants attended the laboratory and performed an acute, exhaustive bout of treadmill running. On Days 5–8, participants returned to the laboratory for follow-up analysis and continued to refrain from all physical activity and followed their prescribed diet. Participants were asked to report any symptoms of illness or fever in the days leading up to and during the study.

Experimental procedures Pre-trial measurements

Platelets (9109L-1)

242 ± 52

261 ± 62

Sodium (mmolL-1)

140 ± 2

141 ± 2

Potassium (mmolL-1)

4.4 ± 0.3

4.4 ± 0.4

3.5–5.1

Creatinine (umolL-1)

101 ± 6

91 ± 5

60–120

Total bilirubin (umolL-1)

13 ± 4

16 ± 10

\25

Conjugated bilirubin (umolL-1)

4±2

6±2

Calcium (mmolL-1)

2.35 ± 0.09

2.39 ± 0.12

2.10–2.55

Albumin (gL-1)

47 ± 2

46 ± 3

34–48

Corrected calcium (mmolL-1)

2.21 ± 0.07

2.26 ± 0.10

2.10–2.55

Participants underwent a full medical examination, including the completion of a medical screening questionnaire, a 12-lead electrocardiogram and urine analyses. Venous blood (10 mL) was collected for routine clinical chemistry (immune and kidney function, electrolytes, calcium and phosphate) and full blood count. Participants completed a 3-day weighed food diary consisting of two weekdays and one weekend day, for the subsequent estimation of habitual daily energy intake (kcal) and macronutrient composition. Participants were issued with a set of calibrated weighing scales to measure food intake, and received both verbal and written instructions. Food diaries were analysed using nutritional analysis software (Microdiet V2, Downlee Systems Limited, UK). To establish the association between oxygen uptake and running velocity during level running, participants completed a 20-min submaximal run on a treadmill (XELG 70 ERGO, Woodway, USA), consisting of four, 5 min stages. After a 30-min rest, participants completed a discontinuous, incremental exercise test to exhaustion to establish

135–145

Values are mean ± 1 SD BMI body mass index, VO2max maximum oxygen uptake *** Different (P \ 0.001) from RA

Controlled diet and restricted training

P1

P2

D2

D1

Preliminary testing

D4

FD20

FD40

D6

D5

D7

D8

Recovery and follow-up analyses

Lead-in period

FD BASE

D3

IEE FD60

EE

R0.5

R1.0

R1.5

R2.0

Blood Samples Meals

Fig. 1 Outline of overall study design (top), and the experimental protocol on Day 4 (D4) (bottom). P1–P2, preliminary days; D1–D8, experimental days. Shaded boxes denote laboratory visits; adjoined boxes denote consecutive days. Hatched box denote exercise: FD,

fixed duration (60 min) exercise at 65 % VO2max; IEE, intermittent, exhaustive exercise at 70 % VO2max). Vertical arrows indicate blood samples and meals: EE, end of exhaustive exercise; R0.5–R2.0, 0.5–2 h of recovery

123

Eur J Appl Physiol

VO2max using a modified Taylor protocol (Froelicher et al. 1974). The results of the two tests were used to estimate the treadmill velocity corresponding to 65 and 70 % VO2max during level running based on the regression line of VO2 and treadmill velocity. A diet consisting of 8 g carbohydrate (CHO)kg FFM-1day-1 and isocaloric with habitual diet was designed for each participant based on individual dietary habits. Participants were provided with three menus that were given in a three-day cyclic order (Menu A on Days 1, 4 and 7, Menu B on Days 2 and 5, and Menu C on Days 3 and 6. Participants attended the laboratory on the morning of Day 8 in a fasted state to provide their final blood sample and were free to resume their normal diet thereafter). During the experimental period, participants provided their own food but were given both verbal and written instructions concerning the quantity and preparation of their meals and timings. Adherence to the food preparation and consumption, instructions was verbally confirmed with each participant throughout the experimental period. Trial procedures After fasting from 2100 h, at 0730 h on Day 4, RA and ET participants voided and had their body mass measured. Participants adopted a semi-recumbent position on a bed, and placed their left hand into a hand-warming unit (Medical Physics, Nottingham, UK) for 20 min. A cannula (18GA 1.2 9 45 mm, Becton–Dickinson, USA) was inserted into a vein in the back of the hand and connected to a three-way tap (Connecta, Becton–Dickinson, USA). The cannula was kept patent with isotonic saline (0.9 % NaCl). The hand was returned to the warming unit for a further 20 min before a baseline (BASE) blood sample was collected for the determination of cytokines, lactate, glucose and CK concentrations. During the experiment, the temperature inside the warming unit was held constant at 55–60 °C. Following a 5-min warm-up at 50 % VO2max and a further 5 min for volitional stretching, participants completed a 60-min run at 65 % of VO2max (fixed duration exercise; FD). Following a 15-min seated rest, participants resumed exercise at 70 % VO2max and then ran to volitional exhaustion, at which point they rested for 5 min before resuming exercise. This work-rest pattern was repeated until participants were not able to complete 5 min of continuous running. Rest breaks were subsequently reduced to 1 min and the work-rest pattern was repeated until participants were not able to complete 2 min of continuous running, at which point exercise was terminated. This work-rest protocol will be referred to as intermittent, exhaustive exercise. Blood samples were collected after 20 (FD20), 40 (FD40) and 60 (FD60) min of

