The Role of Exercise Training on Low-Grade Systemic Inflammation in Adults with Overweight and Obesity: A Systematic Review

Low-grade systemic inflammation leads to critical alterations of several tissues and organs that can promote the appearance of non-communicable diseases, a risk that is increased in adults with obesity. Exercise training may counteract low-grade systemic inflammation, but there is a lack of consensus on how cytokines are modulated by training in adults with obesity. This study aimed of examining the effects of exercise training on circulating pro- and anti-inflammatory cytokines in adults with overweight and obesity, and whether exercise-induced fat mass reduction could mediate that effect. The search was conducted on Medline (Pubmed), SPORTDiscus and Web of Science databases from January 1998 to August 2021, using keywords pertaining to inflammation, exercise, and obesity. A total of 27 studies were selected, in which the circulating concentration levels of cytokines were analyzed. Endurance training (ET) decreased circulating CRP, IL-6 and TNF-α levels. TNF-α was reduced after resistance and concurrent training (CT), while IL-10 increased after resistance training (RT). Changes in IL-10 and CRP coincided with fat mass reduction, while decreased TNF-α levels were concomitant with changes in IL-6 and IL-10. Exercise training may reduce systemic low-grade inflammation profile in adults with overweight and obesity.


Introduction
Physical inactivity is a serious health problem [1] that causes low-grade systemic inflammation and visceral fat mass accumulation [2]. These two intrinsically linked processes are key factors that influence cardio-metabolic, neurodegenerative, and immune disease development [3]. In people with obesity, low-grade systemic inflammation is related to skeletal muscle mass loss and reduced muscle strength production, metabolic and functional limitations observed in people with multiple comorbidities [4], osteoarthritis [5], or who are hospitalized [6]. Unfortunately, low-grade systemic inflammation still generates more questions than answers and whether exercise training could modulate the chronic inflammatory state of a population with excessive fat mass accumulation or not is one of the critical unsolved questions [7].
Visceral fat mass accumulation stimulates the activation of the innate immunity, which promotes a local response to cellular damage facilitated by increased blood flow, immune cell infiltration (i.e., macrophages) and inflammatory mediator production to repair the damaged tissue, as well as neutralizing any toxic agent produced [8,9]. However, when the inflammatory state persists, adipocytes and immune cells release pro-inflammatory cytokines into the circulation, such as C-reactive protein (CRP), interleukin-6 (IL-6), and The present systematic review was developed using guidance from the Preferred Reporting Items for Systematic Reviews and Meta-Analysis protocols (PRISMA-P) [20].

Search Strategy, Information Sources and Data Management
Literature was explored using Medline (PubMed), SPORTDiscus and Web of Science, including articles published between January 1998 and August 2021. The search strategy (inflammation AND exercise AND obesity) is presented in Supplementary Table S1, together with an example of the search engine used. A list of reference of included RCTs was elaborated, and an extra scan was performed for additional RCTs. Study authors were contacted when unpublished studies or additional data were required. The identification, screening and abstraction was performed by two authors working independently, while a third author evaluated the quality of the studies, contact with other authors, and explored non-English articles.

Study Selection
After a pilot-testing eligibility criteria for citations, abstract and full-text articles, screening was conducted by two authors working independently, and when any discrepancy regarding eligibility took place, a third author was involved to solve the conflict. The interventions were coded independently, classified into the following broad categories: cytokines, type of exercise, and exercise-induced fat mass loss. Study quality was evaluated independently by two authors; a third authors was not required for this task, since no conflict appears between authors.
Articles were included when they involved physically inactive (<150 min/week), overweight or obese (BMI > 25 kg/m 2 ) adult populations (18 to 65 years), who participated in an exercise training program. We examined studies including several biomarkers (CRP, IL-1β, IL-1ra, IL-6, IL-8, IL-10, MCP-1, TNF-α); however, for the purpose of this manuscript, only CRP, IL-6, IL-10 and TNF-α were analyzed and discussed. The remaining biomarker (IL-1β, IL-1ra, IL-8 and MCP-1) are presented in Supplementary Tables S2-S5. While articles were excluded if they involved populations whith overweight or obesity and diagnosed by other cardio-metabolic, immunological, or musculoskeletal pathology, or if the training program lasted less than two weeks or was accompanied by a diet program, articles were also excluded if they included older populations (>65 years) or postmenopausal women.

