Exercise and type 2 diabetes: focus on metabolism and inflammation
Abstract
Type 2 diabetes mellitus (T2DM) is associated with metabolic dysregulation and chronic inflammation, and regular exercise may provide a strong stimulus for improving both. In this review, we first discuss the link between inflammation and metabolism. Next, we give an update on the clinical metabolic effects of exercise in T2DM patients with special focus on which parameters to consider for optimizing metabolic improvements. We then discuss the mechanisms whereby exercise exerts its anti-inflammatory and related metabolic effects. Evidence exists that interleukin (IL)-1β is involved in pancreatic β-cell damage, whereas tumor necrosis factor (TNF)-α appears to be a key molecule in peripheral insulin resistance. Mechanistic studies in humans suggest that moderate acute elevations in IL-6, as provoked by exercise, exert direct anti-inflammatory effects by an inhibition of TNF-α and by stimulating IL-1ra (IL-1 receptor antagonist), thereby limiting IL-1β signaling. In addition, IL-6 has direct impact on glucose and lipid metabolism. Moreover, indirect anti-inflammatory effects of exercise may be mediated via improvements in, for example, body composition. While waiting for the outcome of long-term randomized clinical training studies with hard end points, it should be emphasized that physical activity represents a natural strong anti-inflammatory and metabolism-improving strategy with minor side effects.
The prevalence of diabetes mellitus (DM) is rapidly increasing with 382 million adults diagnosed in 2013, increasing to estimated 592 million in 2035.1 Moreover, it is believed that DM is seriously underestimated and that nearly a similar number of undiagnosed compared with the already diagnosed cases exist.2 Different subtypes of DM exist with type 2 diabetes mellitus (T2DM) being far the most prevalent, accounting for estimated 90% of the cases.1
In T2DM subjects, cardiovascular complications are of particular interest as these are mainly causative for the increased morbidity and mortality in T2DM patients compared with healthy subjects.3 For example, the cardiovascular mortality in T2DM subjects without prior myocardial infarction is equal to non-T2DM subjects who have previously had myocardial infarction.4 Metabolic dysregulation and chronic inflammation is considered to be responsible for this increased cardiovascular morbidity and mortality in T2DM patients, and the link between metabolic dysfunction and inflammation is well established.5
Physical activity is a first-line treatment for T2DM and there is consensus that exercise improves metabolic markers in patients with T2DM.6, 7 Moreover, exercise is known to modulate inflammation both acutely and chronically.8 In this review, we discuss the interplay between inflammation and metabolism. Next, we provide an update on the clinical metabolic effects of exercise in T2DM patients with special focus on which parameters to consider for optimizing metabolic improvements. Finally, we will review the mechanisms whereby exercise exerts its anti-inflammatory and related metabolic effects.
Linking inflammation and metabolism
Systemic inflammation, characterized by high levels of circulating inflammatory cytokines, has been causally associated with the development and progression of many chronic diseases, such as T2DM, cardiovascular diseases, depression and dementia.8, 9 Several cross-sectional and prospective studies have described elevated circulating levels of C-reactive protein as well as elevated levels of tumor necrosis factor (TNF)-α, interleukin (IL)-1β, IL-6, IL-1ra (IL-1 receptor antagonist) in T2DM.9
During the past two decades, it has become increasingly clear that chronic inflammation is a crucial factor contributing to the development and progression of chronic non-communicable diseases such as T2DM.10 As diabetic vascular complications are partly mediated by inflammatory processes, targeting inflammation may not only improve glycemic control and slow the progressive β-cell secretory dysfunction but also prevent such comorbidities.9, 11 Evidence exists that IL-1β and TNF-α are key proinflammatory mediators in β-cell damage and insulin resistance.12
Thus, inflammatory changes, including accumulation of macrophages and increased expression of IL-1β, have been documented in T2DM islets.9 It has further been demonstrated that depletion of resident islet macrophages in high-fat-fed transgenic mice with islet amyloid formation can reduce IL-1β expression, improve β-cell insulin secretion and restore glucose tolerance.13 In accordance, therapeutic inhibition of IL-1β ameliorates β-cell dysfunction and glucose homeostasis in individuals with T2DM.14