123

fixed duration exercise and again at the end of intermittent exhaustive exercise (EE) (Fig. 1). On completion of intermittent, exhaustive exercise, participants adopted a semi-recumbent position and returned their hand to the warming unit where it remained while they rested for 2 h. Further blood samples were drawn at 0.5 h (R0.5), 1 h (R1.0), 1.5 h (R1.5) and 2 h (R2.0) of recovery (Fig. 1). Participants remained fasted until the final blood sample was drawn after two h of recovery. On Days 5–8, participants returned to the laboratory at 0800 h following an overnight fast. A blood sample was drawn by venepuncture from a vein in the anticubital fossa (FU1–FU4). Sample collection For the measurement of TNF-a, IL-1b, IL-6 and IL-1ra, blood was transferred into pre-cooled tubes containing 15 %, 0.12 mL of K3 EDTA (Becton–Dickinson Vacutainer System, USA). Tubes were gently inverted 8–10 times and centrifuged immediately at 2,000 rpm and 5 °C for 10 min. Plasma was separated and stored at -70 °C until analysis. For measurement of CK, blood was transferred into pre-cooled standard tubes (Becton Dickinson Vacutainer System, USA) and left to clot at room temperature for 60 min. Tubes were centrifuged immediately at 2,000 rpm and 5 °C for 10 min and serum was separated and stored at -70 °C. For the measurement of glucose and lactate, whole blood was transferred into pre-cooled tubes containing Fluoride-Oxalate. Tubes were gently inverted 8–10 times and analysed in duplicate immediately (Yellow Springs Instruments, 2300 Stat Plus, YSI Ltd, UK). TNF-a, IL-1b, IL-6 and IL-1ra were measured using commercial, solid phase, enzyme-linked, immunosorbent assay (ELISA) (Quantikine HS, R&D Systems Ltd, Abingdon, UK). The TNF-a assay has a detection limit of 0.12 pgmL-1 and an inter/intra assay CV of\14 % across the range 0.5–32.0 ngL-1. The IL-1b assay has a detection limit of 0.1 ngL-1 and an inter/intra assay CV of \12 % across the range 0.5–8.0 ngL-1. The IL-6 assay has a detection limit of 0.039 pgmL-1 and an inter/intra assay CV of \10 % across the range 0.15–10 ngL-1. The IL-1ra assay has a detection limit of 2 ngL-1 and an inter/intra assay CV of \8 % across the range 50–3,000 ngL-1. CK was measured using standard reagents and methodology on a P module (Roche, Lewes UK). The inter/intra assay CV across the measuring range (10–1,000 UL-1) is\5 %. Statistical analysis All data are presented as mean ± 1SD unless otherwise stated and statistical significance was accepted at an alpha

Eur J Appl Physiol

level of P \ 0.05. Student’s t-tests for unpaired data were used to compare participant characteristics and variables relating to exercise performance between the RA and ET groups. Cytokines, glucose, lactate and CK were analysed using a linear mixed model analysis of variance, with the factors Time (of sampling) and Group (RA vs ET) included and with individuals as a random, within-group factor. With the exception of TNF-a, analyses of all variables revealed that the data did not meet the assumptions of the linear mixed model (normality and constant variance). Normality and homogeneity were achieved following log transformations for all variables with the exception of IL-1ra and CK, which required a log-sqrt transformation. Where there was a significant effect of Time but no significant Group 9 Time interaction, each subsequent time point was compared against baseline using a pooled mean from the two groups, using Dunnett’s test with baseline as the control. When the Group 9 Time interaction was significant, time points were compared to baseline within each group using Dunnett’s test with baseline as the ‘‘control’’ and the groups were compared to each other at all time points using the Student Newman–Keuls (SNK) test. Pearson’s correlation coefficient was calculated to examine the relationship between exercise duration and peak concentrations of cytokines and CK, between peak concentrations of IL-6 and IL-1ra and between CK concentrations and IL-6, IL-1ra concentrations on the four follow-up days. All statistical analyses was performed with the SPSS v17 (SPSS Inc., Chicago, IL) with the exception of the Dunnett’s and SNK tests, which were performed with Statistica (StatSoft Inc., Tulsa, OK).

Results Participant characteristics Participant characteristics in the RA and ET groups are shown in Table 1. Compared with RA, ET subjects were of shorter stature (1.73 ± 3 vs 1.84 m, P \ 0.001), had a lower body mass (69.8 ± 6.6 vs 84.4 ± 8.4 kg, P \ 0.001) and a larger VO2max (67.9 ± 6.1 vs 55.5 ± 6.8 mLmin-1kg-1, P \ 0.001) and performed more physical activity (4.5 ± 1.3 vs 1.4 ± 0.4 hwk-1, P \ 0.001). Dietary analysis Compared with RA participants, participants in the ET group reported consuming a greater quantity of CHO (P \ 0.05), CHO relative to fat-free mass (P \ 0.05) and CHO as a percentage of total energy intake (P \ 0.05) (Table 2). There were no other differences between the two

Table 2 Habitual energy intake and dietary macronutrient composition of the recreationally-active (RA) and endurance-trained (ET) groups RA

ET

Energy (MJ)

11.0 ± 1.5

11.6 ± 2.6

CHO (g)

324 ± 59

383 ± 65*

CHO (gkg-1 FFM) CHO (% of total energy)

5.4 ± 0.9 47.3 ± 6.1

6.7 ± 1.3* 53.2 ± 6.5*

Fat (% of total energy)

30.9 ± 4.5

27.2 ± 7.0

Protein (% of total energy)

