Association of Genetic Variances in ADRB1 and PPARGC1a with Two-Kilometre Running Time-Trial Performance in Australian Football League Players: A Preliminary Study

enetic variants in theangiotensin-converting enzyme(ACE) (rs4343),alpha-actinin-3(ACTN3) (rs1815739),adrenoceptor-beta-1(ADRB1) (rs1801253), andperoxisome proliferator-activated receptor gamma coactivator 1-alpha(PPARGC1A) (rs8192678) genes have previously been associated with elite athletic performance. This study assessed the in uence of polymorphisms in these candidate genes towards endurance test

University, Perth 6027, Australia 10 Faculty of Health, Queensland University of Technology, Brisbane 4059, Australia *Correspondence: ryan.anderton@nd.edu.au (R.S.A.); n.hart@ecu.edu.au (N.H.H.) Abstract: Genetic variants in theangiotensin-converting enzyme(ACE) (rs4343),alpha-actinin-3(ACTN3) (rs1815739),adrenoceptor-beta-1(ADRB1) (rs1801253), andperoxisome proliferator-activated receptor gamma coactivator 1-alpha(PPARGC1A) (rs8192678) genes have previously been associated with elite athletic performance. This study assessed the in uence of polymorphisms in these candidate genes towards endurance test performance in 46 players from a single Australian Football League (AFL) team. Each player provided saliva buccal swab samples for DNA analysis and genotyping and were required to perform two independent two-kilometre running time-trials, six weeks apart. Linear mixed models were created to account for repeated measures over time and to determine whether player genotypes are associated with overall performance in the two-kilometre time-trial. The results showed that theADRB1 Arg389GlyCC (p= 0.034) andPPARGC1A Gly482SerGG (p= 0.031) genotypes were signi cantly associated with a faster two-kilometre time-trial. This is the rst study to link genetic polymorphism to an assessment of endurance performance in Australian Football and provides justi cation for further exploratory or con rmatory studies. Keywords:ADRB1;PPARGC1a; genes; sport; performance; Australian Football; endurance 1. Introduction Australian Football (AF) is a multi-dimensional team sport, which requires a com- bination of endurance, strength, power, speed, and competency in sport-speci c skills including kicking, handballing, marking, and tackling [1–7]. The Australia Football League (AFL) represents an elite AF competition and has playing times of four 20-min quarters with time on, with games often spanning beyond 120 min due to stoppages (or, during COVID-19-modi ed seasons, four 16-min quarters with time on, often spanning 100 min due to stoppages). With an oval playing eld of 135 to 185 m in length and 110 to 155 m in width across the competition [8], players consistently run more than 13 km during a typical game [9,10]. Accordingly, AF is characterised as an endurance sport consisting of multiple Sports2021,9, 22.

Sports2021,9, 22 2 of 12 high-intensity and moderate-intensity efforts [11–17], with no movement restrictions on footballers during active play. However, positional differences in movement and match- play pro les exist, with players commonly grouped as nomadic (i.e., high running volumes covering the entire playing surface, such as mid elders) or non-nomadic (key positions, such as ruckmen, forwards, and backs) players [14]. Due to these unique qualities, there are many athletic abilities required to be successful within the elite AFL competition. Athletic ability and performance can be in uenced by multiple variables, including environmental factors such as training history, nutrition, body morphology, cognitive factors, and injury susceptibility. Recently, the genetic underpinning of athletic perfor- mance in elite athletes is gaining ascendency to understand the predictability of an athlete's capacity to perform under various constraints in addition to the potential trainability or responsiveness of athletes to various strength and conditioning modalities. Genetics can affect strength, power, and endurance, additional to other traits such as muscle bre size and composition, exibility, and neuromuscular coordination [18–25]. Athletic status is at least a partially inheritable trait, with upwards of 66% of athlete variance being explained via genetics [26]. Interest in how an individual's genotype can impact phenotypes related to athletic performance has gained traction in recent times in AF. Some of the rst iden- ti ed and most in uential candidate genes associated with athletic performance include angiotensin-converting enzyme(ACE),alpha-actinin-3(ACTN3),adrenoceptor-beta-1(ADRB1), andperoxisome proliferator-activated receptor gamma coactivator 1-alpha(PPARGC1A) [27]. Variants within theACEandACTN3genes have been associated with endurance, strength, and power. TheACEenzyme regulates uid volume within therenin–angiotensin– aldosteronesystem (RAS) [28,29]. Within intron 16 of theACEgene is an insertion/deletion (I/D) polymorphism (rs4343) of an Alu repetitive element, with the I allele being associated with a lower level ofACEenzyme activity [30]. An abundance of evidence indicates that the insertion polymorphism and I allele are associated with elite endurance status in single dis- ciplinary sports such as running (ranging from middle distance to ultramarathon) [31,32], triathletes [33], and rowers [34,35]. The DD genotype has also been found to be bene - cial to sprinters [36]. TheACTN3protein contributes to the formation of skeletal

