Effect Of Aerobic Training And Resistance Training On Circulating .

2016(Kashiwabara et al. 2000). Plasma irisin (catalog#EK067-16) was measured using a commercial EnzymeLinked Immune Sorbent Assays kit (ELISA, PhoenixPharmaceuticals, CA, USA) and a micro-reader.Exercise protocolThe exercise program was executed 5 days/weekduring 8 weeks at the dormitory gymnasium. The aerobictraining program consisted of 60-min sessions of wholebody exercise. The initial duration of each exercise wasgradually progressed to a maximum duration of 50 min,based on the individual’s capabilities. The trainingprogram was performed with 65-80 % of the individualHRmax which was measured at baseline. The subjectsalternated the two modes of aerobic exercise (50 mintreadmill only or the combination of 25 min cycling and25 min treadmill mountain climber) each day during8 weeks.The resistance training program consisted of60-min sessions of exercise for upper or lower body: sixexercises for the upper body parts (shoulder press, seatedrows, lat pull down, bench press, push up, pec deck fly)and five exercises for the lower body parts (squat, legextension, leg curls, leg press, crunch). The upper bodyresistance exercise was performed on Monday andWednesday; the lower body resistance exercise wasperformed on Tuesday and Thursday; and the upper andlower combined resistance training (bench press, lat pulldown, leg curl, and leg press) was performed on Friday.The resistance load was 65-80 % of one repetitionmaximum (1RM) testing which was performed atbaseline and at the end of the study. The subjects wereasked to perform 10-12 repetitions for 3 sets for all upperbody and lower body exercises. Each exercise wasperformed in the moderate contraction velocity (2-3 sconcentric, 2-3 s eccentric). The resting interval betweensets was less than 1 min. Weight was adjustedaccordingly throughout the training program as thesubjects’ strength level gradually increased.Measurement of exercise capacitiesThe isokinetic dynamometer (Humac Norm,CSMi, MA, USA) was used to measure the strength ofthe lower limbs. The isokinetic contraction test was takento evaluate the knee extension and flexion torques of eachlower limb. Prior to testing, participants were asked toperform two familiar practices for the test. After 10 s ofresting, they performed five consecutive cycles ofconcentric knee extension and flexion at a speed of 60 /s.Effect of Aerobic and Resistance Training on Irisin273The grip strength was evaluated by using digital handgrip dynamometer (my-5401, TAKEI, Niigata, Japan).Participants were asked to stand with their arms andwrists in a neutral position. They were trained to performa maximal isometric contraction by pulling the hand gripdynamometer with maximum effort. The test wasrepeated in both arms and the average value of four testswas used for the analysis.In order to determine maximal oxygen uptake(VO2max), Bruce protocol was used on a treadmillergometer as described in a previous study (Tanaka et al.2001). The load was gradually increased by 3 % inclineand 2 km/h after every 3 min from the start. A TreadmillT 150 Cosmed and Quark series by Breath PulmonaryGas Exchange was used for functional diagnostics.Dietary intakeThe participants were required to report a threeday-dietary record during pre-, mid- and post- training.The recording days were comprised of one weekend dayand two weekdays for each three-day-dietary recallperiod. The result of food intake was presented as theamount of total kcal per day. Prior to entering theexercise training program, the nutrition guideline forbalanced meal was provided.Data analysisThe results were expressed as mean SD.Homogeneity of participants among the three groups atbaseline was calculated by One-way Analysis of Variance(ANOVA). The treatment effects (time) x (group) weredetermined by a repeated measure ANOVA. The adjustedp value of less than 0.05 was taken to indicatea significant difference. The differences between final tobasal level of variables were indicated as delta. Toevaluate the correlation among the delta values, Pearson’sr test was conducted. Furthermore, multivariateregression analysis was conducted to adjustment forpotential confounders. All analyses were performed usingSPSS version 18.0 software (SPSS Inc., Chicago, IL,USA).ResultsA total of 28 subjects with a mean age 25.8 yearsand a mean BMI of 26.6 kg/m2 were recruited. Foraerobic and resistance training group, the averageattendance rate in the exercise sessions was 87.4 4.3 %.The characteristics of the study participants before and