Data Extraction and Synthesis
The following information was extracted from each study: author, date of publication, sample size, participants' characteristics, experimental design and procedure, training intervention, and cytokines pre-vs. post-intervention difference. Data were independently extracted by two authors, and any discrepancy was resolved by a third author. Data was collected and presented as the mean ± standard deviation (SD) or mean (SD). When circulating cytokine levels or fat mass loss were provided in figures, values were estimated, and authors were contacted when no data was provided.
Given the heterogeneity of studies (gender, age, duration, intervention, types of training, etc.), it is not possible to combine the studies presented in this systematic review for quantitative analysis (meta-analysis).

C-Reactive Protein (CRP)
In response to ET, the circulating concentration levels of CRP showed a statistically significant reduction (from −9 to −53%) in 8 of 11 studies (Table 1). When the ET was compared to a control group, four of seven studies reported statistically significant decreases in circulating CRP levels [22,26,27,32], and the reduction of this biomarker was concomitant with a reduction in fat mass in 3 of 4 studies [24,29,30,32]. In response to RT, a statistically significant reduction in circulating CRP levels was found, but no differences were detected when RT was compared to a control group [33], whereas Kolahdouzi et al. (2019) did not find statistically significant changes in CRP levels within or between groups after RT [34].
Furthermore, we examined whether gender could influence these results. A statistically significant reduction in circulating CPR levels was observed in all studies conducted in women, while studies in men showed a statistically significant reduction in CPR levels in two of three (Table 1).

Interleukin-6 (IL-6)
Table 2 presents information extracted from the studies that examined circulating IL-6 concentration levels.  16.6 ± 6.1 EC = experimental condition; ET = Endurance training; HIIT = high intensity interval training; HRR = heart rate reserve; HR max = maximal heart rate; R = rest between series; RM = maximal repetition; RT = Resistance training; VO 2peak = peak oxygen uptake; NR = not reported; = estimated data. * p < 0.05 within group comparison; # p < 0.05 between groups comparison (vs control); some data are presented as median (interquartile range). Data are shown as mean ± SD.
In response to ET, circulating IL-6 showed a statistically significant reduction (from −26 to −32%) in 2 of 13 studies [24,26], but only one study reported statistically significant differences when the ET was compared to a control group [26]. Moreover, in Bruun et al., the reduction in IL-6 levels caused by ET was accompanied by a significant reduction in fat mass [24]. In response to the remaining training types, non-statistically significant differences were found for this biomarker.
Regarding gender-differences, in one of six studies conducted in women was found a statistically significant differences in CPR concentrations [26], while non-significant differences were found in any study performed in men for this biomarker ( Table 2). Table 3 presents information extracted from the studies that examined circulating IL-10 concentration levels. 7.17 ± 0.81 EC = experimental condition; ET = Endurance training; HIIT = high intensity interval training; HRR = heart rate reserve; HR max = maximal heart rate; NR = non-reported; R = rest between series; RM = maximal repetition; VO 2peak = peak oxygen uptake;~= estimated data; * p < 0.05 within group comparison; # p < 0.05 between group comparison. Data are shown as mean ± SD.
None of the studies performed included a cohort of women participants which satisfied the inclusion criteria, thus, we could not examine the potential influence of gender on the IL-10 response to exercise training. Table 4 presents information extracted from the studies that examined circulating TNF-α concentration levels.  56.1 ± 13.8 EC = experimental condition; ET = Endurance training; HIIT = high intensity interval training; HRR = heart rate reserve; HR max = maximal heart rate; NR = not reported; R = rest between series; RM = maximal repetition; RT = Resistance training; VO 2peak = peak oxygen uptake; = estimated data; * p < 0.05 within group comparison; # p < 0.05 between group comparison. Data are shown as mean ± SD.
Regarding gender-differences, in one of four of the studies conducted in women was found a statistically significant decrease of IL-6 levels [48]. Similarly, in studies on men, one of six studies reported a significant decrease of this biomarker [46].