18.1 ± 2.3

17.8 ± 3.5

Values are mean ± 1 SD CHO carbohydrate, FFM fat-free mass * Different (P \ 0.05) from RA

groups. The energy content of the experimental diet was not significantly different from habitual energy intake in either group (RA, P = 0.107; ET, P = 0.368), and there were no significant differences between the two groups for any experimental diet variable (P values from 0.053 to 0.667, data not shown). Exercise performance variables and body mass Final exercise intensities were 62.4 ± 3.7 and 67.1 ± 3.5 % VO2max in ET and 60.9 ± 2.8 and 68.8 ± 4.4 % VO2max in RA during FD and intermittent, exhaustive exercise, with no significant differences between the two groups (FD, P = 0.185; Intermittent, exhaustive exercise, P = 0.313). Total exercise time was 116 ± 27 min and 134 ± 14 min for RA and ET, respectively. During intermittent, exhaustive exercise, the ET group ran for 73 ± 14 min compared to 56 ± 27 min in RA (P = 0.079) and ET participants covered a significantly greater distance than RA participants (16.5 ± 3.5 vs 9.8 ± 3.8 km; P \ 0.001). Total distance covered for the whole exercise trial was also significantly greater for ET (29.1 ± 3.8 vs 19.8 ± 3.5 km; P \ 0.001). Body mass data showed a significant main effect of Time (P \ 0.001) but no significant Group 9 Time interaction (P = 0.754) with pooled, mean body mass being significantly lower than BASE at FU1 (-1.1 %, P \ 0.01), FU2 (-0.6 %, P \ 0.01), FU3 (-0.5 %, P \ 0.05) and FU4 (-1.0 %, P \ 0.001) (data not shown). Glucose There was a significant main effect of Time (P \ 0.001) but no significant Group 9 Time interaction (P = 0.754) for glucose concentrations. Pooled, mean concentrations were significantly higher (P \ 0.001) than BASE throughout FD and lower (P \ 0.001) than BASE at the end of exhaustive exercise (RA, 3.62 ± 0.44 mmolL-1; ET, 3.58 ± 0.89 mmolL-1), remaining so until 2 h post-exercise (Fig. 2a).

123

Eur J Appl Physiol

Glucose (mmol·L-1)

7

a

6 5 4 3 2 3

Lactate (mmol ·L-1)

Fig. 2 Glucose (a) and lactate (b) concentrations before (BASE), during (FD20–FD60) and for 2 h after (EE–R2.0) exhaustive running in the RA (open squares) and ET (filled diamonds) groups. Values are mean ± 1SD. LMM revealed the following: a significant main effect of Time (P \ 0.001) but no significant Group 9 Time interaction (P = 0.754) for glucose; a significant main effect of Time (P \ 0.001) and a significant Group 9 Time interaction (P \ 0.01) for lactate. *ET different (P \ 0.05) from RA

*

*

FD20

FD40

b

2

1

0 BASE

FD60

EE

R0.5

R1.0

R1.5

R2.0

Sampling Schedule

Lactate There was a significant main effect of Time (P \ 0.001) and a significant Group 9 Time interaction (P \ 0.01) for lactate concentrations. In RA, concentrations were significantly higher (P \ 0.001) than BASE throughout FD, at the end of intermittent, exhaustive exercise and at 30 min post-exercise (Fig. 2b). Lactate concentrations in ET were significantly higher than BASE at FD20 (P \ 0.01), FD40 (P \ 0.05) and FD60 (P \ 0.01), at the end of intermittent, exhaustive exercise (P \ 0.001) and at 30 min post-exercise (P \ 0.001). The increase in lactate in ET was less marked than in RA resulting in significantly higher lactate concentrations in RA at FD20 (P \ 0.05) and FD40 (P \ 0.05) although not at FD60 (P = 0.06) or thereafter. TNF-a There was no significant difference in baseline TNF-a concentrations between the RA and ET groups (1.39 ± 0.76 vs 1.46 ± 1.12 pgmL-1, P = 0.866). There was a significant main effect of Time (P \ 0.001) but no significant Group 9 Time interaction (P = 0.168) for TNF-a. Pooled, mean TNF-a concentrations were significantly higher (P \ 0.01) than BASE at FD20 and throughout FD (Fig. 3a). Concentrations remained significantly higher (P \ 0.001) than BASE at EE and in the first 2 h of recovery, with peak concentrations occurring at EE (RA, 2.09 ± 0.85 pgmL-1; ET, 2.44 ± 1.50 pgmL-1), increased 1.6- and 2.0-fold from BASE. TNF-a

123

concentrations were not significantly different from BASE on any of the follow-up days (FU1–FU4). IL-1b There was no significant difference in baseline IL-1b concentrations between the RA and ET groups (0.52 ± 0.35 vs 0.50 ± 0.18 pgmL-1, P = 0.897). There was a significant main effect of Time (P \ 0.001) but no significant Group 9 Time interaction (P = 0.118) for IL-1b. Pooled, mean IL-1b concentrations were significantly higher than BASE throughout FD (P \ 0.01) and at EE (P \ 0.001), where peak concentrations occurred (RA, 1.05 ± 0.587 pgmL-1; ET, 1.20 ± 1.19 pgmL-1), increased 2.2- and 2.5-fold (Fig. 3b). Concentrations were not significantly different from BASE at 30 min postexercise or thereafter. Il-6 There was no significant difference in baseline IL-6 concentrations between the RA and ET groups (0.50 ± 0.39 vs 0.58 ± 0.47 ngL-1, P = 0.665). There was a significant main effect of Time (P \ 0.001) but no significant Group 9 Time interaction (P = 0.708) for IL-6. Pooled, mean IL-6 concentrations were significantly higher (P \ 0.001) than BASE throughout FD, with concentrations increased 14.9- and 9.5-fold at FD60 in RA and ET, respectively (Fig. 3c). Peak IL-6 concentrations occurred at the end of intermittent, exhaustive exercise in both groups

a

4 3 2 1 0

IL-1β (pg·ml-1)

