High-intensity interval training attenuates the exercise-induced ...

1098

High-intensity interval training attenuates the exercise-induced increase in plasma IL-6 in response to acute exercise Louise Croft, Jonathan D. Bartlett, Don P.M. MacLaren, Thomas Reilly, Louise Evans, Derek L. Mattey, Nicola B. Nixon, Barry Drust, and James P. Morton

Abstract: This aims of this study were to investigate the effects of carbohydrate availability during endurance training on the plasma interleukin (IL)-6, IL-8, and tumor necrosis factor (TNF)-a response to a subsequent acute bout of highintensity interval exercise. Three groups of recreationally active males performed 6 weeks of high-intensity interval running. Groups 1 (LOW+GLU) and 2 (LOW+PLA) trained twice per day, 2 days per week, and consumed a 6.4% glucose or placebo solution, respectively, before every second training session and at regular intervals throughout exercise. Group 3 (NORM) trained once per day, 4 days per week, and consumed no beverage during training. Each group performed 50 min of high-intensity interval running at the same absolute workloads before and after training. Muscle glycogen utilization in the gastrocnemius muscle during acute exercise was reduced (p < 0.05) in all groups following training, although this was not affected by training condition. Plasma IL-6 concentration increased (p < 0.05) after acute exercise in all groups before and after training. Furthermore, the magnitude of increase was reduced (p < 0.05) following training. This training-induced attenuation in plasma IL-6 increase was similar among groups. Plasma IL-8 concentration increased (p < 0.05) after acute exercise in all groups, although the magnitude of increase was not affected (p > 0.05) by training. Acute exercise did not increase (p > 0.05) plasma TNF-a when undertaken before or after training. Data demonstrate that the exercise-induced increase in plasma IL-6 concentration in response to customary exercise is attenuated by previous exercise training, and that this attenuation appears to occur independent of carbohydrate availability during training. Key words: interval training, IL-6, IL-8, TNF-a, muscle glycogen, gastrocnemius. Re´sume´ : Cette e´tude se propose d’analyser les effets de la disponibilite´ des sucres au cours d’un programme d’entraıˆnement en endurance sur les concentrations plasmatiques de IL-6, IL-8 et de TNF-a note´es au cours d’une se´ance subse´quente d’exercices par intervalle de forte intensite´. Trois groupes d’hommes physiquement actifs sur le plan du loisir participent a` un programme d’entraıˆnement par intervalles a` la course de haute intensite´ et d’une dure´e de six semaines. Les groupes 1 (LOW+GLU) et 2 (LOW+PLA) qui s’entraıˆnent deux fois par jour et deux jours par semaine consomment respectivement une boisson de glucose (6,4 %) et une solution placebo, et ce, avant chaque deuxie`me se´ance d’entraıˆnement et a` intervalles re´guliers au cours de la se´ance d’entraıˆnement. Le groupe 3 (NORM) qui s’entraıˆne une fois par jour, 4 jours par semaine ne consomme aucune boisson tout au long du programme d’entraıˆnement. Tous les groupes participent a` une se´ance de course par intervalles de meˆme intensite´ absolue et d’une dure´e de 50 min, et ce, avant et apre`s le programme d’entraıˆnement. Chez tous les groupes apre`s la fin du programme d’entraıˆnement, on observe au cours de l’effort de bre`ve dure´e une diminution de l’utilisation du glycoge`ne musculaire (p < 0,05) dans le muscle gastrocnemius; le mode d’entraıˆnement n’a rien a` voir avec cette diminution. Avant et apre`s le programme d’entraıˆnement, on observe chez tous les groupes a` la suite de l’effort de bre`ve dure´e une augmentation de la concentration plasmatique de IL-6 (p < 0,05). De plus, on observe une moins grande augmentation (p < 0,05) a` la suite du programme d’entraıˆnement. La moins grande augmentation de la concentration plasmatique de IL-6 cause´e par l’entraıˆnement est semblable chez tous les groupes. Chez tous les groupes, on observe une augmentation de la concentration plasmatique de IL-8 (p < 0,05) apre`s l’effort de bre`ve dure´e, mais le degre´ d’augmentation n’est pas influence´ (p > 0,05) par l’entraıˆnement. Que ce soit avant ou apre`s le programme d’entraıˆnement, l’effort de bre`ve dure´e n’augmente pas la concentration plasmatique de TNF-a (p > 0,05). D’apre`s ces observations, l’augmentation de la concentration plasmatique de IL-6 suscite´e par un effort coutumier est atte´nue´e par un entraıˆnement pre´alable, mais cette atte´nuation se manifeste inde´pendamment de la disponibilite´ des sucres au cours du programme d’entraıˆnement. Received 10 June 2009. Accepted 22 September 2009. Published on the NRC Research Press Web site at apnm.nrc.ca on 11 December 2009. L. Croft,1 J.D. Bartlett, D.P.M. MacLaren, T. Reilly, B. Drust, and J.P. Morton.2 Research Institute for Sport and Exercise Sciences, Liverpool John Moores University, 15-21 Webster Street, Liverpool, L3 2ET, UK. L. Evans. Stepping Hill Hospital, Stockport NHS Foundation Trust, Poplar Grove, Stockport, SK2 7JE, UK. D.L. Mattey and N.B. Nixon. Staffordshire Rheumatology Centre, The Haywood Hospital, High Lane, Burslem, Stoke-on-Trent, ST6 7AG, UK. 1Present

address: School of Sport and Exercise Sciences, Loughborough University, Loughborough, Leicestershire, LE11 3T, UK. author (e-mail: [email protected]).