Discussion
The present systematic review aimed at evaluating the effects of exercise training on pro-and anti-inflammatory cytokines as low-grade systemic inflammation markers in adults with overweight or obesity. The results are summarized and illustrated in Figure 2. Essentially, circulating concentration levels of CRP, IL-6 and TNF-α are reduced in response to endurance training, while TNF-α is also downregulated after resistance and concurrent training. Moreover, IL-10 increases after resistance training. However, only changes in circulating CRP and IL-10 coincides with a reduction in body fat mass.

Discussion
The present systematic review aimed at evaluating the effects of exercise training on pro-and anti-inflammatory cytokines as low-grade systemic inflammation markers in adults with overweight or obesity. The results are summarized and illustrated in Figure 2 Essentially, circulating concentration levels of CRP, IL-6 and TNF-α are reduced in response to endurance training, while TNF-α is also downregulated after resistance and concurrent training. Moreover, IL-10 increases after resistance training. However, only changes in circulating CRP and IL-10 coincides with a reduction in body fat mass.

Adults with overwegith and obesity
Endurance training CRP (together with IL-6) is considered a convincing marker of low-grade systemic inflammation [49]. Hepatocytes are the primary regulators of CRP, whose synthesis is stimulated by IL-6 production from T-cells [50]. One of the main purposes of CRP is to stimulate phagocytosis activity of macrophages to clear damaged cells or bacteria [50].
Most studies examined the effect of ET on the circulating levels of CRP found a significant decrease from −9 to −53% in 8 of 11 studies ( Figure 2 and Table 1). This effect was observed in studies performed ET programs at moderate intensity, but with varied training duration and frequency. Interestingly, the significant reduction of CRP was concomitant with a fat mass diminution [24,29,30,32]. Since no substantial reduction of visceral fat (~5%) is required to promote a reduction of circulating CRP [29], exercise training may indirectly regulate the circulating concentrations levels of this inflammatory marker through stimulating the reduction of body fat mass, consequently promoting a decrease of macrophages infiltration in peripheral tissues.
Initially, IL-6 was categorized as a pro-inflammatory cytokine, given its upregulation during infection. Later, the anti-inflammatory effect of this cytokine was discovered [51] Although mononuclear cells can be responsible for the circulating fluctuation of IL-6, this does not seem to be the case in response to exercise [52]. In fact, skeletal muscle has been identified as a critical regulator of circulating levels of IL-6 after prolonged exercise training [53].
In the present review, it can be observed that in adults with overweight or obesity, ET training causes a decrease in circulating levels of IL-6 (from −2 to −56%) in 11 of 13 studies regardless of exercise-induced fat mass loss ( Figure 2 and Table 2). These results seem to indicate that depending on the prevailing stimuli, training (anti-inflammatory), or fat mass accumulation (pro-inflammatory), IL-6 bioavailability may be regulated by a different cell CRP (together with IL-6) is considered a convincing marker of low-grade systemic inflammation [49]. Hepatocytes are the primary regulators of CRP, whose synthesis is stimulated by IL-6 production from T-cells [50]. One of the main purposes of CRP is to stimulate phagocytosis activity of macrophages to clear damaged cells or bacteria [50].
Most studies examined the effect of ET on the circulating levels of CRP found a significant decrease from −9 to −53% in 8 of 11 studies ( Figure 2 and Table 1). This effect was observed in studies performed ET programs at moderate intensity, but with varied training duration and frequency. Interestingly, the significant reduction of CRP was concomitant with a fat mass diminution [24,29,30,32]. Since no substantial reduction of visceral fat (~5%) is required to promote a reduction of circulating CRP [29], exercise training may indirectly regulate the circulating concentrations levels of this inflammatory marker through stimulating the reduction of body fat mass, consequently promoting a decrease of macrophages infiltration in peripheral tissues.
Initially, IL-6 was categorized as a pro-inflammatory cytokine, given its upregulation during infection. Later, the anti-inflammatory effect of this cytokine was discovered [51]. Although mononuclear cells can be responsible for the circulating fluctuation of IL-6, this does not seem to be the case in response to exercise [52]. In fact, skeletal muscle has been identified as a critical regulator of circulating levels of IL-6 after prolonged exercise training [53].