3

b

2 1 0 60

IL-6 (pg·ml-1)

Fig. 3 TNF-a (a), IL-1b (b), IL-6 (c) and IL-1ra (d) concentrations before (BASE), during (FD20–FD60) and for 2 h after exhaustive running (EE–R2.0), and on four follow-up days (FU1–FU4) in the RA (open squares) and ET (filled diamonds) groups. Values are mean ± 1SD. LMM revealed the following significant main effects of Time for TNF-a, IL-1b, IL-6 and IL-1ra (all P \ 0.001), but no significant Group 9 Time interaction for any variable (TNF-a, P = 0.168; IL-1b, P = 0.118; IL-6, P = 0.708; IL-1ra, P = 0.468)

TNF-α (pg·ml-1)

Eur J Appl Physiol

c

40 20

IL-1ra (x 103 pg·ml-1)

0

d

30 20 10 0

Sampling Schedule

(RA, 29.9 ± 13.2 pgmL-1; ET, 31.9 ± 21.5 pgmL-1), increased 84- and 71-fold from BASE and remained 30-fold higher (P \ 0.001) than BASE in both groups during the first 2 h post-exercise. IL-6 concentrations remained higher (P \ 0.01) than BASE at FU1 (increased 2.1- and 1.5-fold in RA and ET), but not thereafter.

pgmL-1) and R1.5 in ET (12,861 ± 15,492 pgmL-1), increased 52-fold and 64-fold from BASE. IL-1ra concentrations remained significantly higher (P \ 0.01) than BASE at FU1 (increased 1.7- and 1.8-fold in RA and ET) but not thereafter. Creatine kinase

IL-1ra There was no significant difference in baseline IL-1ra concentrations between the RA and ET groups (239 ± 88 vs 189 ± 54 pgmL-1, P = 0.142). There was a significant main effect of Time (P \ 0.001) but no significant Group 9 Time interaction (P = 0.468) for IL-1ra. Pooled, mean concentrations were not significantly different from baseline at FD20 or FD40 but were significantly higher (P \ 0.05) than BASE at FD60, increased 1.6- and 1.8-fold (Fig. 3d). Concentrations were increased further by EE and during the first 2 h post-exercise, with peak IL-1ra concentrations occurring at R2.0 in RA (12,002 ± 10,807

There was no significant difference in baseline CK concentrations between the RA and ET groups (119 ± 66 vs 135 ± 114 pgmL-1, P = 0.696). There was a significant main effect of Time (P \ 0.001) but no significant Group 9 Time interaction (P = 0.966) for CK. Pooled, mean CK concentrations were not significantly different from BASE at FD20 (P = 0.10) but were significantly higher than BASE at FD40 and FD60 (P \ 0.01), at the end of exhaustive exercise and up to 2 h post-exercise (P \ 0.001) (Fig. 4). Peak CK concentrations occurred at FU1 (RA, 855 ± 798 UL-1; ET, 810 ± 587 UL-1), increased 7.0- and 7.1-fold from BASE and remained

123

Eur J Appl Physiol

CK (U·L-1)

1600 1200 800 400 0

Sampling Schedule Fig. 4 CK concentrations before (BASE), during (FD20–FD60) and for 2 h after exhaustive running (EE–R2.0), and on four follow-up days (FU1–FU4) in the RA (open squares) and ET (filled diamonds)

groups. Values are mean ± 1SD. LMM revealed the following: significant main effects of Time (P \ 0.001), but no significant Group 9 Time interaction (P = 0.966)

Table 3 Pearson correlation coefficients on the day of the exercise test (Day 4)

Discussion

TNFapeak Total exercise duration CKpeak

IL1bpeak

IL-6peak

IL1rapeak

0.230

0.069

0.262

0.108

-0.038

-0.234

-0.071

-0.272

IL-1ra @ R1.5

0.489*

IL-1ra @ R2.0

0.508*

CKpeak 0.537*

* P \ 0.05

Table 4 Pearson correlation coefficients on the four follow-up days (FU1–FU4)

IL-6 IL-1ra

CK(FU1)

CK(FU2)

CK(FU3)

CK(FU4)

-0.163

-0.185

-0.208

-0.180

0.083

0.077

-0.064

0.025

higher (P \ 0.001) than BASE on the remaining follow-up days. Pearson correlations Total exercise duration was significantly correlated with peak CK concentrations (r = 0.537, P \ 0.05) but not with peak concentrations of any of the cytokines (r values from -0.108 to 0.262) (Table 3). Peak CK concentrations did not correlate with peak concentrations of any of the cytokines (r values from -0.380 to -0.071). Peak IL-6 concentrations (at EE) were correlated with IL-1ra concentrations at R1.5 (r = 0.489, P \ 0.05) and R2.0 (r = 0.508, P \ 0.05). There were no significant correlations between CK, and IL-6 and IL-1ra concentrations on any of the four follow-up days (r values from -0.208 to 0.083) (Table 4).