2Corresponding

Appl. Physiol. Nutr. Metab. 34: 1098–1107 (2009)

doi:10.1139/H09-117

Published by NRC Research Press

Croft et al.

1099

Mots-cle´s : entraıˆnement par intervalles, IL-6, IL-8, TNF-a, glycoge`ne musculaire, gastrocnemius. [Traduit par la Re´daction]

______________________________________________________________________________________ Introduction The plasma concentration of interleukin (IL)-6 (Nieman et al. 1998; Ostrowski et al. 1998; Starkie et al. 2001a; Steensberg et al. 2001), IL-8 (Nieman et al. 2003), and tumor necrosis factor (TNF)-a (Nieman et al. 2003) increases following a variety of acute exercise challenges. Data also demonstrate that skeletal muscle is one of the major contributing sites of origin for the increased circulating cytokine levels observed following exercise (Starkie et al. 2001b; Steensberg et al. 2002). In this regard, it is increasingly thought that muscle-derived cytokines may have an active role in mediating metabolic adaptations to exercise and training (Pedersen and Febbraio 2008). Indeed, the contraction-induced increase in intramuscular IL-6 gene expression is exacerbated when the exercise protocol is commenced with reduced pre-exercise muscle glycogen stores (Keller et al. 2001) Furthermore, glucose ingestion during exercise attenuates the exercise-induced increase in both muscle IL-6 messenger RNA and plasma IL-6 concentration (Nieman et al. 2003). In addition, muscle IL-8 gene expression is enhanced under conditions of low muscle glycogen (Chan et al. 2004) and attenuated with glucose ingestion during exercise (Nieman et al. 2003), although plasma levels were not affected in either instance. Although the effects of acute exercise on plasma cytokine responses are well studied, there are few investigations of the effects of exercise training on plasma cytokine response to subsequent acute exercise. This paucity of data is surprising, given that exercise training results in skeletal muscle adaptations that function to reduce the degree of metabolic perturbations in homeostasis occurring during acute exercise (Yeo et al. 2008). To our knowledge, the only study of this topic employed a 1-legged knee-extensor exercise and training protocol in which the contraction-induced increase in plasma IL-6 was unaffected by the training period (Fischer et al. 2004). However, in this instance, the post-training protocol was performed at the same relative exercise intensity (hence, an increased absolute workload) as the pretraining exercise challenge, and muscle glycogen utilization during contractions was not reduced following the training period. The primary aim of this study was, therefore, to examine the effects of short-term high-intensity interval exercise training on the plasma cytokine response to acute highintensity interval exercise performed at the same absolute intensities as that undertaken prior to training. Because carbohydrate availability is known to affect the cytokine response (most notably, IL-6) to acute exercise, our secondary aim was to examine how carbohydrate availability during repeated acute exercise exposure (i.e., training) affects the plasma cytokine response to a subsequent acute exercise stress performed in the trained state. Our second aim is important, given the findings from our group (Morton et al. 2009) and others (Hansen et al. 2005; Yeo et al. 2008) that demonstrate that training in conditions of reduced carbohydrate availability induces greater increases in skeletal muscle

oxidative enzyme activity, thus inducing increased rates of fat oxidation during submaximal exercise (Yeo et al. 2008). We hypothesized that exercise training would attenuate the exercise-induced increase in plasma cytokine levels in response to an acute exercise stress, and that performing training sessions in conditions of reduced carbohydrate availability would further attenuate this response.

Materials and methods Subjects Fifteen recreationally active males volunteered to participate in the study. After giving written informed consent to participate, subjects were randomly assigned (according to Altman 1991) to 1 of 3 training groups, each with 5 subjects. The characteristics of those in each group who completed the study are shown in Table 1. There were no significant differences (p > 0.05) in baseline maximal oxygen consumption (V_ O2 max) among groups, as determined by a 1-way between-group general linear model, suggesting that all groups were of similar initial physiological fitness. All subjects refrained from alcohol and caffeine intake for at least 24 h prior to any of the test sessions (excluding training sessions). None of the subjects had any history of neurological disease or musculoskeletal abnormality, and none was under any pharmacological treatment during the course of the study. The study was approved by the Ethics Committee of Liverpool John Moores University (Liverpool, UK). Using the NQuery statistical software package (Statistical Solutions, Cork, Ireland) and appropriate statistical guidelines (Batterham and Atkinson 2005), it was calculated that a sample size of 5 subjects in each group would enable detection of an exercise-induced plasma cytokine increase of 100%, assuming a standard deviation (SD) of differences of 50% and a statistical power of 90%. This effect size and SD are based on relevant investigations that have demonstrated significant increases in plasma cytokine concentration (Febbraio et al. 2004; Starkie et al. 2001a). To estimate sample size for our secondary aim, we performed a 1-way general linear model using the NQuery statistical software package, where the magnitude of the cytokine response was 100% in the control dietary condition but was attenuated by 50% in the carbohydrate intervention conditions. Assuming an SD of differences between groups of 25% and a statistical power of 90%, it was calculated that a sample size of 5 subjects in each group was sufficient. These effect sizes and SDs were based on a review of previous studies that have shown an effect of either endogenous (Keller et al. 2001) or exogenous (Starkie et al. 2001a) carbohydrate availability in determining the cytokine response to acute exercise. Experimental design After having initially been assessed for V_ O2 max and running performance on the Yo-Yo intermittent recovery test 2 Published by NRC Research Press