274Vol. 65Kim H.-J. et al.Table 1. Characteristics of the study participants at pre- and post-training.Control(total 8,Male 4, Female 4)PrePostAnthropometric characteristicsAge (years)25.8 5.5Height (cm)167.2 5.4 167.5 5.5Body weight (kg)74.1 7.674.1 7.7Skeletal muscle mass32.8 4.928.2 4.8(kg)Skeletal muscle mass0.38 0.00.37 0.0(kg) / Body weight (kg)Body fat mass (kg)23.1 4.023.8 3.4Body fat percentage31.1 5.732.1 5.2(%)BMI (kg/m)26.5 2.026.4 2.1Waist (cm)89.7 4.485.5 6.7Hip (cm)104.2 3.2 104.1 4.2WHR0.86 0.00.82 0.0Subcutaneous fat31.9 7.933.3 3.2thickness (mm)Food intake1509.11 88.47(kcal/day)Blood parametersTotal-C (mg/dl)LDL-C (mg/dl)HDL-C (mg/dl)Triglycerides (mg/dl)FFA (μEq/l)Glucose (mg/dl)Insulin (μU/ml)HOMA-IRAerobic training(total 10,Male 6, Female 4)PrePostResistance training(total 10,Male 7, Female 3)PrePostANOVA25.7 4.1168.3 7.7 168.5 7.875.3 12.7 73.1 13.4*28.4 6.428.4 6.526.4 2.9172.1 5.6172.1 5.780.3 12.8 77.7 13.4*32.5 4.832.9 5.10.4250.1420.0300.2710.37 0.00.38 0.0*0.40 0.00.42 0.0** 0.00123.7 4.832.4 4.822.2 4.8*30.6 4.8**22.1 7.328.0 4.419.5 7.3**24.7 5.1** 0.001 0.00126.4 2.490.0 7.6104.9 4.20.86 0.031.3 9.125.3 2.5**80.9 8.8101.5 5.90.79 0.027.4 6.627.0 3.492.8 8.8105.0 5.90.88 0.028.1 8.726.0 3.6*84.6 10.6102.4 6.40.82 0.025.3 8.20.0350.0690.0770.3510.3431533.92 426.121575.65 227.09204.7 37.6 203.1 34.6 187.4 27.6 182.1 25.8 188.1 23.0 169.0 24.6124.2 28.8 121.7 35.0 112.5 29.5 105.5 23.2 116.7 26.0 101.3 26.761.5 16.6 62.6 16.7 59.9 12.8 63.3 12.155.4 8.554.6 13.0106.2 61.7 120.0 74.9 77.2 44.9 87.0 26.388.2 28.391.3 37.7452.2 181.6 466.1 88.7 435.3 108.5 525.8 306.3 458.8 135.8 426.1 125.484.5 9.190.3 9.486.1 6.385.8 7.090.9 7.687.4 9.27.5 3.813.2 9.16.9 3.39.5 4.39.4 7.68.9 5.91.5 0.73.0 2.21.5 0.82.0 0.92.0 1.41.9 hanges of anthropometric and blood profiles of the subjects were measured before and after 8 weeks of aerobic and resistanceexercise training. Each value represents the mean SE. Total-C, total cholesterol; LDL-C, low-density lipoprotein cholesterol; HDL-C,high-density lipoprotein cholesterol. * p 0.05, ** p 0.01 indicate a significant difference between pre- and post-training values withingroup (paired t-test). ANOVA indicates two way repeated ANOVA measures between group and time.after the exercise training were presented in Table 1.There were no significant differences among the threegroups at the baseline anthropometric variables.However, after 8 weeks of exercise training, significantdifferences were found in body composition includingbody weight, body fat mass, BMI and relative skeletalmuscle mass (adjusted by body weight) following8 weeks of training in both of the exercise groups(Table 1). In addition, there were no significant changesin blood profiles including cholesterols, triglycerides, freefatty acid, and insulin among the three experimentalgroups following exercise training (Table 1). Exercisecapacities including isokinetic leg strength, hand gripstrength and cardiovascular fitness (VO2max) weremeasured at pre- and post-training. There were nostatistical changes in exercise capacities among the three

2016275Effect of Aerobic and Resistance Training on IrisinTable 2. Change of exercise capacities of the study participants.ControlAerobic trainingResistance training(N 8)(N 10)(N 10)PreANOVAPostPrePostPrePostp152.6 61.283.1 30.5142.8 58.281.0 21.1179 60.274.6 28.5157.0 55.169.2 27.9169.8 57.684.4 32.5161.3 59.779.9 28.4180.8 32.6101.4 23.6170.8 33.499.0 23.9193.5 31.3107.8 19.9184.8 28.9102.2 22.00.0730.4390.6920.504Peak torque 60 /s / Body weight (N m/kg)Extensor R207 55.7199.8 64.3Flexor R107.0 41.4 109.3 30.5Extensor L179.5 51.4 186.5 59.4Flexor L103.3 32.1 106.5 32.1232.5 44.796.7 24.2204.4 43.389.5 23.1228.6 46.3112.7 25.5215.1 48.1106.8 20.4226.9 38.7126.5 28.1213.8 34.7124.1 31.4250.5 28.6**138.8 18.2239.6 29.4133.7 24.20.0360.2890.4210.384Hand grip strength (kg) / Body weight (kg)28.9 9.128.4 8.435.2 9.935.1 11.137.1 7.040.6 7.8**0.03134.8 5.437.9 4.9**36.1 