Objective: Inflammatory mediators, such as tumour necrosis factor alpha (TNFα), may contribute to delayed-onset muscle soreness. The effect of neutralising TNFα with etanercept, a soluble TNFα receptor, on delayed-onset muscle soreness (DOMS) induced in the quadriceps muscle was analysed.
Design: On two separate occasions at least 6 weeks apart, etanercept 25 mg or vehicle was given subcutaneously 1 hour before unaccustomed exercise to 12 healthy men in a randomised double-blind cross-over format. To induce DOMS, subjects completed 4 sets of 15 repetitions at 80% of their one-repetition maximum (1RM), using a 45° inclined leg press. Muscle soreness was assessed using a 100-mm visual analogue scale (VAS), and pressure pain threshold (PPT) on the thigh before and 24, 48 and 72 hours after exercise. Changes in the subject’s muscle strength were detected by reassessing the subject’s 1RM 24, 48 and 72 hours after exercise.
Results: Muscle strength decreased 24 and 48 hours after exercise regardless of agent administered (analysis of variance, p<0.001). At 72 hours after exercise, muscle strength was significantly greater (p<0.01) after etanercept than after placebo. The exercise protocol induced significant DOMS for up to 72 hours, as reflected by reduced PPT and increased VAS scores (p<0.001). Etanercept had no effect on PPT or VAS.
Conclusion: TNFα does not affect muscle soreness associated with unaccustomed exercise, but may improve the recovery of muscle function.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Delayed-onset muscle soreness (DOMS) results in muscle pain after eccentric muscle contraction or unaccustomed exercise,1 2 and is perceived as muscle soreness during passive stretch and as muscle contraction. Mechanical hyperalgesia, an increased sensitivity to painful pressure applied to the affected muscle, is also present.3 DOMS is most painful 24–48 hours after exercise, and resolves 5–7 days after exercise.1 4 The exact mechanism of the muscle soreness associated with DOMS is not yet fully understood. Myofibrillar microtrauma5 and associated inflammation6 are the most likely peripheral mechanism initiating DOMS, but the muscle soreness is intractable against cyclo-oxygenase inhibitors, which inhibit prostaglandin synthesis.7 8 It is possible that if muscle inflammation does occur, chemical mediators upstream of prostaglandin synthesis, such as cytokines, would be released into the muscle, contributing to the hyperalgesia.
Cytokines are key contributors to the hyperalgesia induced by cutaneous tissue injury and inflammation.9 In cutaneous tissue, tumour necrosis factor-alpha (TNFα) is a key mediator and initiator of the cytokine cascade.10 TNFα sensitises cutaneous nociceptors,11 12 but also stimulates the release of other mediators, such as interleukin (IL)-1β, IL-6 and prostaglandins, which sensitise cutaneous nociceptors. Whether TNFα contributes to pain induced during muscle inflammation, in particular exercise-induced muscle inflammation, is not yet known.
Physical exercise has been shown to influence the production of cytokines.4 13 14 Increases in IL-1β and TNFα plasma concentrations are dependent on the intensity, duration and type of exercise, but the concentration of these cytokines in muscle after injury and particularly after eccentric exercise is less clear. In addition, these cytokines are pleiotropic, with TNFα and IL-6 playing a role in glucose metabolism, protein synthesis and protein degradation in muscle.6 15 Although IL-6 is released during muscle contraction, it is unlikely to contribute to muscle soreness, as an increased concentration of IL-6 during concentric muscle contraction occurs in the absence of muscle pain.16 17 Therefore, although cytokine concentrations increase after eccentric exercise, it remains to be elucidated whether cytokines released in muscle after eccentric exercise contribute to the soreness associated with DOMS. Therefore the aim of our study was to identify whether inhibiting the action of TNFα, using a soluble TNFα receptor, etanercept, attenuates DOMS.