1100

Appl. Physiol. Nutr. Metab. Vol. 34, 2009

Table 1. Subject characteristics of each training group. Characteristics Age (y) Height (m) Body mass (kg)

LOW+GLU (n = 5) 20±1 1.78±0.04 80±7

LOW+PLA (n = 5) 21±1 1.78±0.02 73±8

NORM (n = 5) 20±1 1.8±0.03 76±9

Note: LOW+GLU, subjects trained twice per day, 2 days per week, and consumed a 6.4% carbohydrate beverage before every second training session and at regular intervals throughout exercise; LOW+PLA, subjects trained twice per day, 2 days per week, and consumed placebo solution before every second training session and at regular intervals throughout exercise; NORM, subjects trained once per day, 4 days per week, and consumed no beverage throughout training.

(YoYoIRT2) (see section on Assessment of physiological fitness, below), subjects completed 6 weeks of high-intensity interval running 4 times per week. Each group, therefore, performed a total of 24 training sessions throughout the 6-week period. Groups 1 and 2 trained twice per day, 2 days per week (once in the morning and once in the afternoon; training sessions were separated by a 3- to 4-h rest period), whereas Group 3 trained once per day, 4 days per week. In this way, each group performed the same amount of work throughout the training period, yet groups 1 and 2 performed every second training session with reduced pre-exercise muscle glycogen levels (Morton et al. (2009) demonstrated that subjects commence the afternoon training session with glycogen stores depleted by approximately 45% in the gastrocnemius muscle). To allow us to also examine the effects of exogenous glucose supplementation, subjects in Group 1 (LOW+GLU) consumed a 6.4% carbohydrate beverage (GlaxoSmithKline Consumer Healthcare, UK) immediately prior to and throughout every second (i.e., afternoon) training session; subjects in Group 2 (LOW+PLA) consumed an identical amount of a taste-, consistency-, and odourmatched placebo solution at identical times. Subjects consumed 8 mL
kg–1 of the relevant solution in a 10–15 min period immediately prior to exercise, and a further 3 mL
kg–1 in the active recovery periods following interval 1 and interval 3 (see section on High-intensity intermittent training protocol, below, for a description of the training stimulus). In contrast, subjects in Group 3 (NORM) commenced every training session with normal glycogen stores and consumed no beverages during any of their training sessions. To examine the effects of training on the plasma cytokine response to exercise, the first training session served as a ‘‘criterion’’ exercise session. At 72 h after completion of the last training session, all subjects completed an additional criterion exercise session, which was performed at the same absolute running speed as the first criterion exercise session. Venous blood samples and muscle biopsies (from the gastrocnemius muscle) were obtained immediately before and after both of these criterion exercise sessions. Biopsy samples were assessed for muscle glycogen concentration, and blood samples were analysed for plasma IL-6, IL-8, TNF-a, and lactate concentration. A schematic illustration and information on the experimental design are shown in Fig. 1. Dietary interventions On each day of testing or training, subjects arrived in the laboratory 2 h after having consumed a standardized break-

fast (3.64 MJ, 863 kcal: 63% carbohydrate, 25% fat, 12% protein). For those in groups that trained twice per day (i.e., LOW+GLU and LOW+PLA), the morning and afternoon training sessions were separated by a 3- to 4-h recovery period. At the midpoint of this interval (i.e., 1.5 to 2 h following completion of the morning exercise session), each subject consumed a standardized low-carbohydrate lunch, containing 1.1, and a plateau of oxygen consumption despite increasing workload. At least 24 h after completion of this test, subjects were assessed for intermittent exercise performance using the YoYoIRT2 (Krustrup et al. 2006). This test consists of two 20-m runs at a progressively increased running speed, controlled by audio beeps from a tape recorder. Between each running bout, subjects have a 10-s rest period. When the subject twice fail to reach the finish line in the required time, the distance covered is recorded, and repre-

sents the test result. The test was performed on an indoor running surface on a 4-m wide and 20-m long running lane marked by cones. The coefficient of variation for test–retest reliability of this test in our laboratory is 8.8%. Each subject performed pre- and post-training fitness assessments on the same of day to prevent any diurnal variations in performance (Reilly and Brooks 1986). High-intensity intermittent training protocol The intermittent training protocol consisted of highintensity running and was performed on a motorised treadmill. The protocol commenced with a 10-min warm-up at a running velocity corresponding to 70% of V_ O2 max, followed by five 3-min bouts at a running velocity corresponding to 90% V_ O2 max. The high-intensity efforts were separated by 3-min active recovery periods (1.5 min at a velocity corresponding to 25% V_ O2 max, followed by 1.5 min at a velocity corresponding to 50% V_ O2 max). Following the interval and recovery periods, subjects performed a 10-min cool-down period at a running velocity corresponding to 70% of V_ O2 max. The exercise protocol, therefore, gave a total of 15 min of interval exercise and 15 min of active recovery time, for a total time for the intermittent training protocol of 30 min. When warm-up and cool-down times were included, the total duration of the exercise protocol was 50 min. Training intensities were increased by 5% of the original V_ O2 max following 2 weeks of training, and by a further 5% after 4 weeks of training. Heart rate was measured continuously during each training session (Polar S610i, Kempele, FinPublished by NRC Research Press