2.939.4 5.9*0.023Peak torque 60 /s (N m)Extensor R156.5 55.9Flexor R81.6 37.7Extensor L135.3 49.3Flexor L77.6 32.1VO2max (ml/kg/min)32.9 6.730.7 6.9Exercise capacities including isokinetic leg strength, hand grip strength, and cardiovascular fitness were measured before and after8 weeks of aerobic and resistance exercise training. * p 0.05, ** p 0.01 indicate a significant difference between pre- and posttraining values within group (paired t-test). ANOVA indicates two way repeated ANOVA measures between group and time.Fig. 1. Effect of 8 weeks of aerobic and resistance exercisetraining on circulating irisin level in control versus aerobic andresistance exercise training group. The white bar means preexercise and black bar means post exercise training. * Indicatea significant difference between pre and post values (p 0.05).groups at the baseline. After 8 weeks of training, legextensor muscle (right leg, adjusted with body weight)showed a significant improvement (p 0.001) at 60 /s inresistance training group (Table 2). In addition, hand gripstrength was significantly increased (p 0.001) only inresistance training group (Table 2). The values ofVO2max were improved in both of aerobic training(p 0.001) and resistance training (p 0.05) (Table 2).The concentration of circulating irisin wasmeasured at pre- and post-training. Compared to controlgroup and aerobic training group, level of circulatingirisin in resistance training group was significantlyincreased (p 0.001) following 8 weeks of training(Fig. 1). As shown in Table 3, multivariate linearregression analysis adjusted with potential confoundersincluding age, gender, body composition and food intakewas performed. Significant correlations of irisin withexercise training persisted after adjustment for age,gender, skeletal muscle mass (baseline), fat mass(baseline), and daily calorie intake (Table 3). Thealternation of circulating irisin level was correlated withincrease in skeletal muscles (R 0.432, p 0.022) anddecrease in fat mass (R 0.407, p 0.031) (Fig. 2A, B).This result suggests that change of circulating irisinfollowing exercise training was closely related to positiveimprovement of body composition in overweight/obese.

276Vol. 65Kim H.-J. et al.Table 3. Multivariate regression analysis of anthropometric factors with change of circulating irisin levels as a dependent variable.Aerobic trainingResistance trainingAgeGenderMuscle massFat massFood 590.7510.1130.030*0.754* Indicate significant correlation in multivariate linear regression adjusted with other variables (p 0.05).β1: unadjusted bivariate linear regression coefficient.β2: multivariate linear regression coefficient adjusted for age.β3: multivariate linear regression coefficient adjusted for age, gender.β4: multivariate linear regression coefficient adjusted for age, gender, baseline skeletal muscle mass.β5: multivariate linear regression coefficient adjusted for age, gender, baseline skeletal muscle mass, baseline fat mass.β6: multivariate linear regression coefficient adjusted for age, gender, baseline skeletal muscle mass, baseline fat mass, food intake.DiscussionFig. 2. Association of circulating irisin level with (A) muscle mass(r 0.432, p 0.022), and (B) body fat mass (r 0.407, p 0.031)in 28 overweight/obese adults after 8 weeks of exercise training.In the present study, circulating irisin proteinconcentration was significantly increased in 8 weeks ment of body composition. And we found thatalternation of circulating irisin was highly correlated withincrease in skeletal muscle and decrease in fat mass. Toour knowledge, this is the first study that investigated theeffect of two different types of exercise training oncirculating irisin and its association with bodycomposition changes in overweight/obese adults.Although irisin is known as exercise inducedmyokine, many studies carried out inconsistent results(Hecksteden et al. 2013, Huh et al. 2012, Norheim et al.2014, Pekkala et al. 2013, Timmons et al. 2012).Timmons et al. (2012) reported a significant difference inFNDC5 mRNA only in the older subjects with endurancetraining from their experiment that analyzed gene chipdata from 6 weeks of endurance training and 20 weeks ofstrength training in younger and older groups. From theresult, Timmons et al. (2012) suggested that FNDC5 isnot systematically affected by exercise. However,Boström et al. (2012) pointed out that the gene chipanalyses showed smaller effect sizes and lack of increasein the expression of PGC-1α. Huh et al. (2012)investigated acute and 8-week-lasting sprint training inhealthy adults. Interestingly, the irisin levels wereincreased in acute training but remained unchanged in8 weeks of training. Furthermore, similar to Huh et al.(2012), Norheim et al. (2014) concluded that 12 weeks of