An abundance of evidence indicates that the insertion polymorphism and I allele are associated with elite endurance status in single dis- ciplinary sports such as running (ranging from middle distance to ultramarathon) [31,32], triathletes [33], and rowers [34,35]. The DD genotype has also been found to be bene - cial to sprinters [36]. TheACTN3protein contributes to the formation of skeletal muscle bres [37], and helps coordinate type II fast-twitch muscle bre contraction [22,23,37]. The R577Xpolymorphism (rs1815739) within theACTN3gene can encode for a premature stop codon (T allele), which is associated with improved endurance performance [22,38]. The TT genotype frequencies were found to be signi cantly lower in bodybuilders and power athletes [39], whereas the T allele and TT genotype frequencies are likely higher in endurance athletes such as endurance running [40–42], road cyclists [40], and rowers [40]. Research into the in uence of theACTN3 R577Xpolymorphism in sports has found higher frequencies of the C allele and CC genotype in strength, power, or speed sports such as speed skating [43], track sprinters [36,41,42,44], and eld athletes [41,42]. TheACTN3gene has also been investigated in soccer, with Santiago et al. [45] nding that the CC and CT genotypes were signi cantly higher. Variants within theADRB1 Arg389Gly(rs1801253) andPPARGC1A Gly482Ser(rs8192678) genes have also been linked with endurance performance [24,25,46–52]. TheADRB1gene encodes for G-coupled receptors in cardiac tissue that impacts cardiac output [25,53]. Due to this, most research intoADRB1genetic variation has been conducted in patients with cardiac conditions, such as idiopathic or ischemic cardiomyopathy. For example, Wagoner, Craft, Zengel, McGuire, Rathz, Dorn, and Liggett [52] investigated a variant (rs1801253) within theADRB1gene and found that patients with the C allele demonstrated an increase in maximum rate of oxygen consumption (VO2max), exercise time, and endurance perfor- mance. However, Wessner et al. [54] found that the GG genotype of thisADRB1variant was more prevalent in international or highest national level handball and soccer players. ThePPARGC1Agene is involved in glucose regulation and lipid metabolism, as well as determination of bre type and skeletal muscle bre formation [24,46]. ThePPARGC1A Gly482Servariant within this gene has been associated with athletic performance, with the A

Wessner et al. [54] found that the GG genotype of thisADRB1variant was more prevalent in international or highest national level handball and soccer players. ThePPARGC1Agene is involved in glucose regulation and lipid metabolism, as well as determination of bre type and skeletal muscle bre formation [24,46]. ThePPARGC1A Gly482Servariant within this gene has been associated with athletic performance, with the A allele found to be at a lower frequency in Israeli endurance runners, with signi cant