1102

land). Our intention for choosing this duration of training and specific training protocol was an attempt to recreate training practices that are currently employed in those sports characterized by intermittent activity profiles, such as soccer (e.g., Helgerud et al. 2001; Hoff and Helgerud 2004). Furthermore, the 6-week intervention period is representative of the preseason period in these sports, and the initial study using this protocol (Morton et al. 2009) demonstrated that this intensity and duration of training is sufficient to induce improvements in V_ O2 max, high-intensity interval running performance, and oxidative enzyme adaptations of the gastrocnemius muscle. Furthermore, interval training is receiving increased recognition as a superior training method, compared with continuous training approaches (Helgerud et al. 2007). Tissue collection and biochemical analysis Muscle biopsies (~10 mg) were obtained from the lateral portion of the gastrocnemius muscle under local anesthesia (0.5% marcaine), using a Pro-Mag 2.2 biopsy gun (MDTECH, Manan Medical Products, Northbrook, Ill.). Once the biopsy needle is inserted through the fascia, the ‘‘firing’’ of the biopsy gun operates with a feed-forward of up to 2.5– 3.5 cm, depending on the angle of insertion of the needle. Once obtained, samples were immediately frozen in liquid nitrogen and stored at –80 8C for later analysis. Muscle glycogen concentration was determined in wet mass, according to the hexokinase method (Lowry and Passonneau 1972). Briefly, samples were homogenized in 1 mol
L–1 HCl and boiled for 10 min. Degradation of glycogen was completed after 3 h of incubation, through the enzymatic action of amyloglucosidase. Samples were subsequently neutralized with 1 mol
L–1 NaOH and assayed for glucose using the hexokinase method. Blood samples (~6 mL from each time of sampling) were drawn from a superficial vein in the anticubital crease of the forearm, using standard veinepuncture techniques (Becton Dickinson Vacutainer Systems, Europe). Samples were collected in vacutainers containing EDTA and stored on ice until centrifugation at 2000 r
min–1 (1000g) for 15 min at 4 8C. Following centrifugation, aliquots of plasma were stored at –80 8C for later analysis. Samples were analysed for plasma lactate (Randox Laboratories Ltd. Co., Antrim, UK), IL-6, IL-8, and TNF-a concentration using a multiplex, bead-based assay system (Fluorokine multi-analyte profiling kit, human cytokine panel A, R & D Systems, Minneapolis, Minn.) measured on a Bio-Plex 200 analyser (Bio-Rad Laboratories, Hemel Hempstead, Hertfordshire, UK). All postexercise blood samples were corrected for plasma volume changes, according to the methods of Dill and Costill (1974). Statistical analysis Statistical analysis was conducted using the Statistical Package for Social Sciences software programme (version 15). Changes in V_ O2 max and YoYoIRT2 were assessed using a 2-way mixed design repeated-measures general linear model, where the within factor was time (pretraining vs. post-training) and the between factor was training condition (LOW+GLU vs. LOW+PLA vs. NORM). Changes in the plasma cytokine–lactate response and muscle glycogen utilization during the criterion exercise session were assessed

Appl. Physiol. Nutr. Metab. Vol. 34, 2009 Table 2. The effects of the training period on maximal oxygen consumption and distance covered on the Yo-Yo intermittent recovery test 2 (YoYoIRT2) test. Pretraining –1 _ VO2 max (mL
kg
min–1) LOW+GLU 51.8±5.1 LOW+PLA 59±3.4 NORM 55.9±6.8 YoYoIRT2 (m) LOW+GLU 407±103 LOW+PLA 440±75 NORM 393±133

Post-training 57.4±5.7* 62.8±2.1* 60.8±5.3* 473±120* 512±82* 480±199*

Note: V˙2 max, maximal oxygen consumption. *Significant main effect of training, p < 0.05.

using a 3-way mixed design repeated-measures general linear model, where the within factors were acute exercise response (pre-exercise vs. postexercise) and training status (pretraining vs. post-training) and the between factor was training condition (LOW+GLU vs. LOW+PLA vs. NORM). Where there was a significant interaction effect, Bonferroni post hoc tests were used to locate the differences. All data in text, figures, and tables are presented as means ± SD (as opposed to SEM, which does not provide an appropriate estimation of the variability observed among the actual individual observations in a sample mean), with p values £0.05 indicating statistical significance.

Results Changes in physiological fitness following training Each group’s V_ O2 max and distance covered on the YoYoIRT2 before and after training are displayed in Table 2. There was a significant (p < 0.05) main effect of training for both performance parameters in all groups. There was, however, no significant difference (p > 0.05) in the traininginduced increase in performance between training conditions. Effects of training on the plasma cytokine response to the criterion exercise session Both before and after the training period, there was a significant main effect (p < 0.05) of acute exercise on plasma IL-6 concentration in all groups (Fig. 2A). The increase in plasma IL-6 following the acute exercise protocol was significantly greater before training (LOW+GLU, 3.3 pg
mL–1; LOW+PLA, 3.7 pg
mL–1; NORM, 2.8 pg
mL–1) than after training (LOW+GLU, 2.2 pg
mL–1; LOW+PLA, 1.4 pg
mL–1; NORM, 1.6 pg
mL–1). There was no significant 3-way interaction of acute exercise, training, and group, demonstrating that the training-induced attenuation in the magnitude of increase in exercise-induced plasma IL-6 was similar among groups. Before and after training, there was a significant main effect (p < 0.05) of acute exercise on plasma IL-8 concentration in all groups (Fig. 2B). There was, however, no significant (p > 0.05) effect of training on the magnitude of the increase in plasma IL-8 in response to the acute exercise protocol (pretraining: LOW+GLU, 1.2 pg
mL–1; LOW+PLA, 3.1 pg
mL–1; NORM, 1.9 pg
mL–1; post-training: LOW+GLU, 1 pg
mL–1; LOW+PLA, 1.6 pg
mL–1; NORM, 1.1 pg
mL–1). Published by NRC Research Press

Croft et al.