123

The main findings from this study are (1) acute, exhaustive treadmill running results in a cascade-link increase circulating cytokines, with modest increases in TNF-a and IL-1b and a marked increase in IL-6, preceding a marked increase in IL-1ra, and; (2) when performed at the same relative exercise intensity, improved training status did not affect the cytokine response to exhaustive running. This is the first study to compare the cytokine response to acute exercise at the same relative exercise intensity in two groups of different training status, and reports no effect of training status on basal cytokine concentrations or on changes in cytokines with exhaustive running. In contrast to our findings, exercise training is reported to reduce basal cytokine levels (Nicklas et al. 2008). The effect of exercise training maybe to restore elevated cytokines to their normal levels, whereas in our study, basal concentrations were already low in both groups, which might explain the lack of any effect of improved training status. The IL-6 response to acute, endurance exercise is increased with exercise intensity (Helge et al. 2003; Scott et al. 2011) and lower muscle glycogen content (Chan et al. 2004), and attenuated by carbohydrate ingestion (Starkie et al. 2001; Nieman et al. 2003). During exercise, IL-6 might serve to mobilise hepatic glucose (Febbraio et al. 2004) and increase lipolysis and fat oxidation in skeletal muscle (Wolsk et al. 2010), which might explain the apparent acceleration in the increase in IL-6 towards the later part of exercise in our study and in the work of others (Ostrowski et al. 1998a; Steensberg et al. 2002). Consistent with this idea, Fischer (2006) has calculated that 51 % of the variation in the increase in IL-6 can be explained by exercise duration, while IL-6 concentrations are higher following a marathon compared with a half marathon (Reihmane et al. 2012). However, Fischer’s calculations were based on continuous exercise protocols, whereas our exercise was intermittent, while Reihmane et al. (2012)

Eur J Appl Physiol

were not able to control relative exercise intensity. Like Neubauer et al. (2008), we showed no significant correlation between peak IL-6 concentrations and total exercise duration. Adaptations to exercise training include increased skeletal muscle glycogen content and enhanced fat oxidisation, which result in a greater capacity for mechanical work and less dependence on plasma glucose (Holloszy and Booth 1976; Saltin and Rowell 1980; Hawley 2002). Consistent with a role in mediating glucose homeostasis, exercise training appears to attenuate the IL-6 response to acute exercise. This includes similar changes in IL-6 before and after 10 weeks of dynamic knee-extensor exercise training, despite a higher absolute workload (Fischer et al. 2004) and, using the same absolute workload, an attenuated IL-6 response following 6 weeks of high-intensity interval running (Croft et al. 2009). In the study of Croft et al. (2009), the attenuation occurred despite a greater exercise duration in the post-training test, although using the same absolute workload for both pre- and post-exercise tests, the lower relative exercise intensity with the post-exercise test might also have contributed to the attenuation of IL-6 (Helge et al. 2003; Scott et al. 2011). Using the same relative exercise intensity, we did not observe an attenuation of the IL-6 response to exhaustive exercise with endurance training. After 60 min of exercise, however, whilst IL-6 concentrations were similar between the two groups, the fold increase in concentrations from pre-exercise was more than 50 % greater in RA compared with ET (14.9-fold vs 9.5-fold). If endurance training does result in adaptations that reduce dependence on plasma glucose and muscle glycogen as substrate (Holloszy and Booth 1976; Saltin and Rowell 1980; Hawley 2002), an attenuation (or delayed onset) of the increase in IL-6 during exercise might be expected. We did not measure fat oxidation, but total fat oxidation might have been higher in ET compared with RA, as a result of increased oxidation of intramuscular triglycerides (Phillips et al. 1996). Lactate levels were approximately 40 % lower in ET compared with RA during the first 60 min of exercise, which might reflect the increased capacity to oxidise lactate with training (Donovan and Brooks 1983; Bergman et al. 1999). This, in turn, could have decreased glucose oxidation by exercising muscle resulting in a decreased demand for blood glucose and glucose production to maintain blood glucose homeostasis (Miller et al. 2002). Fischer et al. (2004) observed that the attenuation of the IL-6 response following training was accompanied by an attenuated decrease in plasma glucose levels. We, however, observed no difference in glucose concentrations between RA and ET up to 60 min of exercise although, in the study of Fischer et al., the attenuation of the blood glucose response was only evident after 5 h of exercise, with no difference

after 3 h (Fischer et al. 2004). Taken together, these data suggest that the attenuated increase in IL-6 in ET during the first 60 min of exercise might be accounted for by differences in fat and lactate oxidation between ET and RA as a result of ET’s training history reducing the demand for glucose production. It is also possible, therefore, that we examined a longer period of continuous exercise, a more marked effect of training status on blood glucose and IL-6 might have been observed. In contrast to a more prolonged period of continuous exercise, however, all participants subsequently ran to volitional exhaustion and, likely as a result of a similar relative exercise intensity, the ET group ran for approximately *50 % longer than the RA group. Although we did not observe an association between peak IL-6 concentrations and total exercise duration, the longer duration of exercise may have masked any effect of training status. Equally, however, as exercise duration is an important determinant of the IL-6 response (Fischer 2006; Reihmane et al. 2012), a greater increase in IL-6 might have been expected in the ET group. It is possible, therefore, that the similar post-exercise IL-6 concentrations in the two groups despite a greater exercise duration in ET might, itself, reflect an attenuation of the IL-6 response as a result of improved training status. The time course of IL-1ra response to exercise in the present study is in line with previous findings (Ostrowski et al. 1998a, b, 1999). The only known biological role of IL-1ra is to bind to the IL-1 receptor, inhibiting the functions of IL-1a and IL-1b (Dinarello, 1998). However, increasing plasma concentrations of IL-6 are reported to increase IL-1ra concentrations to levels similar to those seen with strenuous exercise (Steensberg et al. 2003). In our study, the initial increase in IL-6 preceded that IL-1ra and peak IL-1ra concentrations occurred 1–2 h after those of IL-6. We also observed a significant correlation between peak IL-6 and peak IL-1ra concentrations, and together, our findings support the idea that the IL-6 response to exercise mediates the subsequent increase in IL-1ra. In this instance, the similar IL-1ra responses in the two groups in the present study are consistent with the lack of an effect of training status on the preceding increase in IL-6. In contrast to IL-6, the source of the increased TNF-a and IL-1b with exercise remains uncertain. Unlike with inflammation, monocyte activity does not increase with exercise (Starkie et al. 2001; Bernecker et al. 2011). Circulating lipopolysaccharides (LPS) are known stimulators of TNF-a and may increase during exercise (Jeukendrup et al. 2000). In animals at least, Kupffer cells are a major source of circulating TNF-a following LPS-challenge (Jiang et al. 1999) making the liver a possible source of TNF-a during exercise. However, increased LPS levels are reported during exercise without concomitant increases in