Sports2021,9, 22 3 of 12 differences between those endurance runners and sprinters of the same level [55]. In addition, the AA genotype was found to be the more favourable genotype for a population of Russian and Lithuanian powerlifters compared to controls [56]. In a meta-analysis of the PPARGC1ars8192678 variant [57], the A allele and the AA genotype were suggested to be bene cial for athletic performance regardless of the type of sport; however, studies in AF are lacking. Due to the potential in uence of these candidate genetic variants on endurance performance in athletic pursuits and non-elite AF populations, the primary purpose of this study was to investigate any associations of theACE(rs4343),ACTN3(rs1815739),ADRB1 (rs1801253), andPPARGC1A(rs8192678) polymorphisms with two-kilometre time-trial endurance test performance in elite AF players. The secondary purpose of the study was to determine if there was a genetic difference between nomadic and non-nomadic players, to examine if particular genotypes were more favourable for certain positions. 2. Materials and Methods 2.1. Participants Forty-six (n= 46) elite male AF players recruited from an AFL football club participated in the study. All players were injury-free at the time of testing. To ensure anonymity, players were assigned a randomised, non-identi able code. All players were provided with information letters outlining the purpose of the study, along with its potential bene ts and risks, and provided written informed consent for their participation. The study was approved by the Edith Cowan University Human Research and Ethics Committee (ID: 2019-00181-JACOB). 2.2. Sample Collection and DNA Analysis Buccal saliva samples were collected via mouth swabs with participants instructed to brush the edge of a soft tip swab along the insides of their cheek and gums for 30 s [58,59]. Players were asked not to consume coffee, alcohol, or food two hours prior to saliva collection. Collected samples were labelled with a numeric code for de-identi cation and were sent to the Australian Genome Research Facility (AGRF; Brisbane, QLD, Australia; NATA 17025) for DNA extraction and genotyping using the Agena Bioscience MassARRAY system (AGRF). A summary of the genetic variants investigated in this study is presented in Table.

or food two hours prior to saliva collection. Collected samples were labelled with a numeric code for de-identi cation and were sent to the Australian Genome Research Facility (AGRF; Brisbane, QLD, Australia; NATA 17025) for DNA extraction and genotyping using the Agena Bioscience MassARRAY system (AGRF). A summary of the genetic variants investigated in this study is presented in Table. 2.3. Endurance Testing Endurance performance was evaluated twice to obtain accurate and reliable results for two-kilometre time-trials performed six weeks apart. Trials were conducted on a certi ed athletics track (Oceania Athletics Association) during pre-season training in conjunction with the football club's regular pre-season testing. During the six-week training block, players participated in the same structured pre-season training program overseen by the club's high-performance department. All times were visually veri ed through video recording in conjunction with the AFL club's guidelines for two-kilometre time-trials. 2.4. Statistical Analysis Data were statistically analysed using SPSS V.24 (IBM, Armonk, NY, USA). A gen- eralised linear mixed model (GLMM) was created to analyse the relationship between covariates and overall performance in the two-kilometre time-trial. Separate GLMMs were created to account for repeated measures over time, and to determine whether player genotypes are associated with overall performance in the two-kilometre time-trial. Beta ( ) coef cients have been reported as a standardised measure of effect (i.e., effect size) for the GLMMs. Further analysis was completed with the genotype and allele frequencies, which were compared using Pearson's Chi-square ( 2) tests between nomadic and non- nomadic positions, as well as T-tests, or a non-parametric alternative, to determine the mean difference between groups. A signi cant nominalp-value of <0.05 was employed.

Sports2021,9, 22 4 of 12 ETA square values ( 2) for Chi-square analyses were also calculated to determine the effect of any observed associations, de ned as none ( 2< 0.010), small ( 2< 0.060), moderate ( 2 < 0.140), and large ( 2 < 0.200). Table 1.Variant distribution in elite Australian Football (AF) athletes. Elite AFn(%) ACTN3 R577X CC 21 (45.7%) CT 23 (50.0%) TT 2 (4.3%) C allele 65 (70.7%) T allele 27 (29.3%) ACE I/D II 11 (23.9%) ID 23 (50.0%) DD 12 (26.1%) I allele 45 (48.9%) D allele 47 (51.1%) ADRB1 Arg389Gly CC 25 (54.3%) CG 18 (39.1%) GG 3 (6.5%) C allele 68 (73.9%) G allele 24 (26.1%) PPARGC1a Gly482Ser GG 23 (50.0%) GA 18 (39.1%) AA 5 (10.9%) G allele 64 (69.6%) A allele 28 (30.4%) 3. Results This study is preliminary in nature with access to a single AFL football team producing a sample size of 46 elite AF players. Demographic characteristics of these male players were: age = 24.4 4.0 years; weight = 88.3 8.1 kg; height = 187.8 6.3 cm; body mass index (BMI) = 24.9 1.4. Two-kilometre time-trials produced completion times of 406.9 ( 22.0) seconds for the rst trial and 400.9 ( 17.0) seconds for the second trial. The players' genotype and allele frequencies are presented in Table. Age (p= 0.750; 95% CI [ 1.191, 0.861]), height (p= 0.086; 95% CI [ 6.964–104.060]), weight (p= 0.110; 95% CI [ 0.086–0.827), and BMI (p= 0.674; 95% CI [ 2.164, 3.332]) were not signi cantly associated with two-kilometre performance (Table). Therefore, these variables were not considered as covariates in subsequent GLMMs investigating individual genetic variants. Table 2.Generalised linear mixed model of covariates and the two-kilometre time-trial time. Variable Coef cient Standard Error t Value Signi cance 95% CI Time 5.467 4.184 1.307 0.195 2.847–13.781 Age 0.165 0.516 0.320 0.750 1.191–0.861 Height 0.524 0.302 1.735 0.086 0.076–1.124 Weight 0.378 0.231 1.635 0.106 0.082–0.838 BMI 0.584 1.382 0.423 0.674 2.164–3.332 Note: coef cient = standardised effect size. To account for repeated measures, separate GLMMs between the rst time-trial (time point 1) and