1103

Fig. 2. The effects of the training period on the plasma (A) IL-6, (B) IL-8, and (C) TNF-a response to the criterion exercise session. *, Significant main effect of exercise, p 0.05) in the training-induced reduction in muscle glycogen utilization between training conditions. Plasma lactate significantly (p < 0.05) increased after acute exercise (Fig. 4) when the protocol was performed either before or after training. Similar to muscle glycogen utilization, plasma lactate showed a significant main effect of training; the magnitude of the increase in plasma lactate in response to acute exercise was significantly reduced (p < 0.05) following training. There was no significant main effect of group, demonstrating that the trainingPublished by NRC Research Press

1104 Fig. 3. The effects of the training period on muscle glycogen utilization of the gastrocnemius muscle during the criterion exercise session. *, Significant main effect of training, p < 0.05.

induced reduction in plasma lactate response was similar between groups.

Discussion The aims of this study were 2-fold. First, we sought to examine the effects of short-term high-intensity interval training on the magnitude of exercise-induced increases in plasma IL-6, IL-8, and TNF-a levels following an acute high-intensity interval exercise challenge performed at the same absolute intensities as those undertaken prior to training. Second, because carbohydrate availability is known to influence the cytokine response (most notably, IL-6) to acute exercise, we aimed to determine how carbohydrate availability during repeated acute exercise exposure (i.e., training) affects the plasma cytokine response to an acute exercise stress performed in the trained state. Using a whole-body and ecologically valid (i.e., applicable to the general population) exercise and training protocol, we showed, for the first time, that the increased circulating plasma IL-6 levels following an acute customary exercise protocol is reduced with training. Furthermore, this traininginduced attenuation in IL-6 levels following high-intensity interval exercise appeared to occur independent of carbohydrate availability during training. We also present novel findings, in that acute high-intensity interval exercise induces statistically significant increases in plasma IL-8, which were apparent in both the untrained and trained state. Consistent with previous data (Pedersen and Febbraio 2008; Steensberg et al. 2002), we showed that plasma TNF-a levels are unaffected by acute nondamaging-type exercise protocols, despite the high-intensity exhaustive intervals associated with our chosen intervention. Plasma IL-6 has been consistently shown to increase following acute knee-extensor, cycling, running, and resistance exercise protocols (for a review, see Fischer 2006). Steensberg et al. (2001) demonstrated that IL-6 release from the contracting skeletal muscles can account for this increase, where it is thought to exert an endrocrine-like action upon the liver to maintain glucose homeostasis (Pedersen and Febbraio 2008). In general, the magnitude of this increase

Appl. Physiol. Nutr. Metab. Vol. 34, 2009

appears to be largely dependent on the duration of the exercise stimulus where a linear relationship is evident (Fischer 2006). However, short-duration (e.g., as little as 6 min) high-intensity maximal effort can induce approximately 2-fold increases in plasma IL-6 (Nielsen et al. 1996), suggesting that exercise intensity may also be an important factor influencing the IL-6 response to acute exercise. In this study, we used a high-intensity intermittent protocol as both our exercise and training stimulus, where the majority of the stress is provided during the five 3-min bouts undertaken at a running velocity corresponding to 90% V_ O2 max. Prior to the training period, this protocol also resulted in approximately 2-fold increases in circulating IL-6 levels, further highlighting that relatively short duration but high-intensity effort can markedly influence systemic IL-6 levels. Given that considerable evidence exists suggesting that IL-6 functions as a myokine involved in energy metabolism (particularly that of maintaining glucose homeostasis (Pedersen and Febbraio 2008)), it seems reasonable to postulate that the magnitude of the IL-6 increase in response to exercise in the trained state would be significantly reduced because of the metabolic adaptations evident in trained skeletal muscle. However, Fischer et al. (2004) observed that the plasma IL-6 response to acute knee-extensor exercise is unaffected by 10 weeks of training. In contrast, our data demonstrate that the magnitude of the exercise-induced plasma IL-6 increase in response to acute whole-body exercise undertaken in the trained state is effectively halved, compared with pretraining values. Our study differs in methodological design from that of Fischer et al. (2004), in that our post-training exercise protocol was performed at the same absolute running speeds (hence, reduced relative workload) as the pretraining exercise stress. Furthermore, whereas Fischer et al. (2004) observed no significant reduction in muscle glycogen utilization following training, our data demonstrate a training-induced reduction in muscle glycogen utilization and postexercise blood lactate concentration, indicating a reduced reliance on carbohydrate stores to support ATP production. It is, therefore, possible that the reduced circulating IL-6 levels following acute exercise undertaken following short-term endurance training may be due to a training-induced reduction in carbohydrate utilization during exercise, thus reducing the need for muscle-derived IL-6 to stimulate hepatic glucose production during exercise (Febbraio et al. 2004). Alternatively, the training-induced attenuation in plasma IL-6 in response to acute exercise may be due to reduced exercise-induced reactive oxygen species and (or) heat shock protein formation that is evident in the trained state (Morton et al. 2008; Brooks et al. 2008), given that both of the aforementioned factors are known regulators of IL-6 gene transcription (Fischer 2006). It should be noted, however, that it is difficult to make firm conclusions regarding the apparent training-induced reduction in circulating IL-6 levels in response to acute exercise because the measurement of plasma IL-6 per se cannot provide definitive insight into the effects of endurance training on IL-6 production from muscle cells versus clearance from nonmuscle cells. Typically, the plasma concentration of IL-8 has only been shown to increase following prolonged damaging exercise protocols, such as marathon running (Nieman et al. 2003, Published by NRC Research Press