MATERIALS AND METHODS
The experimental protocol was approved by the University of the Witwatersrand Committee for Research on Human Subjects (M050311) and all subjects gave written informed consent for participation.
In total, 12 healthy men (mean (SD) age 24 (3) years; height 1.78 (0.06) m; weight 72.5 (7.5) kg, who were recreationally active, participated in the study. Subjects were asked to refrain from exercise, massage, electrotherapy techniques, hot packs, and caffeine and alcohol ingestion, for 4 days before and for the duration of each trial. Any subject taking pain medication was excluded from the study.
Subjects reported to our exercise laboratory on two separate occasions at least 6 weeks apart (range 6 to 14 weeks). Each trial lasted 4 days, and subjects completed each day of the trial at the same time of day. In randomised double-blind crossover conditions, subjects were injected with either a soluble TNFα receptor (etanercept) or vehicle (sterile water). Etanercept 25 mg (Enbrel, Wyeth, Johannesburg, South Africa), was administered once as a 1-ml subcutaneous injection, 1 hour before the exercise session. The time to peak concentration after administration of 25 mg etanercept in serum is 69 (34) hours and etanercept has a half-life of 102 (30) hours.20 The stated dose of 25 mg etanercept is the manufacturer’s recommended dose for the treatment of rheumatoid arthritis.21
On the first day of each trial, we determined each subject’s one-repetition maximum (1RM) for the leg, using a 45° incline leg-press machine (Cardio Genesis Fitness Systems, South Africa). The 1RM is the mass (kg) that a subject is able to leg-press once only, and is an indication of the muscle strength of the hamstrings, quadriceps and gluteal muscle groups. In order to induce lower limb muscle damage, subjects completed four sets of 15 repetitions at 80% of their 1RM, using the leg-press machine, or until voluntary muscle fatigue. This unaccustomed exercise has a component of both concentric and eccentric muscle contraction, particularly in the quadriceps muscle group. Each subject’s leg-press 1RM was retested 24, 48 and 72 hours after the initial exercise session.
Immediately before the exercise session and 24, 48 and 72 hours after the exercise, 5 ml of blood was collected from the brachial vein by venepuncture and placed into EDTA collecting tubes. Blood was centrifuged at 2000 g at 4°C for 10 minutes, then the plasma was removed and stored at −80°C until analysis. Plasma creatine kinase (CK) concentration was measured using a commercially available calorimetric assay kit (Roche Diagnostics, Johannesburg, South Africa).
Immediately before the exercise session, and 24, 48 and 72 hours after exercise, we applied a pressure algometer (Somedic, AB, Sweden) with a 1 cm2 probe to three points on the right thigh (1) midway between the hip and the knee on the quadriceps femoris muscle (5 cm above the superior border of the patella), (2) midway between the medial femoral epicondyle and the patella, and (3) on the same point laterally. We applied the algometer at each point with a uniformly increasing pressure, until the subject indicated that the pressure stimulus had become painful. Two measurements taken 30 seconds apart were performed on each site, and pressure pain threshold (PPT, kPa) was recorded as a mean of all six measurements.
In order to quantify the intensity of quadriceps muscle pain, a 100-mm visual analogue scale (VAS) anchored with “no pain” on the left and “worst pain ever experienced” on the right was used after subjects performed a simple squat (body mass only). On the VAS, each subject marked the intensity of pain experienced after the squat. This procedure was carried out before, and then 24, 48 and 72 hours after the exercise session.
Muscle biopsies from the left vastus lateralis muscle were obtained from 5 of the 12 subjects. The muscle biopsies were taken immediately before exercise in one exercise session to obtain a pre-exercise sample, and 2 and 24 hours after both exercise sessions. A local anaesthetic (0.5%, Macaine, Adcock Ingram Limited, South Africa) was injected into the dermis and subcutaneous tissue over the left vastus lateralis muscle, 10 minutes before the muscle tissue was excised. A 2-cm incision was made through the skin and underlying fascia using a surgical blade. A biopsy needle (Stille Surgical, AB, Sweden) with a diameter of 6 mm was inserted into the muscle tissue, suction was applied and approximately 100 mg of muscle tissue was obtained. The muscle tissue was weighed, fresh-frozen in liquid nitrogen and stored at −80°C for later analysis of cytokine concentrations.