Standard Error t Value Signi cance 95% CI Time 5.467 4.184 1.307 0.195 2.847–13.781 Age 0.165 0.516 0.320 0.750 1.191–0.861 Height 0.524 0.302 1.735 0.086 0.076–1.124 Weight 0.378 0.231 1.635 0.106 0.082–0.838 BMI 0.584 1.382 0.423 0.674 2.164–3.332 Note: coef cient = standardised effect size. To account for repeated measures, separate GLMMs between the rst time-trial (time point 1) and the second time-trial (time point 2) revealed thatADRB1 Arg389Glyand PPARGC1a Gly482Servariants were signi cantly associated with two-kilometre time-trial performance (Table). Participants with the ADRB1 Arg389GlyCC genotype were 17.568 s

Sports2021,9, 22 5 of 12 faster (p= 0.034; 95% CI [ 33.766, 1.371]) in the two-kilometre time-trial when compared to participants carrying the GG genotype. Similarly, individuals with thePPARGC1a Gly482SerGG genotype were 14.421 s (p= 0.031; 95% CI [1.311, 27.531]) faster in the two-kilometre time-trial than individuals with the AA genotype. Table 3.Generalised linear model of genetic variables and the two-kilometre time-trial time. Variable Coef cient Standard Error Signi cance 95% CI ADRB1 Intercept 416.292 7.826 0.000 400.738–431.849 Time point 1 5.467 4.128 0.189 2.740–13.673 2 0 * ADRB1 Arg389Gly CC 17.568 8.128 0.034 33.766– 1.371 CG 13.806 8.297 0.100 30.301–2.688 GG 0 * ACE Intercept 397.802 4.403 0.000 389.049–406.556 Time point 1 5.467 4.198 0.196 2.878–13.812 2 0 * ACE I/D II 6.049 5.815 0.301 5.510–17.608 ID 4.146 4.999 0.409 5.791–14.084 DD 0 * ACTN3 Intercept 399.571 9.807 0.00 380.074–419.067 Time point 1 5.467 4.222 0.199 2.927–13.860 2 0 * ACTN3 R577X CC 0.748 10.153 0.941 19.436–20.932 CT 2.929 10.093 0.772 17/135–22.993 TT 0 * PPARGC1a Intercept 389.763 6.175 0.00 377.488–402.038 Time point 1 5.467 4.111 0.187 2.706–13.640 2 0 * PPARGC1a Gly482Ser GG 14.421 6.595 0.031 1.311–27.531 AG 11.293 6.800 0.100 2.224–24.810 AA 0 * Note: Signi cant effects are bolded. * indicated comparison group. coef cient = standardised effect size. Nomadic players had an average height of 185.08 4.58 cm, average weight of 84.78 5.61 kg , and average BMI of 24.75 1.37. Non-nomadic players had an average height of 197.35 3.36 cm, average weight of 100.57 5.36 kg, and average BMI of 25.83 1.27. A T-test was conducted been nomadic and non-nomadic players for height (p= 0.312; 95% CI [ 15.234– 9.312]), while Mann–Whitney tests were conducted for weight and BMI. Non-nomadic players scored higher in weight (Mdn = 40.58) and BMI (Mdn = 30.54) than nomadic players (weight: Mdn = 18.31; BMI: Mdn = 21.76; weight: U = 409.00;p= 0.00; BMI: U = 288.500;p= 0.048). There was a signi cant difference inADRB1 Arg389Glygenotype frequency between nomadic and non-nomadic positions ( 2 = 6.293,p= 0.037), with the CC genotype being signi cantly overrepresented in nomadic positions