Whereas IL-1β appears to be a direct cause of β-cell dysfunction, evidence exists that the proinflammatory cytokine TNF-α is a key molecule in insulin resistance. This evidence is based on classical studies in cultured cells and in TNF-α knockout mice, reviewed elsewhere.15, 16 Moreover, in healthy humans, TNF-α was shown to inhibit whole-body insulin-mediated glucose uptake and signal transduction by inhibiting peripheral insulin-stimulated glucose uptake via impaired phosphorylation of Akt substrate 160, a key step in the canonical insulin signaling cascade regulating GLUT4 translocation and glucose uptake.17
The role of IL-6 in metabolism is debated18 and although increased plasma levels of IL-6 have been associated with T2DM in correlational studies,9 mechanistic studies in humans suggest that moderate acute elevations in IL-6 may inhibit TNF-α production and stimulate IL-1ra, thereby limiting IL-1β signaling, reviewed in Knudsen and Pedersen.9
Clinical exercise training studies—metabolic effects
There is solid evidence for the beneficial effects of structured exercise on metabolic markers in patients with T2DM. Studies have consistently showed exercise-induced improvements in glycemic control, insulin sensitivity, body composition and cardiorespiratory fitness, with some studies also supporting beneficial effects on lipid status and blood pressure.19, 20, 21, 22, 23 These factors are all considered to be important for preventing cardiovascular complications in T2DM, and observational studies support that exercise reduces cardiovascular events and mortality.3, 24
Traditionally, continuous endurance exercise with moderate intensity has been recommended for patients with T2DM. Although this approach definitely has positive effects, the best way to prescribe exercise for optimizing the metabolic effects is unknown. Many different aspects need consideration and a uniform model for all patients does not exist. Here we will discuss some of the aspects to consider when recommending exercise for T2DM patients.
Training modality
Most exercise training studies performed in T2DM patients have evaluated endurance training, and, as said, endurance training is typically recommended for T2DM patients. Interestingly, in studies where resistance training has been evaluated, it has been shown that this has effects comparable to endurance training on body composition, glycemic control and insulin sensitivity and -signaling,25, 26 why tradition, convenience and feasibility are the most probable reasons for these recommendations. Convincingly, the Health Benefits of Aerobic and Resistance Training in individuals with type 2 diabetes (HART-D) and Diabetes Aerobic and Resistance Exercise (DARE) studies demonstrated that whereas both endurance and resistance training each have positive effects, the combination of endurance and resistance training was superior for improving glycemic control, cardiorespiratory fitness and body composition.21, 22 This has led to updated recommendations from both the European and the American diabetes associations, which now both recommend the combination of endurance and resistance exercise for optimal T2DM care.6, 7
Training supervision
Most physical activity interventions in T2DM subjects have used fully or partially supervised training programs. Although the evidence for the effectiveness of such programs in terms of improving glycemic control and other metabolic risk factors is convincing,20 the evidence base for non- or minimally supervised interventions is smaller. Although some studies have reported increased physical activity and fitness level as a result of training counseling,27 and comparable effects of counseling and supervised training on fitness level,28 no studies have, to our knowledge, compared the effect of a supervised vs a non-supervised but otherwise similar training regime in T2DM subjects. A meta-analysis concluded that whereas supervised training was efficient, training advice alone was not sufficient to improve glycemic control in T2DM subjects.19 This may be due to a lower adherence to non-supervised training interventions,29 and to the fact that exercise intensity typically is lower in T2DM subjects who do non-supervised training.30 When bearing the large number of prevalent T2DM cases in mind, a challenge remains in developing non- or minimally supervised exercise interventions that both have high adherence and improve metabolic risk factors.
Training volume and intensity
A dose–response relationship between training volume and metabolic effects is generally accepted.19 However, some studies with fairly large training volumes show minor effects whereas other studies, with small training volumes, show substantial effects and studies that have compared smaller with larger training volumes have found either no or only trivial benefits of larger training volumes.31 Thus, a ceiling effect of training volume does clearly exist and training volume alone is potentially not as important as previously thought.