Muscle TNFα assay
Each muscle tissue section was weighed and placed in a glass tissue homogeniser, together with 250 μl of cell lysing solution (Bio-Plex Cell Lysis Kit, Bio-Rad Laboratories, California, USA), ground and then frozen at −80°C. After thawing, samples were broken down in a sonicator (Cole-Palmer Instrument Company, Chicago, USA) on ice for 3 minutes. Samples were then spun in a centrifuge at 4000 g for 4 minutes at 4°C. The supernatant was collected, and the protein concentration determined using a Bradford protein assay (Bio-Rad Laboratories).
TNFα concentrations were measured using a bead-array analysis (Bio-Rad Laboratories). The pre-mixed beads, coated with target antibodies (50 μl) were added to a 96-well plate. After washing with wash dilution buffer, 50 μl of standard or undiluted sample was added, with each sample measured in triplicate. The plate was incubated in the dark for 60 minutes at room temperature (RT). The plate was washed, detection antibodies (25 μl) were added and the plate was incubated in the dark for 30 minutes (RT). After washing the plate again, 50 μl of streptavidin–phycoerythrin was added and incubated in the dark for 10 minutes at 20 g (RT). The beads then were resuspended in 125 μl of Bio-Plex assay buffer, and the plate was read using the Bio-Plex suspension array system.
Data are shown as mean (SD) unless otherwise stated. VAS data were normalised using the arcsine transform and recorded as mean (SD). Two-way repeated measures analysis of variance, with time and agent administered as the main effects, was used to compare the 1RM, PPT and VAS. Tukey post hoc tests were used where significant differences were found. We used the non-parametric Friedman test, with Dunn multiple comparisons, to analyse for differences in plasma CK concentration and TNFα concentration. Significance was set at p<0.05.
Muscle TNFα concentration
Table 1 shows TNFα in the muscle (pg/mg/protein) before exercise, and 2 and 24 hours after administration of placebo or etanercept. The detection limit of the assay for TNFα was 6.32 pg/ml, or 0.331 pg/mg protein. No TNFα was detected in muscle 2 hours after exercise in the subjects receiving etanercept, indicating that the dose of etanercept was sufficient to neutralise the muscle TNFα.
Figure 1 shows the 1RM of the quadriceps muscle before and 24, 48 and 72 hours after the exercise session, in the groups receiving placebo or etanercept. The mean (SD) 1RM before exercise was 317 (6) kg, with no significant difference in muscle strength before exercise between the two trials (p = 0.91). There was a significant time effect (F3,63 = 16.74, p<0.0001), with 1RM at 24 and 48 hours being lower than at 72 hours and before exercise. There was no significant agent effect (F1,21 = 0.84, p = 0.369), but there was a significant interaction (F3,63 = 3.24, p<0.05). Muscle strength after exercise and etanercept was significantly lower 24 hours after than before exercise, but by 48 hours, muscle strength had returned to pre-exercise values. After placebo, muscle strength was significantly lower 24 and 48 hours after exercise compared with the pre-exercise value.
Figure 2 shows the plasma CK concentration (median and interquartile range) before and 24, 48 and 72 hours after exercise. There was no significant difference in plasma CK concentration before exercise between the group receiving placebo and the group receiving etanercept (Friedman statistic 13.9, p = 0.053). After placebo, there was a significant increase in plasma CK concentration at 24 hours (median 291 μmol/l, 95% CI 24 to 1111 μmol/l) compared with pre-exercise concentration (63 μmol/l, 137 to 666 μmol/l, p<0.01). However, after etanercept, there was no significant change in plasma CK concentrations at any time after the exercise compared with pre-exercise concentrations (83 μmol/l, −137 to 1043 μmol/l).