weight (Mdn = 40.58) and BMI (Mdn = 30.54) than nomadic players (weight: Mdn = 18.31; BMI: Mdn = 21.76; weight: U = 409.00;p= 0.00; BMI: U = 288.500;p= 0.048). There was a signi cant difference inADRB1 Arg389Glygenotype frequency between nomadic and non-nomadic positions ( 2 = 6.293,p= 0.037), with the CC genotype being signi cantly overrepresented in nomadic positions (63.6%) compared with non-nomadic positions (25.0%; Table). Furthermore, the C allele was signi cantly overrepresented in nomadic positions (80.3%) compared to non-nomadic positions (19.7%; 2 = 6.148,p= 0.017).

Sports2021,9, 22 6 of 12 Table 4.Genotype and allele distribution between nomadic and non-nomadic positional categories. Nomadicn(%) Non-Nomadic n(%) Signi cance (p) ETA Squared ( 2 ) ACE I/D II 8 (24.2%) 3 (25.0%) ID 16 (48.5%) 7 (58.3%) 0.904 0.013 Small DD 9 (27.3%) 2 (16.7%) I allele 32 (48.5%) 13 (54.2%) 0.812 0.005 Small D allele 34 (51.5%) 11 (45.8%) ACTN3 R577X CC 14 (42.4%) 6 (50.0%) CT 17 (51.5%) 6 (50.0%) 1.000 0.019 Small TT 2 (6.1%) 0 (0.0%) C allele 45 (68.2%) 18 (75.0%) 0.611 0.009 None T allele 21 (31.8%) 6 (25.0%) ADRB1 CC 21 (63.6%) 3 (25.0%) Arg389Gly CG 11 (33.3%) 7 (58.3%) 0.037 0.140 Large GG 1 (3.0%) 2 (16.7%) C allele 53 (80.3%) 13 (54.2%) 0.029 0.137 Moderate G allele 13 (19.7%) 11 (45.8%) PPARGC1a GG 17 (51.5%) 6 (50.0%) Gly482Ser AG 11 (33.3%) 6 (50.0%) 0.393 0.055 Small AA 5 (15.2%) 0 (0.0%) G allele 45 (68.2%) 18 (75.0%) 0.611 0.009 None A allele 21 (31.8%) 6 (25.0%) Note: Signi cant effects are bolded. 4. Discussion This preliminary study investigates the frequency of genotypes from a group of candidate genes, which may contribute to the differences in endurance performance of elite AF players. The study further investigated the presence of any associations between candidate variants and performance in the two-kilometre time-trial for elite AF players. This study is the rst to investigate the frequencies of theADRB1 Arg389GlyandPPARGC1a Gly482Servariants in elite AF players. The C allele ofADRB1 Arg389Glyand the G allele of PPARGC1a Gly482Serhad higher frequencies than their respective allele counterparts, with the homozygous genotypes for those alleles also having a greater frequency than the other genotypes. Secondly, the results from the current study found a signi cant association between two-kilometre time-trial performance andADRB1 Arg389GlyandPPARGC1a Gly482Servariants, indicating these may contribute to endurance performance. TheADRB1gene encodes for the beta-adrenergic receptor, with stimulation resulting in the activation and phosphorylation of targeted proteins in cardiac tissue, regulating cardiac function [60–62]. In previous literature, positive associations between the C allele of theADRB1 Arg389Glypolymorphism have been seen in aerobic capacity performance in heart disease populations [52]. Sawczuk

This study investigates genetic influences on endurance performance in AFL players.