2016endurance training did not enhance the plasmaconcentration of irisin in an older group. They found thatthe expression of UCP1 mRNA is not correlated with theFNDC5 expression in subcutaneous adipose tissue orskeletal muscle. Hecksteden et al. (2013) also found anunchanged irisin level after 26 weeks of aerobic andstrength endurance training. However, the interpretationof the study results was difficult. First of all, the trainingprotocol (four or six bouts of 80 m sprints) in Huh et al.(2012) was an unusual preventive training that is differentfrom the original work of Boström et al. (2012).Moreover, Norheim et al. (2014) found that increasedFNDC5 mRNA by exercise was not translated into anincrease in irisin circulation. Hecksteden et al. (2013)suggested that a degeneration of irisin level is possible infrozen samples. In this regard, our data with anunchanged circulating irisin level in aerobic exercisetraining showed a similar pattern to the previous studies.However, considering the key role of irisin as a myokine(cytokine which is produced, expressed and released byskeletal muscle), the result of a changed circulating irisinlevel in the resistance exercise training group seemsmeaningful. The result is similar to that in acuteresistance training in obesity group (Pekkala et al. 2013).An interesting point is that the circulating irisin level wasnot changed by chronic strength training in healthy men(Hecksteden et al. 2013). Recently, Loffler et al. (2015)reported that increase irisin levels after acute strenuousexercise and 30-min bout of intensive exercise in childrenand young adults, whereas longer or chronic increases inphysical activity did not affect irisin levels. There weredifferences compared to our experiments: 1) subjects(they recruited school children, whereas we recruitedyoung adults), 2) intensity of exercise (they applied lowintensity physical activity, whereas we applied moderatevigorous intensity supervised training). Therefore, futurestudies are needed to investigate the effect of resistanceexercise on circulating irisin in obese and healthy humanswith the same study design.To reveal an association between circulatingirisin and obesity, several researchers investigated thecorrelation of circulating irisin with obesity parameters inhumans. Huh et al. (2012) and Stengel et al. (2013)reported that circulating irisin showed a positivecorrelation with BMI. The researchers speculated that thepositive association might be a compensatory mechanism.On the other hand, Choi et al. (2013) reported thatcirculating irisin showed a negative correlation with BMI.The differences among the studies might come from theEffect of Aerobic and Resistance Training on Irisin277diversity in subjects’ characteristics and age. Although,several studies investigated the correlation of circulatingirisin level with obesity, there is not enough research thatshows different contribution by different type of exerciseto the circulating irisin level in obesity. According to therecent studies by Boström et al. (2012) and Kelly (2012)irisin is likely to play a pivotal role in browning of whitefat cell leading to heat production and energyexpenditure. It also suggested that exercise inducesmuscle FNDC5 expression causing increased circulatingirisin levels. Therefore, it is speculated that in our study,exercise training might have decreased fat mass and fatpercentage which were potentially mediated by increasedcirculating irisin levels. In this study it could be noticedthat circulating irisin level has a correlation with positiveimprovement of body composition in overweight/obese(Fig. 2). However, there is still lack of study to elucidatethe complex association among exercise training,exercise-induced irisin levels, and