123

Eur J Appl Physiol

TNF-a (Jeukendrup et al. 2000), whilst even reports of increased LPS concentrations during exercise are contested (Davis et al. 2007). Damage to muscle during exercise may result in neutrophil infiltration and pro-inflammatory cytokine accumulation (Fielding et al. 1993). This might explain the accumulation of TNF-a in skeletal muscle following running (Nieman et al. 2003), although there is conflicting evidence concerning changes in IL-1b mRNA in muscle (Nieman et al. 2003; Ostrowski et al. 1998b). TNF-a does not accumulate in (Steensberg et al. 2002; Chan et al. 2004), nor is it released from, contracting muscle during either cycling (Febbraio et al. 2003) or knee-extensor exercise (Steensberg et al. 2002), neither of which result in muscle cell disruption (Starkie et al. 2005; Steensberg et al. 2000), although this has never been evaluated with running. In our study, the elevation in TNF-a and IL-1b concentrations coincided with the initial, small increase in CK, suggesting a possible association between muscle cell disruption and changes in pro-inflammatory cytokines. That said, repeated contraction reduces muscle fibre membrane resistance and increases permeability without damage to the fibre itself (Fink and Lu¨ttgau 1976; Fink et al. 1983), which might, in part, account for the increase in CK and possibly its timing (after 40 min of exercise). The pronounced increase in CK on the first follow-up day (FU1) that decreased thereafter is consistent with muscle damage (Serraa˜o et al. 2003), although elsewhere, a similar time course of changes in serum CK has been reported following a 21 km run—comparable to the total distance covered in the present study—despite no evidence of damage (Goodman et al. 1997). If the increase in CK was a result of changes in membrane permeability rather than muscle damage, this might explain why we did not observe any correlations between peak concentrations of CK (on FU1), and peak TNF-a and IL-1b concentrations (at FD60). Alternatively, any association might have been confounded by the delay in the elimination of CK from the extracellular compartment due to the ratio of the enzyme’s life-span to biological half-life. If pro-inflammatory cytokines are released from skeletal muscle during running as a result of muscle damage or increased membrane permeability, higher intensity and or longer duration exercise might have been expected to display more marked increases in TNF-a and IL-1b concentrations as a result of a greater muscle damage. One study reports increased TNF-a at 70 % VO2max but not 50 % VO2max with 30 min of cycling (Kimura et al. 2001), with the authors attributing this effect to greater damage, although no measures of muscle cell damage were taken, while in a more recent study, both TNF-a and CK concentrations were significantly higher following a marathon

123

compared with a half marathon (Reihmane et al. 2012). In our study, total exercise duration in the ET group was approximately 50 % greater than in the RA group and we observed an association between peak CK concentrations and total exercise duration across both groups. However, there were no differences in peak CK or TNF-a and IL-1b concentrations between the ET and RA groups. In animals at least, both the lung (Rosa Neto et al. 2009) and adipose tissue (Colbert et al. 2001) also increase their expression of TNF-a during acute exercise so can not be excluded as possible contributory sources to increased concentrations in our study. Finally, the activity of cytokine-releasing macrophages involved in muscle repair (Fielding et al. 1993) might explain the prolonged increases in cytokines following exercise. Consistent with our findings, increased IL-6 concentrations have been reported 24 h or more after longduration, endurance exercise (Starkie et al. 2001). Significant correlations have also been reported between pre- and post-race changes in TNF-a, IL-1b and IL-6 and markers of muscle damage (Starkie et al. 2001; Neubauer et al. 2008; Ostapiuk-Karolczuk et al. 2012). In the present study, however, TNF-a and IL-1b concentrations were not significantly elevated from baseline on the four follow-up days, and we observed no associations between IL-6 or IL1ra and CK. This might indicate that the elevated IL-6 and IL-1ra concentrations at 24 h post-exercise simply reflect the marked changes that occurred during exercise the previous day.

Conclusions In conclusion, acute, exhaustive running at 65–70 % VO2max was associated with a cascade-like increase in cytokines, with modest and transient increases in TNF-a and IL-1b and more marked and prolonged increases in IL-6 and IL-1ra. Changes in cytokine concentrations were, however, not significantly different between recreationallyactive and endurance-trained individuals, suggesting that this response is not affected by training status. Exercise also increased circulating CK concentrations, but there was no association with changes in any cytokines, possibly indicating that changes in CK resulted from increased muscle membrane permeability rather than cell damage. Acknowledgments The authors would like to acknowledge Mrs Anne Wright and Mr Nicholas Moon for their assistance with the collection of data, and all the subjects that participated in the study, without whose considerable effort, the study would not have been possible. The study was funded by the Human Capability Domain of the UK Ministry of Defence Scientific Research Programme. JPR Scott receives an Industrial Fellowship from the Royal Commission for the Exhibition of 1851.