Croft et al.

1105

Fig. 4. The effects of the training period on the plasma lactate response to the criterion exercise session. *, Significant main effect of exercise, p < 0.05; {; significant main effect of training, p < 0.05.

2005, 2001), whereas shorter duration nondamaging exercise protocols, such as cycling (Chan et al. 2004), do not increase IL-8 levels. It should be noted, however, that highintensity exhaustive cycling (10–15 min) has shown small but statistically significant increases in circulating IL-8 levels (Mucci et al. 2000). Our data appear to be in agreement with those of Mucci et al. (2000), in that the short-duration high-intensity intermittent exercise protocol we employed also induced small but significant increases in plasma IL-8, which, unlike IL-6, occurred in both the untrained and trained state. Collectively, these data suggest that detectable increases in plasma IL-8 can be observed if a threshold (and yet to be determined) exercise intensity is exceeded. Skeletal muscle has been identified as the likely source of the small increases in plasma IL-8 following exercise (Akerstrom et al. 2005; Nieman et al. 2003), and it is suggested that this cytokine primarily mediates its effects in a paracrine-like fashion, acting locally (as opposed to IL-6) where, in the absence of muscle damage, it is thought to stimulate angiogenesis (Pedersen and Febbraio 2008). Further research is needed, however, to elucidate the precise biological significance of the local and apparent small systemic exerciseinduced increases in IL-8. In contrast to IL-6 and IL-8, our data demonstrate that plasma TNF-a levels do not appear to increase during short-duration nondamaging exercise, even if the exercise is intense. These data are consistent with previous findings (Steensberg et al. 2002), and suggest that more prolonged and damaging exercise protocols, such as marathon running (Ostrowski et al. 1999), are needed to detect increases in circulating TNF-a concentration. To address our second aim, we employed an experimental design in which both endogenous and exogenous carbohydrate availability during training were manipulated. This is important for 2 reasons: the plasma cytokine response to acute exercise is known to be accentuated during conditions of low carbohydrate availability; and training in conditions of reduced carbohydrate availability can augment the capacity for lipid oxidation (i.e., improved oxidative enzyme

activity (Hansen et al. 2005; Morton et al. 2009; Yeo et al. 2008)), which, in turn, may attenuate the plasma cytokine response to exercise undertaken in the trained state. However, despite the manipulation of carbohydrate availability between training conditions, all groups displayed a similar attenuation of the plasma IL-6 response to exercise. This was also the case for data concerning muscle glycogen utilization and postexercise blood lactate concentrations. Taken together, these data suggest that the training-induced alterations in metabolism during the high-intensity intermittent exercise protocol observed here occurred independent of carbohydrate availability during the 6 weeks of high-intensity intermittent training. It is possible, however, that if we had employed a continuous exercise protocol at submaximal intensities, such as at 60% V_ O2 max (as opposed to highintensity intermittent exercise), we may have seen a more pronounced attenuation of the IL-6 response in subjects who trained in the LOW+PLA condition, because such training conditions can enhance the activity of key oxidative enzymes involved in lipid metabolism (Hansen et al. 2005; Morton et al. 2009; Yeo et al. 2008). Indeed, whereas lipid oxidation rates can increase with training during these submaximal intensities (Yeo et al. 2008), it is well recognised that lipid oxidation cannot be concomitantly increased during high-intensity exercise (such as the acute exercise protocol studied here), despite the higher oxidative enzyme activity that has previously been observed in the LOW+PLA training conditions. Further research using a variety of criterion exercise protocols, therefore, appears warranted. We acknowledge that our sample size of 5 subjects in each group may have been underpowered to address our secondary aim. It was difficult to perform detailed a priori sample size estimation, given the lack of research that has chronically manipulated carbohydrate availability. We therefore based our calculations on acute studies in which carbohydrate availability appears to affect exercise-induced plasma IL-6 levels in the order of 50% (Keller et al. 2001; Steensberg et al. 2001). Nevertheless, it is possible that, in Published by NRC Research Press

1106

a chronic situation, an affect of carbohydrate availability is more subtle than this magnitude and, in this regard, larger sample sizes would be needed to maintain statistical power at 90%. In summary, our data show, for the first time, that the exercise-induced increase in plasma IL-6 concentration following acute whole-body exercise is attenuated with exercise training. Such findings may be due to a training-induced reduction in carbohydrate utilization during exercise (thereby reducing the need for IL-6 to stimulate hepatic glucose production), although further research is warranted to provide more mechanistic insight into the effects of exercise training on muscle-derived IL-6 production versus clearance from nonmuscle cells. Our preliminary evidence also suggests that reduced carbohydrate availability during training does not further attenuate the exercise-induced increase in plasma IL-6 levels following acute high-intensity exercise.