change of bodycomposition.Limitations to the current study include: 1) smallsample size and lack of consideration for genderdifference, 2) high drop-out rate, 3) analysis of bodycomposition by BIA, and 4) lack of evidence of browningmarker. Firstly, our data have a lack of consideration forgender differences among the three groups. This limitationalso includes the hormonal and metabolism influence indifferent gender. We additionally compared the genderdependent effect on irisin levels and there was nosignificant difference in men and women at baseline andtheir response to the training (data not shown). At thebaseline, we designed the group by similar ratio of genderbut the drop-out caused the gender differences at the final.Also, high drop-out rate was the limitation of the study.Total number of drop-out in each group was as follows;aerobic group (total 7, male 5, female 2) and resistancegroup (total 8, male 2, female 6). We compared the basiccharacteristics of participants and drop-out subjects in eachtraining group. However, there was no significantdifference between drop-out participants and trainingparticipants. Secondly, although several researchesreported that BIA has a good agreement with DXA inmeasuring fat free mass, fat mass and body fat percentage(Fox et al. 1996, Pratley et al. 2000), BIA alone could notprovide conclusive information of body composition thatDual Energy X-ray Absorptiometry (DEXA) couldprovide. Lastly, we were limited to estimate browningmarker such as uncoupling protein 1 (UCP1), PR domaincontaining 16 (PRDM16) and other cytokines which might

278Vol. 65Kim H.-J. et al.be related to circulating irisin level.In conclusion, the findings of our researchprovided an insight for an approach to increasecirculating irisin level in overweight/obese adults withresistance exercise training. In addition, although anassociation between exercise-induced irisin and bodycomposition change was found in obese adults, we couldnot evaluate their causal relationship in this study.Therefore, the future study should cover the role ofexercise-induced irisin on body composition includingmuscle mass and fat mass.Conflict of InterestThere is no conflict of interest.AcknowledgementsThis work was supported by National ResearchFoundation of Korea (NRF) funded by Ministry , NRF-2013M3A9D5072550, andMEST 2011-0030135) and Ministry of Education (NRF2014R1A1A2058645).ReferencesAYDIN S, KULOGLU T, AYDIN S, EREN MN, CELIK A, YILMAZ M, KALAYCI M, SAHIN I, GUNGOR O,GUREL A, OGETURK M, DABAK O: Cardiac, skeletal muscle and serum irisin responses to with or withoutwater exercise in young and old male rats: cardiac muscle produces more irisin than skeletal muscle. Peptides52: 68-73, 2014.BOSTROM P, WU J, JEDRYCHOWSKI MP, KORDE A, YE L, LO JC, RASBACH KA, BOSTROM EA, CHOI JH,LONG JZ, KAJIMURA S, ZINGARETTI MC, VIND BF, TU H, CINTI S, HOJLUND K, GYGI SP,SPIEGELMAN BM: A PGC1-alpha-dependent myokine that drives brown-fat-like development of white fatand thermogenesis. Nature 481: 463-468, 2012.CHO KO, JO YJ, SONG BK, OH JW, KIM YS: Colon transit time according to physical activity and characteristics inSouth Korean adults. World J Gastroenterol 19: 550-555, 2013.CHOI YK, KIM MK, BAE KH, SEO HA, JEONG JY, LEE WK, KIM JG, LEE IK, PARK KG: Serum irisin levels innew-onset type 2 diabetes. Diabetes Res and Clin Pract 100: 96-101, 2013.FOX AA, THOMPSON JL, BUTTERFIELD GE, GYLFADOTTIR U, MOYNIHAN S, SPILLER G: Effects of dietand exercise on common cardiovascular disease risk factors in moderately obese older women. Am J Clin Nutr63: 225-233, 1996.HECKSTEDEN A, WEGMANN M, STEFFEN A, KRAUSHAAR J, MORSCH A, RUPPENTHAL S, KAESTNER L,MEYER T: Irisin and exercise training in humans – results from a randomized controlled training trial.BMC Med 11: 235, 2013.HUH JY, PANAGIOTOU G, MOUGIOS V, BRINKOETTER M, VAMVINI MT, SCHNEIDER BE, MANTZOROSCS: FNDC5 and irisin in humans: I. Predictors of circulating conc

aerobic and resistance training, showed significant improvement in anthropometric parameters and exercise capacities including maximal oxygen uptake and muscle strength. Interestingly, the circulating irisin was significantly increased in resistance training group (p 0.002) but not in aerobic training (p 0.426) compared to control group.