Variable number of tandem repeat polymorphisms of the interleukin-1 receptor antagonist gene IL-1RN: a novel association with the athlete status

Factors that affect athletic performance are only partially defined. However, there are accumulating data indicating that both pro- and anti-inflammatory responses can play a role. Prolonged strenuous exercise results in an activation of the acute phase response with some similarities to sepsis [1]. The induction of a systemic cytokine response, including elevated plasma levels of interleukin (IL)-1 receptor antagonist (IL-1ra), IL-6, IL-8, IL-10, and granulocyte colony-stimulating factor (G-CSF), following intense physical activity, is well documented [1‐5]. IL-1ra is a member of the IL-1 family, which includes the classical IL-1 (α and β) cytokines, IL-18 and the newly described IL-1F 5-11 [6]. IL-1ra acts as an antagonist of the IL-1 receptor type I (IL-1RI) and prevents IL-1 (α and β)-dependent signaling. In the absence of IL-1ra, the activity of IL-1 is unopposed, promoting rampant inflammation. Recent research has shown that deficiency of IL-1ra in humans can cause IL-1-mediated systemic and local inflammation that affect infants soon after birth [7]. IL-1α is an IL-1RI agonist with proinflammatory action, that acts primarily as an intracellular transcriptional regulator; it is not found in high concentration in the circulation. In contrast, IL-1β can be secreted from cells in large quantities and thus it is commonly found in serum and secretions. In general, IL-1β acts synergistically with tumor necrosis factor (TNF)-α, activating proinflammatory responses in a wide range of cells and promoting the acute phase response [6]. For these reasons IL-1β levels have been studied in relation to physical activity. However, the circulating concentration of IL-1β is either unaffected by exercise, or it exhibits relatively small, delayed changes [1, 2]. In contrast to minimal changes in serum circulating IL-1β after exercise, there is evidence of increased local IL-1β levels after exercise within skeletal muscle; this is most likely associated with response to micro-injury of skeletal muscle caused by intense physical activity [8]. Interestingly, Mahoney and colleagues [9] demonstrated that elevated gene expression of IL-1RI is induced in skeletal muscle after eccentric exercise. Inflammatory and chemoattractant mediators likely promote phagocytosis of trauma-induced cellular debris by macrophages and these cells can continue to secrete IL-1β up to 5 days post-injury [8]. It is of note that accumulation and activation of muscle resident macrophages is a rich source of growth factors postulated to stimulate myogenesis [4]. Thus, inflammation may serve as a mechanism promoting hypertrophy. However, pre- and post-exercise levels of inflammatory factors display considerable variation among people, and this is likely influenced, at least partially, by genetic variation [10, 11].

IL-1β plays an important role in the regulation of acute inflammatory response [12]. After an insult to the body, IL-1β is one of the first pro-inflammatory cytokines to be activated. Intense physical activity can cause micro-injury to skeletal muscle, and cytokines such as IL-1β and TNF-α are thought to initiate and regulate the repair process. IL-1β is able to induce the secretion of several inflammatory factors, such as IL-6, IL-8, TNF-α, and GM-CSF, by many different cell types including myoblasts, smooth and skeletal muscle cells, fibroblasts, macrophages, peripheral blood mononuclear cells (PBMC) and endothelial cells [12‐14]. IL-1β binds IL-1RI on the surface of a variety of cells and initiates a cascade of events leading to recruitment and activation of macrophages and neutrophils. However, inflammation must be strictly regulated to avoid secondary tissue damage and myopathy. The activity of IL-1β is modulated mainly by a receptor antagonist (IL-1ra) that inhibits IL-1 action by competing for its receptor [6, 15]. In healthy individuals, IL-1ra is easily detectable in plasma (concentrations are hundreds of pg/mL), but IL-1β levels are usually undetectable (very few or less than pg/mL) [15, 16]. In general, after an insult to the body both IL-1β and IL-1ra increase, and thus, plasma levels of these cytokines are positively correlated [17]. However, the final inflammatory activity of IL-1β is considered to be dependent on the actual ratio of IL-1β over IL-1ra [7, 18].