Acknowledgements This study was supported by a research grant from GlaxoSmithKline Consumer Healthcare (UK). The authors acknowledge the technical assistance of Dave Stokes, Andrew Hulton, and Padraic Phibbs during data collection, and the outstanding efforts of all subjects during the intensive training period. Finally, the independent statistical advice of Prof. Greg Atkinson is greatly appreciated.

References Akerstrom, T., Steensberg, A., Keller, P., Keller, C., Penkowa, M., and Pedersen, B.K. 2005. Exercise induces interleukin-8 expression in human skeletal muscle. J. Physiol. 563(Pt. 2): 507–516. doi:10.1113/jphysiol.2004.077610. PMID:15618276. Altman, D.G. (Editor). 1991. Practical statistics for medical research. Chapman and Hall, London, UK. Batterham, A., and Atkinson, G.A. 2005. How big does my sample size need to be? A primer on the murky world of sample size estimation. Phys. Ther. Sport, 6(3): 153–163. doi:10.1016/j.ptsp. 2005.05.004. Brooks, S.V., Vasilaki, A., Larkin, L.M., McArdle, A., and Jackson, M.J. 2008. Repeated bouts of aerobic exercise lead to reductions in skeletal muscle free radical generation and nuclear factor kappaB activation. J. Physiol. 586(16): 3979–3990. doi:10.1113/jphysiol.2008.155382. PMID:18591188. Chan, M.H., Carey, A.L., Watt, M.J., and Febbraio, M.A. 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(2): R322–R327. PMID:15072962. Dill, D.B., and Costill, D.L. 1974. Calculation of percentage changes in volumes of blood, plasma, and red cells in dehydration. J. Appl. Physiol. 37(2): 247–248. PMID:4850854. Febbraio, M.A., Hiscock, N., Sacchetti, M., Fischer, C.P., and Pedersen, B.K. 2004. Interleukin-6 is a novel factor mediating glucose homeostasis during skeletal muscle contraction. Diabetes, 53(7): 1643–1648. doi:10.2337/diabetes.53.7.1643. PMID: 15220185. Fischer, C.P. 2006. Interleukin-6 in acute exercise and training: what is the biological relevance? Exerc. Immunol. Rev. 12: 6– 33. PMID:17201070. Fischer, C.P., Plomgaard, P., Hansen, A.K., Pilegaard, H., Saltin, B., and Pedersen, B.K. 2004. Endurance training reduces the contraction-induced interleukin-6 mRNA expression in human skele-

Appl. Physiol. Nutr. Metab. Vol. 34, 2009 tal muscle. Am. J. Physiol. Endocrinol. Metab. 287(6): E1189– E1194. doi:10.1152/ajpendo.00206.2004. PMID:15304377. Hansen, A.K., Fischer, C.P., Plomgaard, P., Andersen, J.L., Saltin, B., and Pedersen, B.K. 2005. Skeletal muscle adaptation: training twice every second day vs. training once daily. J. Appl. Physiol. 98(1): 93–99. doi:10.1152/japplphysiol.00163.2004. PMID: 15361516. Helgerud, J., Engen, L.C., Wisloff, U., and Hoff, J. 2001. Aerobic endurance training improves soccer performance. Med. Sci. Sports Exerc. 33(11): 1925–1931. doi:10.1097/00005768200111000-00019. PMID:11689745. Helgerud, J., Høydal, K., Wang, E., Karlsen, T., Berg, P., Bjerkaas, M., et al. 2007. Aerobic high-intensity intervals improve VO2max more than moderate training. Med. Sci. Sports Exerc. 39(4): 665–671. doi:10.1249/mss.0b013e3180304570. PMID: 17414804. Hoff, J., and Helgerud, J. 2004. Endurance and strength training for soccer players: physiological considerations. Sports Med. 34(3): 165–180. doi:10.2165/00007256-200434030-00003. PMID: 14987126. Keller, C., Steensberg, A., Pilegaard, H., Osada, T., Saltin, B., Pedersen, B.K., and Neufer, P.D. 2001. Transcriptional activation of the IL-6 gene in human contracting skeletal muscle: influence of muscle glycogen content. FASEB J. 15(14): 2748– 2750. PMID:11687509. Krustrup, P., Mohr, M., Nybo, L., Jensen, J.M., Nielsen, J.J., and Bangsbo, J. 2006. The Yo-Yo IR2 test: physiological response, reliability, and application to elite soccer. Med. Sci. Sports Exerc. 38(9): 1666–1673. doi:10.1249/01.mss.0000227538.20799. 08. PMID:16960529. Lowry, O.H., and Passonneau, J.V. (Editors). 1972. A flexible system of enzymatic analysis. Academic Press, New York. Morton, J.P., Maclaren, D.P.M., Cable, N.T., Campbell, I.T., Evans, L., Kayani, A.C., et al. 2008. Trained men display increased Basal heat shock protein content of skeletal muscle. Med. Sci. Sports Exerc. 40(7): 1255–1262. doi:10.1249/MSS. 0b013e31816a7171. PMID:18580405. Morton, J.P., Croft, L., Bartlett, J.D., Maclaren, D.P.M., Reilly, T., Evans, L., et al. 2009. Reduced carbohydrate availability does not modulate training-induced heat shock protein adaptations but does upregulate oxidative enzyme activity in human skeletal muscle. J. Appl. Physiol. 106(5): 1513–1521. doi:10.1152/ japplphysiol.00003.2009. PMID:19265068. Mucci, P., Durand, F., Lebel, B., Bousquet, J., and Pre´faut, C. 2000. Interleukins 1-beta, -8, and histamine increases in highly trained, exercising athletes. Med. Sci. Sports Exerc. 32(6): 1094–1100. doi:10.1097/00005768-200006000-00009. PMID:10862535. Nielsen, H.B., Secher, N.H., Christensen, N.J., and Pedersen, B.K. 1996. Lymphocytes and NK cell activity during repeated bouts of maximal exercise. Am. J. Physiol. 271(1 Pt. 2): R222–R227. PMID:8760224. Nieman, D.C., Nehlsen-Cannarella, S.L., Fagoaga, O.R., Henson, D.A., Utter, A., Davis, J.M., et al. 1998. Influence of mode and carbohydrate on the cytokine response to heavy exertion. Med. Sci. Sports Exerc. 30(5): 671–678. doi:10.1097/00005768199805000-00005. PMID:9588607. Nieman, D.C., Henson, D.A., Smith, L.L., Utter, A.C., Vinci, D.M., Davis, J.M., et al. 2001. Cytokine changes after a marathon race. J. Appl. Physiol. 91(1): 109–114. PMID:11408420. Nieman, D.C., Davis, J.M., Henson, D.A., Walberg-Rankin, J., Shute, M., Dumke, C.L., et al. 2003. Carbohydrate ingestion influences skeletal muscle cytokine mRNA and plasma cytokine levels after a 3-h run. J. Appl. Physiol. 94(5): 1917–1925. PMID:12533503. Published by NRC Research Press