In the present review, it can be observed that in adults with overweight or obesity, ET training causes a decrease in circulating levels of IL-6 (from −2 to −56%) in 11 of 13 studies, regardless of exercise-induced fat mass loss ( Figure 2 and Table 2). These results seem to indicate that depending on the prevailing stimuli, training (anti-inflammatory), or fat mass accumulation (pro-inflammatory), IL-6 bioavailability may be regulated by a different cell type (immune and skeletal muscle cell) facilitating the dual pro-vs. anti-inflammatory function attributed to IL-15 [54], a cytokine of the same family. This dual function of IL-6 may be supported by the consistent but non-significant decrease of IL-6 observed in response not only after endurance, but also after resistance, concurrent, and HIIT training ( Figure 2).
In humans, IL-10 synthesis is primarily regulated by monocytes [55], facilitating the inhibition of pro-inflammatory cytokines production in macrophages [56]. Although IL-10 does not seem to be expressed by myocytes [57], in healthy individuals, exercise promotes an upregulation (>80%) of the circulating concentration levels of this cytokine [58]. However, in this review, only RT reported a consistent increase of IL-10, which was concomitant with a reduction in fat mass ( Figure 2 and Table 3). It can be argued that in skeletal muscle, the increased expression levels of microRNAs and/or myokines after exercise [59], may increase IL-10 production from monocytes regulating the pro-inflammatory signaling pathway in peripheral tissues, such as adipose tissue, in which macrophages and neutrophils are infiltrated.
Furthermore, in healthy humans, acute bouts of exercise activate myocytes TNF-α expression levels, whereas the small increase of circulating concentrations suggests that skeletal muscle may not be the major regulator of this cytokine [58]. In fact, it seems that monocytes can be the key regulator of this cytokine [60,61], which can be responsible for TNF-α reduction found after ET, RT and HIIT in this review.
Interestingly, a decreased concentration levels of TNF-α has been linked with the downregulation of IL-6 [62] and upregulation of IL-10 [56]. Regarding the TNF-α/IL-6 relationship, most of the analyzed studies reported a coinciding decrease between these two cytokines [24,30,35,37,39]. This decrease of TNF-α and IL-6 can reduce the risk of insulin resistance in a population with obesity [63]. Moreover, a coinciding inverse relationship between TNF-α and IL-10 was also observed in most studies [23,42,46]. Therefore, this evidence may support the regulative role of TNF-α on IL-6 and IL-10 concentration levels in adults with overweight or obesity who experimented an alteration of monocyte metabolism after exercise training. This effect facilitates a diminished inflammatory state by downregulating (IL-6) and upregulating (IL-10).
Despite the regulative role of exercise training on the circulating concentration levels of CRP, IL-6, IL-10 and TNFα observed here, the amount of evidence and the disparity among participants' characteristics, training protocols, and methods used to determine cytokines concentration levels in blood were major limitations of this systematic review, all of which complicate the understanding of exercise training effects on low-grade systemic inflammation in adults with overweight and obesity. Therefore, further studies with homogeneous characteristics are needed to conduct meta-analysis and to decode the bioavailability of cytokines as markers of the intricate pro-and anti-inflammatory response to training in this population.

Conclusions
In summary, prolonged doses of endurance, resistance and high-intensity interval training promote a reduction on the circulating concentration of pro-inflammatory cytokines (IL-6, CRP and TNF-α), while resistance training stimulates an increase in antiinflammatory cytokines (IL-10). This regulative role of exercise training on systemic low-grade inflammation seems to be independent of exercise-induced fat mass loss.
Therefore, the present systematic review may help clinicians to attenuate low-grade systemic inflammation by prescribing the adequate dose of exercise training according to the alteration of pro-and anti-inflammatory cytokines of each patient. However, additional clinical trials are required to further elucidate the regulative role of exercise on the cytokines responsible for low-grade systemic inflammation.

Conflicts of Interest:
The authors declare no conflict of interest.