Traditionally, high-intensity training has not been recommended for subjects with T2DM, partly due to a fear of inducing complications, and partly as it was believed that moderate-intensity training was sufficient to elicit beneficial changes.32 Moreover, it has been found that high-intensity training reduces adherence and, as such, training volume compared with lower-intensity training, thereby potentially limiting the effectiveness of high-intensity training.33 Indeed, moderate-intensity training has shown improvements in glycemic control and insulin sensitivity comparable to higher-intensity training matched for total amount of energy expenditure in T2DM patients,34 or has found only trivial superiority of increased intensity.35 One study has even reported that a single endurance exercise bout with lower intensity reduces hyperglycemia during the following 24 h more than an exercise bout with higher intensity in T2DM patients.36 Conversely, other studies have found that training intensity is the major determinant for improving metabolic risk factors. For example, DiPietro et al.37 found that endurance training performed at 80% of peak oxygen consumption (VO2peak) was superior to an energy expenditure-matched endurance training program performed at 65% of VO2peak with regards to improving insulin sensitivity, despite no differential outcome in body composition. As such, the impact of training intensity on metabolic risk factors in T2DM patients is still unclear.
Interval training
Interval training regimes have gained increasing attention during the past years. A hallmark study was performed by Tjonna et al.38 who compared aerobic interval with continuous training over 12 weeks in subjects with the metabolic syndrome. They showed that interval training was superior in improving almost all aspects of the metabolic syndrome, despite training volumes were comparable between the two groups. Following this, an increasing amount of interval-training programs have been tested in subjects with metabolic disease, including T2DM. In general, the interval-training regimes are superior in improving metabolic parameters compared with continuous training regimes, or they show improvements equal to continuous regimes but with a lower training volume.39, 40 In that respect, we have shown that aerobic interval walking training is superior to continuous walking training matched for volume, energy expenditure and mean intensity, for improving body composition, physical fitness, glycemic control and peripheral glucose disposal in patients with T2DM both acutely and after a long-term training intervention.41, 42, 43
Thus, interval training seems promising as a tool for optimizing metabolic effects in T2DM patients, also when performed very briefly.44 The long-term impact of interval-training modalities is however unknown. This has to be taken into consideration given that free-living interval-training adherence may be low.45
The anti-inflammatory effects of exercise
In both longitudinal and cross-sectional studies it has been described that regular physical activity, also without weight loss, diminishes systemic chronic inflammation, although the prescription of exercise as a potential anti-inflammatory tool is a relatively new concept.9 The protective effect of exercise against diseases associated with chronic inflammation may to some extent be ascribed to an anti-inflammatory effect of regular exercise, which could be mediated either by an acute anti-inflammatory effect with each bout of exercise, or by long-term anti-inflammatory effects via exercise-induced improvements in body composition, physical capacity, comorbidities and cardiovascular risk factors.
The direct anti-inflammatory effect of exercise
It has been suggested that contracting skeletal muscle produces, expresses and releases cytokines and other peptides, which exert autocrine, paracrine or endocrine effects, and, as such, should be classified as myokines. Today, hundreds of secreted peptides have been established as part of the muscle secretome.46, 47, 48 This finding provides a conceptual basis and a new paradigm for understanding how muscles communicate with other organs, such as adipose tissue, liver, pancreas, bones and brain, thus potentially affecting systemic hormones and the inflammatory milieu.
As previously reviewed16 the myokines myostatin, leukemia inhibitory factor, IL-4, IL-6, IL-7 and IL-15 appear to be involved in muscle hypertrophy and myogenesis; brain-derived neurotropic factor and IL-6 stimulate fat oxidation; and IL-6 and IL-15 may mediate lipolysis. IL-6 also appears to have systemic effects on the liver, adipose tissue and the immune system, and mediates cross talk between intestinal L cells and pancreatic islets. Insulin-like growth factor-1, fibroblast growth factor-2 and transforming growth factor-β are osteogenic factors, follistatin-related protein 1 may improve endothelial function of the vascular system and irisin has been shown to drive a brown-fat-like phenotype. In addition, meteorin like has been identified as a myokine that regulates immune–adipose interactions to increase beige fat thermogenesis.16