Eur J Appl Physiol Conflict of interest

The authors declare no conflicts of interest.

References Bergman BC, Wolfel EE, Butterfield GE, Lopaschuk GD, Casazza GA, Horning MA, Brooks GA (1999) Active muscle and whole body lactate kinetics after endurance training in men. J Appl Physiol 87:1684–1696 Bernecker C, Scherr J, Schinner S, Braun S, Scherbaum WA, Halle M (2011) Evidence for an exercise induced increase of TNF-alpha and IL-6 in marathon runners. Scand J Med Sci Sports. doi: 10.1111/j.1600-0838.2011.01372.x Chan MH, Carey AL, Watt MJ, Febbraio MA (2004) Cytokine gene expression in human skeletal muscle during concentric contraction: evidence that IL-8, like IL-6, is influenced by glycogen availability. Am J Physiol Regul Integr Comp Physiol 287:R322–R327 Colbert LH, Davis JM, Essig DA, Ghaffar A, Mayer EP (2001) Tissue expression and plasma concentrations of TNFalpha, IL-1beta, and IL-6 following treadmill exercise in mice. Int J Sports Med 22:261–267 Croft L, Bartlett JD, MacLaren DP, Reilly T, Evans L, Mattey DL, Nixon NB, Drust B, Morton JP (2009) High-intensity interval training attenuates the exercise-induced increase in plasma IL-6 in response to acute exercise. Appl Physiol Nutr Metab 34:1098–1107 Davis JM, Nieman DC, Murphy AE (2007) Response to ‘‘unusually low plasma concentrations of lipopolysaccharide following 160-km race’’. Brain Behav Immun 21:515 Dinarello CA (1998) Interleukin-1, interleukin-1 receptors and interleukin-1 receptor antagonist. Int Rev Immunol 16:457–499 Donovan CM, Brooks GA (1983) Endurance training affects lactate clearance, not lactate production. Am J Physiol 244:E83–E92 Febbraio MA, Steensberg A, Starkie RL, McConell GK, Kingwell BA (2003) Skeletal muscle interleukin-6 and tumor necrosis factoralpha release in healthy subjects and patients with type 2 diabetes at rest and during exercise. Metabolism 52:939–944 Febbraio MA, Hiscock N, Sacchetti M, Fischer CP, Pedersen BK (2004) Interleukin-6 is a novel factor mediating glucose homeostasis during skeletal muscle contraction. Diabetes 53:1643–1648 Fielding RA, Manfredi TJ, Ding W, Fiatarone MA, Evans WJ, Cannon JG (1993) Acute phase response in exercise. III. Neutrophil and IL-1 beta accumulation in skeletal muscle. Am J Physiol 265:R166–R172 Fink R, Lu¨ttgau HC (1976) An evaluation of the membrane constants and the potassium conductance in metabolically exhausted muscle fibres. J Physiol 263:215–238 Fink R, Hase S, Lu¨ttgau HC, Wettwer E (1983) The effect of cellular energy reserves and internal calcium ions on the potassium conductance in skeletal muscle of the frog. J Physiol 336:211–228 Fischer CP (2006) Interleukin-6 in acute exercise and training: what is the biological relevance? Exerc Immunol Rev 12:6–33 Fischer CP, Plomgaard P, Hansen AK, Pilegaard H, Saltin B, Pedersen BK (2004) Endurance training reduces the contractioninduced interleukin-6 mRNA expression in human skeletal muscle. Am J Physiol 287:E1189–E1194 Froelicher VF Jr, Brammell H, Davis G, Noguera I, Stewart A, Lancaster MC (1974) A comparison of three maximal treadmill exercise protocols. J Appl Physiol 36:720–725 Gokhale R, Chandrashekara S, Vasanthakumar KC (2007) Cytokine response to strenuous exercise in athletes and non-athletes: an adaptive response. Cytokine 40:123–127