There is evidence that susceptibility to inflammation is influenced by genetic variation in cytokine genes [14, 19]. Several studies showed that polymorphisms in the IL-1β (IL-1B) and IL-1ra (IL-1RN) genes correlate with altered protein expression in vitro [14, 20] and in vivo [10, 16, 21, 22]. These two genes map close to each other on human chromosome 2 [12]. Polymorphisms in the IL-1 gene cluster associate with many clinical conditions, including inflammatory and infectious diseases (periodontal and arterial diseases, altered metabolic conditions, type 2 diabetes, type 2 diabetes complications, gastric cancer, rheumatoid arthritis, ankylosing spondylitis, systemic inflammation, myopathies, Alzheimer's disease, malaria, and bacterial vaginosis) [6, 12, 17, 19, 21, 23‐31]. Mostly, two single nucleotide polymorphisms (SNPs) in IL-1B have been studied for disease predisposition, one at position -511 in the promoter region [32] and another at position +3954 in exon 5 (TaqI restriction site polymorphism) [18]. In addition, a pentaallelic polymorphic site in intron 2 of the IL-1RN gene consisting in a variable number tandem repeats (VNTR, 86 bp repeats) has been extensively investigated in relation to a variety of pathological conditions, including inflammatory myopathies [6, 18, 19, 29, 33, 34].

Recently, a study performed on 24 sedentary subjects (of unspecified ethnicity) selected on the basis of their haplotype pattern in the IL-1 gene cluster (based on specific combinations of +4845 IL-1A, +3954 IL-1B, -511 IL-1B, and -3737 IL-1B polymorphisms) showed that subjects C/C for IL-1B +3954 who carry allele 2 at IL-1RN +2018 were associated with the inflammation of skeletal muscle, following acute resistance exercise [10].

The aim of this study was to investigate whether polymorphisms in the IL-1β and IL-1ra genes affect athletic status, competitive athlete vs. non-athlete. We evaluated whether the genotype frequencies of -511 and +3954 SNPs of IL-1B, and intron 2 VNTR of IL-1RN vary between athletes and non-athletes. Athletes of National or Regional competitive standard were recruited. This analysis was performed in a relatively homogeneous ethnic group of Caucasian Italian subjects because associations between polymorphisms in immune system-related genes may be confounded by allele frequency differences between ethnic groups [35, 36]. We examined whether the polymorphisms of these genes are associated with being a professional (high-grade) or non-professional (medium-grade) athlete.

The study population consisted of 205 athletes; of these 53 were professional (high-grade, National level) and 152 were non-professional (medium-grade, Regional level, recreational athletes not paid for performing competitions) athletes regularly participating in athletic competitions. Controls were 458 healthy non-athletes. Demographic characteristics and sport disciplines practiced by the 205 athletes are shown in Table 1.

A total of 663 study subjects were genotyped for the three gene loci (IL-1B in position -511 and +3954; and IL-1RN intron 2 VNTR).

Neither case nor control samples deviated from Hardy Weinberg equilibrium (HWE) at site -511 or site +3954. The controls deviated from HWE at the VNTR in IL-1RN (P = 0.001); while athletes marginally deviated from HWE (P = 0.097). Athletes and controls did not differ in the distribution of -511 IL-1B genotypes; 38.8% athletes vs. 44.8% controls were homozygous CC, 48.8% athletes vs. 45.9% controls were heterozygous CT, and 12.4% athletes vs. 9.4% controls were homozygous TT (Table 2). The two groups also did not differ significantly in allele frequency (P = 0.11).

Genotype and allele distribution of IL-1B +3954 gene polymorphism are shown in Table 3. Athletes and controls did not differ; 60.0% athletes vs. 62.2% controls were homozygous CC, 36.1% athletes vs. 32.3% controls were heterozygous CT, and 3.9% athletes vs. 5.5% controls were homozygous TT. The two groups also did not differ significantly in allele frequency (P = 0.89).

For the IL-1RN VNTR polymorphism three alleles (allele 1, 2 and 3) were common in our study; allele 4 was found in 3 subjects (one genotype 1/4 and two genotypes 2/4, all in the control group) and allele 5 was detected in only 1 subject (one genotype 1/5, in the athlete group). The most common allele, the IL-1RN*1 allele, was less frequent in athletes than in controls (65.4% vs. 74.0%, P = 0.001), whereas the second commonest allele, IL-1RN*2, had an allele frequency higher in athletes than controls (32.2% vs. 22.9%, P < 0.001), finally, IL-1RN*3 had similar frequency in athletes and controls (P = 0.57) (overall alleles, P = 0.001) (Table 4). Specifically, athletes were less likely to have the 1/1 genotype (OR = 0.55, 95% CI 0.40-0.77), and 2-fold more likely to be 1/2 (OR = 1.93, 95% CI 1.37-2.74). Athletes and controls did not differ in frequency of the 2/2 genotype. However, we observed increased frequencies of cumulative 1/2, 2/2, 2/3 and 2/4 genotypes in athletes (53.7%) compared with controls (36.7%), OR = 2.00 (95% CI = 1.43-2.79), suggesting a dominant effect of the 2 allele. These observations were significant in both male and female athletes when analyzed separately (data not shown, see additional File 1). Genotypes that included the 3 allele did not differ between athletes and controls.