Croft et al. Nieman, D.C., Davis, J.M., Henson, D.A., Gross, S.J., Dumke, C.L., Utter, A.C., et al. 2005. Muscle cytokine mRNA changes after 2.5 h of cycling: influence of carbohydrate. Med. Sci. Sports Exerc. 37(8): 1283–1290. doi:10.1249/01.mss. 0000175054.99588.b1. PMID:16118573. Ostrowski, K., Hermann, C., Bangash, A., Schjerling, P., Nielsen, J.N., and Pedersen, B.K. 1998. A trauma-like elevation of plasma cytokines in humans in response to treadmill running. J. Physiol. 513(Pt. 3): 889–894. doi:10.1111/j.1469-7793.1998. 889ba.x. PMID:9824725. Ostrowski, K., Rohde, T., Asp, S., Schjerling, P., and Pedersen, B.K. 1999. Pro- and anti-inflammatory cytokine balance in strenuous exercise in humans. J. Physiol. 515(Pt. 1): 287–291. doi:10.1111/j.1469-7793.1999.287ad.x. PMID:9925898. Pedersen, B.K., and Febbraio, M.A. 2008. Muscle as an endocrine organ: focus on muscle-derived interleukin-6. Physiol. Rev. 88(4): 1379–1406. doi:10.1152/physrev.90100.2007. PMID: 18923185. Reilly, T., and Brooks, G.A. 1986. Exercise and the circadian variation in body temperature measures. Int. J. Sports Med. 7(6): 358–362. doi:10.1055/s-2008-1025792. PMID:3804546. Shirreffs, S.M., Taylor, A.J., Leiper, J.B., and Maughan, R.J. 1996. Post-exercise rehydration in man: effects of volume consumed and drink sodium content. Med. Sci. Sports Exerc. 28(10): 1260–1271. PMID:8897383.

1107 Starkie, R.L., Arkinstall, M.J., Koukoulas, I., Hawley, J.A., and Febbraio, M.A. 2001a. Carbohydrate ingestion attenuates the increase in plasma interleukin-6, but not skeletal muscle interleukin-6 mRNA, during exercise in humans. J. Physiol. 533(Pt. 2): 585–591. doi:10.1111/j.1469-7793.2001.0585a.x. PMID: 11389214. Starkie, R.L., Rolland, J., Angus, D.J., Anderson, M.J., and Febbraio, M.A. 2001b. Circulating monocytes are not the source of elevations in plasma IL-6 and TNF-a levels after prolonged running. Am. J. Physiol. Cell Physiol. 280(4): C769–C774. PMID: 11245592. Steensberg, A., Toft, A.D., Schjerling, P., Halkjaer-Kristensen, J., and Pedersen, B.K. 2001. Plasma interleukin-6 during strenuous exercise: role of epinephrine. Am. J. Physiol. Cell Physiol. 281(3): C1001–C1004. PMID:11502577. Steensberg, A., Keller, C., Starkie, R.L., Osada, T., Febbraio, M.A., and Pedersen, B.K. 2002. IL-6 and TNF-a expression in, and release from, contracting human skeletal muscle. Am. J. Physiol. Endocrinol. Metab. 283(6): E1272–E1278. PMID:12388119. Yeo, W.K., Paton, C.D., Garnham, A.P., Burke, L.M., Carey, A.L., and Hawley, J.A. 2008. Skeletal muscle adaptation and performance responses to once a day versus twice every second day endurance training regimens. J. Appl. Physiol. 105(5): 1462– 1470. doi:10.1152/japplphysiol.90882.2008. PMID:18772325.

Published by NRC Research Press