Delayed-onset muscle soreness
Figure 3 shows the VAS scores of the subject’s perceived soreness in the quadriceps muscle after exercise. Subjects reported significantly higher VAS at 24, 48 and 72 hours compared with pre-exercise VAS scores (F3,66 = 59.40, p<0.0001), regardless of the agent administered. The VAS scores were significantly higher at 48 hours compared with those at 24 hours (p<0.05) and 72 hours (p<0.001) after exercise. Etanercept did not significantly change the intensity of muscle pain (F1,22 = 0.78, p = 0.39) at any time compared with that after placebo.
Pressure pain threshold
Figure 4 shows the PPT of the quadriceps muscle before exercise, and 24, 48 and 72 hours after exercise, after placebo or etanercept. There was no significant difference between the subject’s pre-exercise PPT of trial 1 (455 (170) kPa) and trial 2 (515 (194) kPa, p = 0.93). The PPT, 24, 48 and 72 hours after exercise was significantly lower than pre-exercise PPT values (F3,66 = 24.63, p<0.0001). PPT was also significantly lower 24 hours and 48 hours after exercise compared with 72 hours after exercise. Etanercept had no significant effect on PPT at any time compared with placebo administration (F1,22 = 0.01, p = 0.93).
We successfully induced DOMS in the quadriceps muscle of subjects completing a bout of unaccustomed exercise on a leg-press machine. According to the VAS scores, the subjects experienced mild pain at 24 hours and moderate pain at 48 hours.22 All the subjects’ quadriceps muscles were significantly more sensitive to pressure 24 and 48 hours after a bout of unaccustomed exercise on the leg-press machine compared with pre-exercise values and 72 hours after exercise, regardless of the agent administered. The time course of the reduced PPT correlated with the time course of the subjects’ perceived soreness as reflected on the VAS. Despite neutralisation of TNFα in muscle having no effect on the development of muscle soreness, administration of etanercept did improve the functional recovery of muscle strength. That is, when subjects received etanercept, they regained normal muscle strength, as tested using 1RM, 72 hours after the exercise, whereas in the group given placebo, muscle strength was still significantly reduced 72 hours after exercise compared with pre-exercise concentrations. Therefore, although TNFα does not appear to be instrumental in mediating the pain component of DOMS, it may contribute to tissue repair and hence functional recovery.
In our study, we found TNFα in skeletal muscle at rest, before exercise. Although our values are lower than those reported in healthy male volunteers in previous studies,23 24 etanercept sufficiently neutralised the muscle TNFα protein. These data indicate that increased TNFα synthesis after unaccustomed exercise is not a primary mediator of pain, but rather that this basal production of TNFα in muscle compromises the functional recovery of muscle after exercise. Further evidence to suggest the effect of TNFα on muscle function is the lack of increase in CK, a marker of muscle injury,1 at 24 hours after exercise in those given etanercept compared with those given placebo. It is possible that muscle TNFα concentration does not increase in exercise-induced muscle damage, as our results concur with a previous study that found no change in TNFα gene expression in muscle immediately and 2 hours after a marathon run.25
Although TNFα concentration in muscle was not raised in our study, production of TNFα has been found after muscle injury, such as a freeze or crush injury, in rats.19 26 However, in both these models, the injury was severe, and no correlation was made between the increased TNFα concentration in muscle after injury and the presence of hyperalgesia. In contrast, carrageenan, an inflammatory agent, when injected into the muscle of rats, does not result in increased muscle TNFα concentration up to 24 hours after the injection.27 Therefore, it may be that TNFα is increased in muscle in more severe injuries, such as a freeze or crush injury, but not in mild injury, such as DOMS or carrageenan-induced inflammation.