In a secondary analysis (Table 5), we compared IL-1RN genotypes between the 53 professional athletes and the 152 non-professional athletes. Interestingly, we found that the 1/2 VNTR genotype was almost 2-fold more frequent in professional than in non-professional athletes (OR = 1.92, 95% CI = 1.02-3.61). No other significant differences were noted between professional and recreational athletes.

Comparison of professional athletes with non-athletes revealed that genotype 1/2 VNTR was 3-fold more frequent in high-grade athletes (OR = 3.12, 95% CI = 1.75-5.56) (Table 5). The same genotype was only 1.6-fold more frequent in medium grade/recreational athletes than in non-athletes (OR = 1.62, 95% CI = 1.10-2.40) (Table 5). The 1/1 genotype was 2.5-fold more frequent in non-athletes than in professional athletes (OR = 2.52, 95% CI = 1.40-4.56), but 1.6-fold more frequent in non-athletes than in recreational athletes (OR = 1.61, 95% CI = 1.11-2.33). Overall (frequencies already reported in Table 4), the 1/1 genotype was 1.8-fold more frequent in non-athletes than in all athletes (OR = 1.80, 95% CI = 1.29-2.51).

Haplotype analyses were performed, using all three polymorphisms (Table 6). There was virtually no LD between any of the three sites in athletes (Figure 1) or non-athletes (r² ≤ 0.05) (Figure 2). The three site haplotype was differentially distributed between athletes and controls (P = 0.012). Individually only one haplotype associated with athlete status. Specifically, (-511)C-(+3954)T-(VNTR)2 was 3 times more common in athletes than non-athletes (OR = 3.02, 95%CI = 1.15-7.87). We also analyzed the three pair wise haplotypes and found that only those that included the VNTR were significant, indicating that the association was related to the VNTR or a marker in linkage disequilibrium with it.

Our study supports the notion that the IL-1 family of genes is significantly associated with physical performance. The finding of a higher frequency of allele 2, and specifically of the 1/2 genotype VNTR IL-1RN, in athletes with respect to non-athletes highlights that immune genetic variants are involved in athletic predisposition. Based on previous evidence of inflammatory effects of the VNTR IL-1RN allele 2, the association of IL-1 related genes with athlete status appears to associate a genetic predisposition of a moderately enhanced inflammatory response. However, we cannot exclude much more indirect effects modulated by IL-1ra. Immunity in physical exercise is likely regulated by subtle tuning of cytokine levels; for example, a partial reduction in IL-1ra could reinforce immune defenses against exercise-associated micro-injuries.

It was beyond the scope of this study to investigate the effects of IL-1 polymorphisms on exercise-induced systemic and local inflammation, recovery capacity and type of sport activity (aerobic vs. anaerobic; endurance vs. sprint), which should be clarified in future studies. Of major interest will be the comparison of aerobic vs. anaerobic athletes; however, in the present study this kind of evaluation was not possible due to small number of athletes participating in purely anaerobic sports. Future studies relating IL-1RN genotypes and cytokine levels should focus preferably on immune genetic profiles and inflammatory markers evaluated before and after exercise. Although our results indicate an association between the IL-1 family of cytokines and athlete status, it is not clear how, specifically, the VNTR IL-1RN polymorphism affects the phenotype promoting a relationship between this allele and grade of athletic performance.

In this study we focused on 2 SNPs of the IL-1B gene and 1 VNTR in the IL-1RN gene because these are the most frequently studied polymorphisms of the IL-1β and IL-1ra cytokines, which have been associated with a variety of human conditions. Interestingly, athlete status may constitute a confounding factor contributing to variability of results concerning the association of the IL-1RN polymorphisms with pathologic conditions [61].

Our present results should prompt larger studies on additional polymorphisms in the IL-1 family in relation to physical performance and elite athletic grade, particularly, those located in the IL-1ra gene, as the SNP at position +2018.

In conclusion, we demonstrated in a cross-sectional study that a polymorphism potentially able to modulate the levels of the IL-1ra cytokine, and indirectly IL-1 (α and β) activity, is associated with athlete status. Further research into immune factors that increase athletic performance will help to inform the design of how immune modulators may improve recovery or implement cure interventions after injuries. Our finding suggests that clinical trials performed to assess efficacy of various treatments for athlete's recovery should probably differentiate subjects on a genetic basis in addition to etiology to maximize the chances of observing statistically significant and reproducible results. Our study suggests cytokines are important molecules to further explore in relation to sports activity. Overall, this study may contribute to an increased understanding of the genetic basis of athletic performance. It may also increase our understanding of inflammatory mechanisms associated with intensive exercise.