Interleukin-6 is the myokine prototype and although IL-6 is generally viewed as a proinflammatory cytokine, evidence exists that muscle-derived IL-6 has anti-inflammatory properties.49 It has been suggested that both the upstream and downstream signaling pathways for IL-6 differ markedly between myocytes and macrophages.50
In macrophages, IL-6 signaling is dependent upon activation of the nuclear factor-KappaB signaling pathway. However, it appears that intramuscular IL-6 expression is regulated by a network of signaling cascades that is likely to involve a cross talk between the Ca2+/nuclear factor of activated T cells and glycogen/p38 mitogen-activated protein kinases pathways.51 Therefore, it has been suggested that when IL-6 is signaling in monocytes or macrophages, it creates a proinflammatory response, whereas contraction-induced activation of IL-6 and its signaling in muscle cells is totally independent of a preceding TNF response or nuclear factor-KappaB.50
Some myokines can induce an anti-inflammatory response with each bout of exercise. For example during exercise, IL-6 is the first detectable cytokine released into the blood from the contracting skeletal muscle fibers. IL-6 increases with exercise and contributes to a marked increase in circulating levels of IL-6. Both the amount of muscle mass engaged in the exercise and the duration of exercise matter.50
Given that interval training appears to be superior to continuous training in patients with T2DM,41, 43 it is of interest that IL-6 is also dependent on the exercise intensity.52
Of note, in contrast to the cytokine response elicited by sepsis, muscle-derived IL-6 occurs without a preceding increase in TNF-α. In relation to physical activity, IL-6 induces a subsequent increase in the production of IL-1ra and IL-10 by blood mononuclear cells, thus stimulating the occurrence of anti-inflammatory cytokines.50
A model of ‘low-grade inflammation’ was previously established in our laboratory. A very low dose of Escherichia coli endotoxin was administered to healthy subjects, who were randomized to either rest or exercise before endotoxin administration. In resting subjects, endotoxin induced a two- to three-fold increase in circulating levels of TNF-α. In contrast, when participants performed 3 h of ergometer cycling and received the endotoxin bolus at 2.5 h, the TNF-α response was blunted, suggesting that acute exercise may inhibit TNF-α production. The effects of exercise could be mimicked by an infusion of IL-6, which suppressed the endotoxin-induced TNF-α production.53
Thus, an acute bout of exercise induces a strong anti-inflammatory effect, which at least in part is mediated by IL-6, which has suppressive effects on TNF-α and stimulates the occurrence of IL-1ra and IL-10. Adding to the anti-inflammatory effects of IL-6, the Brüning group has identified that IL-6 signaling is an important determinant of the alternative activation of macrophages. It is suggested that during inflammatory conditions (that is, in the presence of a greater abundance of free fatty acids and/or bacterial lipopolysaccharide), IL-6 limits the expression of genes encoding inflammatory cytokines and augments the responsiveness of macrophages to IL-4 thereby counterbalance the typical shift of macrophage populations toward a proinflammatory M1 phenotype and ultimately diminishes inflammation and the associated resistance to insulin.54
The anti-inflammatory effect of exercise is at least partly mediated via IL-6 and may have several beneficial metabolic effects. Given the role of TNF-α in insulin resistance, exercise may enhance insulin sensitivity at least in part by its suppressive effect on TNF-α. In parallel, given the involvement of IL-1β in pancreatic β-cell damage, the finding that exercise provokes an increase in circulating IL-1ra may contribute to protect against IL-1-mediated destruction of β-cells.
However, IL-6 has also direct metabolic effects as it stimulates β-cell proliferation, prevents apoptosis caused by metabolic stress and regulates β-cell mass in vivo.55 Thus, it appears that exercise-induced IL-6 production may be involved in the expansion of pancreatic β-cell mass that is needed for functional β-cell compensation when an increased metabolic demand is present.55 It has furthermore been shown that elevated IL-6 concentrations in response to exercise stimulate glucagon-like peptide-1 secretion from intestinal L cells and pancreatic β-cells, improving insulin secretion and glycemia, suggesting that IL-6 is involved in an endocrine loop implicating IL-6 in a beneficial regulation of insulin secretion, which may be useful in T2DM.56
An acute challenge with IL-6 has also been shown to enhance peripheral glucose uptake and to induce lipid oxidation via a mechanism that includes activation of adenosine monophosphate-activated protein kinase. Finally, IL-6 is an essential regulator of satellite cell-mediated skeletal muscle hypertrophy, reviewed in Munoz-Canoves et al.57
Thus, although the role of IL-6 in metabolism is debated,18 mechanistic studies in humans suggest that moderate acute elevations in IL-6, as provoked by exercise, exert anti-inflammatory effects by an inhibition of TNF-α and by stimulating IL-1ra, thereby limiting IL-1β signaling. In this context, the possibility exists that with regular exercise, the anti-inflammatory effects of each acute bout of exercise will protect against chronic systemic low-grade inflammation.