Goodman C, Henry G, Dawson B, Gillam I, Beilby J, Ching S, Fabian V, Dasig D, Kakulas B, Morling P (1997) Biochemical and ultrastructural indices of muscle damage after a twenty-one kilometre run. Aust J Sci Med Sport 29:95–98 Hawley JA (2002) Adaptations of skeletal muscle to prolonged, intense endurance training. Clin Exp Pharmacol Physiol 29:218–222 Helge JW, Stallknecht B, Pedersen BK, Galbo H, Kiens B, Richter EA (2003) The effect of graded exercise on IL-6 release and glucose uptake in human skeletal muscle. J Physiol 546:299–305 Holloszy JO, Booth FW (1976) Biochemical adaptations to endurance exercise in muscle. Annu Rev Physiol 38:273–291 Jeukendrup AE, Vet-Joop K, Sturk A, Stegen JH, Senden J, Saris WH, Wagenmakers AJ (2000) Relationship between gastrointestinal complaints and endotoxaemia, cytokine release and the acute-phase reaction during and after a long-distance triathlon in highly trained men. Clin Sci (Lond) 98:47–55 Jiang Q, DeTolla L, Kalvakolanu I, Singh IS, Fitzgerald B, Lipsky MM, Kane AS, Cross AS, Hasday JD (1999) Febrile range temperature modifies early systemic TNFa expression in bacterial endotoxin-challenged mice. Infect Immun 67:1539– 1546 Kimura H, Suzui M, Nagao F, Matsumoto K (2001) Highly sensitive determination of plasma cytokines by time-resolved fluoroimmunoassay; effect of bicycle exercise on plasma level of interleukin-1 alpha (IL-1 alpha), tumor necrosis factor alpha (TNF alpha), and interferon gamma (IFN gamma). Anal Sci 17:593–597 Miller BF, Fattor JA, Jacobs KA, Horning MA, Navazio F, Lindinger MI, Brooks GA (2002) Lactate and glucose interactions during rest and exercise in men: effect of exogenous lactate infusion. J Physiol 544:963–975 Neubauer O, Konig D, Wagner KH (2008) Recovery after an ironman triathlon: sustained inflammatory responses and muscular stress. Eur J Appl Physiol 104:417–426 Nicklas BJ, Hsu F-C, Brinkley TJ, Church T, Goodpaster BH, Kritchevsky SB, Pahor M (2008) Exercise training and plasma C-reactive protein and interleukin-6 in elderly people. J Am Geriatr Soc 56:2045–2052 Nieman DC, Nehlsen-Cannarella SL, Fagoaga OR, Henson DA, Utter A, Davis JM, Williams F, Butterworth DE (1998) Influence of mode and carbohydrate on the cytokine response to heavy exertion. Med Sci Sports Exerc 30:671–678 Nieman DC, Davis JM, Henson DA, Walberg-Rankin J, Shute M, Dumke CL, Utter AC, Vinci DM, Carson JA, Brown A, Lee WJ, McAnulty SR, McAnulty LS (2003) Carbohydrate ingestion influences skeletal muscle cytokine mRNA and plasma cytokine levels after a 3-h run. J Appl Physiol 94:1917–1925 Ostapiuk-Karolczuk J, Zembron-Lacny A, Naczk M, Gajewski M, Kasperska A, Dziewiecka H, Szyszka K (2012) Cytokines and cellular inflammatory sequence in non-athletes after prolonged exercise. J Sports Med Phys Fit 52:563–568 Ostrowski K, Hermann C, Bangash A, Schjerling P, Nielsen JN, Pedersen BK (1998a) A trauma-like elevation of plasma cytokines in humans in response to treadmill running. J Physiol 513:889–894 Ostrowski K, Rohde T, Zacho M, Asp S, Pedersen BK (1998b) Evidence that interleukin-6 is produced in human skeletal muscle during prolonged running. J Physiol 508:949–953 Ostrowski K, Rohde T, Asp S, Schjerling P, Pedersen BK (1999) Proand anti-inflammatory cytokine balance in strenuous exercise in humans. J Physiol 515:287–291 Phillips SM, Green HJ, Tarnopolsky MA, Heigenhauser GJ, Hill RE, Grant SM (1996) Effects of training duration on substrate turnover and oxidation during exercise. J Appl Physiol 81:2182–2191

123

Eur J Appl Physiol Reihmane D, Jurka A, Tretjakovs P, Dela F (2012) Increase in IL-6, TNF-a, and MMP-9, but not sICAM-1, concentrations depends on exercise duration. Eur J Appl Physiol. doi:10.1007/s00421012-2491-9 Rosa Neto JC, Lira FS, Oyama LM, Zanchi NE, Yamashita AS, Batista ML Jr, Oller do Nascimento CM, Seelaender M (2009) Exhaustive exercise causes an anti-inflammatory effect in skeletal muscle and a pro-inflammatory effect in adipose tissue in rats. Eur J Appl Physiol 106:697–704 Saltin B, Rowell LB (1980) Functional adaptations to physical activity and inactivity. Fed Proc 39:1506–1513 Scott JP, Sale C, Greeves JP, Casey A, Dutton J, Fraser WD (2011) Effect of exercise intensity on the cytokine response to an acute bout of running. Med Sci Sports Exerc 43:2297–2306 Serraa˜o FV, Foerster B, Spada S, Morales MM, Monteiro-Pedro V, Tannu´s A, Salvini TF (2003) Functional changes of human quadriceps muscle injured by eccentric exercise. Braz J Med Biol Res 36:781–786 Starkie RL, Rolland J, Angus DJ, Anderson MJ, Febbraio MA (2001) Circulating monocytes are not the source of elevations in plasma

123

IL-6 and TNF-alpha levels after prolonged running. Am J Physiol 280:C769–C774 Starkie RL, Hargreaves M, Rolland J, Febbraio MA (2005) Heat stress, cytokines, and the immune response to exercise. Brain Behav Immun 19:404–412 Steensberg A, van Hall G, Osada T, Sacchetti M, Saltin B, KlarlundPedersen B (2000) Production of interleukin-6 in contracting human skeletal muscles can account for the exercise-induced increase in plasma interleukin-6. J Physiol 529:237–242 Steensberg A, Keller C, Starkie RL, Osada T, Febbraio MA, Pedersen BK (2002) IL-6 and TNF-alpha expression in, and release from, contracting human skeletal muscle. Am J Physiol 283:E1272– E1278 Steensberg A, Fischer CP, Keller C, Moller K, Pedersen BK (2003) IL-6 enhances IL-1ra, IL-10, and cortisol in humans. Am J Physiol 285:E433–E437 Wolsk E, Mygind H, Grøndahl TS, Pedersen BK, van Hall G (2010) IL-6 selectively stimulates fat metabolism in human skeletal muscle. Am J Physiol Endocrinol Metab 299:E832–E840