What is already known on this topic
The exact mechanism of the muscle soreness associated with delayed-onset muscle soreness is not yet fully understood, but myofibrillar microtrauma and associated inflammation are the most likely peripheral mechanism initiating DOMS.
However, the muscle soreness is not affected by cyclo-oxygenase inhibitors, which inhibit prostaglandin synthesis, and so decrease inflammation.
What this study adds
Tunour necrosis factor-α, a pro-inflammatory cytokine that indirectly mediates the synthesis of prostaglandins, does not contribute to the muscle soreness during DOMS, inferring that DOMS may not be an inflammatory-mediated pain.
We have shown that the mechanism behind the muscle soreness induced during DOMS is independent of the muscle strength impairment during DOMS, as TNFα attenuated the muscle strength deficit but not the pain.
Exogenous TNFα injected into the muscle of rats induces muscle hyperalgesia.3 However, the dose of TNFα in that study was pharmacological rather than physiological. Therefore, TNFα may produce hyperalgesia in muscle but not at physiological doses. In our study, etanercept did not attenuate the subjective pain, as measured by VAS, or hyperalgesia, as measured by PPT, after unaccustomed exercise. It is possible that the dose we used was too low to diminish the muscle pain generated by the unaccustomed exercise. However, etanercept, at the same dose we used, is used successfully to treat inflammation and hyperalgesia associated with rheumatoid arthritis (a more intense pain than that of DOMS) and other inflammatory-mediated musculoskeletal diseases.28–30 Therefore, a 25 mg dose of etanercept effectively alleviates mild to moderate cytokine-induced pain, but does not appear to attenuate DOMS.
TNFα is not likely to sensitise muscle nociceptors during DOMS, but it exerts other effects in muscle such as inhibiting the differentiation of myoblasts and promotes protein degradation.18 31–33 Therefore, an increase in TNFα in muscle results in muscle catabolism and muscle wasting.34 TNFα may play a direct role in regulating muscle function and homeostasis, but may also induce insulin resistance via down-regulation of expression of the glucose transporter GLUT-4, and may suppress protein synthesis via ubiquitin and nuclear factor kappa beta pathways.6 35 36 It is proposed that TNFα release is suppressed in normal muscle, but during certain diseases, such as diabetes and rheumatoid arthritis, the normal regulatory mechanisms that suppress TNFα are down-regulated, allowing TNFα release to increase, producing catabolic effects on muscle.37
TNFα is also involved in muscle repair and in the recovery of muscle function.6 19 However, whether an increased concentration of muscle TNFα exerts protein degradation or muscle repair seems to be dependent on the type and severity of the injury. After a freeze injury in TNFα receptor knock-out mice and wild-type mice given neutralising TNFα antibodies, poorer functional muscle recovery was noted compared with that of injured wild-type mice.19 However, the effect of TNFα in repair was only apparent 13 days after injury. In our study, we found that neutralising the TNFα produced significant improvement in muscle function 3 days after injury. Nonetheless, our study confirms a previous study that neutralising TNFα in the acute phase of injury has no effect on muscle strength initially but may improve functional muscle recovery later.19 It is possible that in our study, where the muscle injury was less severe than the freeze injury in the mice, etanercept administration prevented the proteolytic effects of TNFα on muscle, resulting in improved muscle strength 72 hours after the eccentric exercise. Future studies should evaluate the effect of neutralising TNFα on functional muscle recovery after exercise-induced muscle injury.
In conclusion, our study has shown that the pain during DOMS was not attenuated by inhibiting TNFα, inferring that DOMS may not be an inflammatory-mediated pain. We have shown, however, that the mechanism behind the strength impairment during DOMS is independent of the hyperalgesia induced during DOMS, as TNFα attenuated the muscle strength deficit but not the pain.
This research was supported by the Thuthuka Programme, National Research Foundation, South Africa and the Iris Ellen Hodges Trust, University of the Witwatersrand. We are grateful to M Badenhorst for assistance and to the participants who gave of their time.
Competing interests: None.