The indirect anti-inflammatory effect of exercise
Regular exercise may also mediate anti-inflammatory effects by protecting against accumulation of visceral fat. It seems well documented that independent of body mass index, physical inactivity is a cause of central obesity and evidence exists that visceral fat is more inflamed than subcutaneous fat.58
Moreover, both physical inactivity8 and abdominal adiposity50 are associated with persistent, systemic low-grade inflammation and a direct link between physical inactivity and visceral fat has been established. In a study in which healthy men reduced their daily activity levels from >10 000 to <1500 ‘steps’ for 14 days, intra-abdominal fat mass increased without a change in total fat mass.59
It is possible that myokines may also have a role in limiting accumulation of visceral fat. IL-15 appears to be involved in the regulation of visceral fat as overexpression of IL-15 has been shown to protect mice from obesity, especially with regard to accumulation of visceral fat.60 Moreover, muscular IL-15 is increased in mice following weeks of running and a marked and robust upregulation of IL-15 protein in muscle is found after endurance training in humans.60
As described above, we suggest that both IL-15 and IL-6 may have important roles in lipid metabolism. However, other myokines may be involved in the regulation of adipose tissue mass. In this regard, brain-derived neurotropic factor has been suggested as a myokine that works in an autocrine or paracrine fashion with effects on peripheral metabolism, including fat oxidation with a subsequent effect on the size of adipose tissue, reviewed in Pedersen.61 Patients with type 2 diabetes frequently suffer from muscle waste due to decreased physical activity. A large muscle mass may protect against accumulation of visceral fat and several myokines, including myostatin, leukemia inhibitory factor, IL-4, IL-6, IL-7 and IL-15 appear to be involved in the regulation of skeletal muscle growth and maintenance,61 and the myokines meteorin like62 and irisin63 are involved in browning of adipose tissue.
Moreover, IL-6 and other myokines, such as IL-15 and follistatin-related protein 1, mediate long-term exercise-induced improvements in cardiovascular risk factors (for example, fat distribution and endothelial function), thus potentially having indirect anti-inflammatory effects, reviewed in Pedersen and Febbraio.63
We suggest that exercise may induce anti-inflammatory effects by limiting the amount of visceral fat accumulation. We have previously proposed that a state of chronic inflammation may be accompanied by anemia, tiredness and muscle waste, which together with other comorbidities will lead to further deconditioning and a worsening of the chronic inflammation, which again will negatively affect cardiovascular performance and physical activity, thus establishing a ‘vicious circle of chronic inflammation and physical inactivity’.16
Conclusion and perspective
As discussed above, there is good evidence to suggest that regular exercise improves metabolic markers and has anti-inflammatory properties. Exercise may mediate its anti-inflammatory effects directly with each bout of exercise or indirectly via long-term metabolic improvements in, for example, body composition, physical fitness, lipid and glucose metabolism. Thus, although only limited research has been carried out in this area, the various factors discussed above that advantageously can be taken into account for optimizing metabolic outcome of exercise, are probably also important factors for optimizing the anti-inflammatory properties of exercise.
These anti-inflammatory properties of exercise might be a common denominator for the protective effect of exercise on chronic diseases such as T2DM, various cancers, cardiovascular diseases, depression and dementia, given the previously proposed ‘diseasome of physical inactivity’.8 This claims that there is a 2–3 times increased risk of each of the other diseases in the diseasome if a person already has one of the diseases.
Although the beneficial metabolic and anti-inflammatory effects of exercise, as reviewed in the current paper, are well established, the long-term outcome of exercise interventions in T2DM patients is largely unknown. Cohort studies indicate that increased levels of physical activity and increased physical fitness level do improve hard end points,3, 24, 64 but there is, to our knowledge, no randomized controlled trials that have evaluated the impact of physical activity on hard end points in T2DM patients. Although huge obstacles (blinding issues, compliance issues, long follow-up time, large sample size) complicate such an randomized controlled trials, it must be performed, to fully elucidate the potential of physical activity for the treatment of T2DM. However, while waiting for the outcome of such a study, it should be emphasized that physical activity represents a natural strong anti-inflammatory and metabolism-improving strategy with minor side effects.
ACKNOWLEDGEMENTS
The Centre for Physical Activity Research (CFAS) is supported by a grant from Trygfonden. CIM/CFAS is a member of DD2—the Danish Center for Strategic Research in Type 2 Diabetes (the Danish Council for Strategic Research, grant no. 09-067009 and 09-075724).
CONFLICT OF INTEREST
The authors declare no conflict of interest.
