Effects of exercise and dietary intervention on serum metabolites in men with insomnia symptoms : a 6-month randomized-controlled trial

This is a sel -a chi ed e sion o an o iginal a icle. This e sion may di e om he o iginal in pagina ion and ypog aphic de ails. Au ho (s): Ti le: Yea : Ve sion: Copy igh : Righ s: Righ s u l: Please ci e he o iginal e sion: CC BY 4.0 h ps://c ea i ecommons.o g/licenses/by/4.0/ E ec s o exe cise and die a y in e en ion on se um me aboli es in men wi h insomnia symp oms : a 6-mon h andomized-con olled ial © 2020 he Au ho s Published e sion Zhang, Xiaobo; Wang, Xiuqiang; Le, Shenglong; Ojanen, Xiaowei; Tan, Xiao; Wiklund, Pe i; Cheng, Sulin Zhang, X., Wang, X., Le, S., Ojanen, X., Tan, X., Wiklund, P., & Cheng, S. (2020). E ec s o exe cise and die a y in e en ion on se um me aboli es in men wi h insomnia symp oms : a 6- mon h andomized-con olled ial. Spo s Medicine and Heal h Science, 2(2), 95-101. h ps://doi.o g/10.1016/j.smhs.2020.05.002 2020 O iginal Resea ch E ec s o exe cise and die a y in e en ions on se um me aboli es in men wi h insomnia symp oms: A 6-mon h andomized con olled ial Xiaobo Zhang a , b , c , Xiuqiang Wang b , c , Shenglong Le b , c , d , Xiaowei Ojanen a , b , Xiao Tan e , Pe i Wiklund b , c , Sulin Cheng b , c , d , * a School o Li e Sciences and Bio echnology, Shanghai Jiao Tong Uni e si y, Shanghai, China b Key Labo a o y o Sys ems Biomedicine Minis y o Educa ion, and Exe cise T ansla ional Medicine Cen e , Shanghai Cen e o Sys ems Biomedicine, Shanghai Jiao Tong Uni e si y, Shanghai, China c Exe cise, Heal h, and Technology Cen e , Depa men o Physical Educa ion, Shanghai Jiao Tong Uni e si y, Shanghai, China d Facul y o Spo and Heal h Sciences, Uni e si y o Jy € askyl€ a, Jy € askyl€ a, Finland e Depa men o Neu oscience, Uppsala Uni e si y, Uppsala, Sweden ARTICLE INFO Keywo ds: Die Exe cise Insomnia Me abolomics ABSTRACT Accumula ing e idence show ha exe cise and die in e en ions a e associa ed wi h imp o ed sleep quali y. S udies in es iga ing he e ec s o exe cise and die ing on ci cula ing me abolomics in people wi h sleep diso - de s, pa icula ly insomnia, a e sca ce. This 6-mon h andomized s udy aimed o assess he e ec s o exe cise and die a y in e en ions on se um me aboli es in men wi h insomnia symp oms. Se en y- wo Finnish men (age: 51.6 10.1 yea s) wi h ch onic insomnia symp oms who we e assigned o di e en in e en ion g oups comple ed his s udy (exe cise, n¼24; die , n¼27; and con ol, n¼21). The Shapi o-Wilk W- es , Le ene es , Spea man co ela ion analysis, and analysis o a iance we e used o da a analysis. We ound ha exe cise and die in e en ion we e associa ed wi h imp o ed sleep quali y and wi h a numbe o me aboli es ac oss di e en biochemical pa hways. Al hough we could no show causali y, ou findings p o ide new insigh in o he biological mechanisms unde lying he heal h e ec s o physical ac i i y, die , and sleep quali y. Fu he in es iga ion is needed o be e unde s and he link among li es yle, sleep quali y, and me abolic heal h. In oduc ion Insomnia, he mos p e alen sleep diso de , is common in mode n socie y and has become an impo an public heal h issue. 1 The p e a- lence o insomnia symp oms is 21.4% in adul s om he Uni ed S a es, as assessed by c i e ia om he Diagnos ic and S a is ical Manual o Men al Diso de s, ou h edi ion. 2 Fu he mo e, sleep loss o sleep dis up ions induced by insomnia symp oms a e also associa ed wi h an inc ease in all-cause mo ali y and can lead o de imen al e ec s on neu oendoc ine sys ems. 3 To da e, he s udies conce ning he e ec o exe cise in- e en ions on sleep ha e ocused on good sleepe s, 4,5 fi a hle es, 6,7 o pa ien s wi h sleep apnea. 8,9 The e ec s o exe cise on sleep in an un- ained popula ion and insomnia ha e ye o be es ablished. 10 We ha e shown ha 6 mon hs o exe cise and die in e en ion sho ened objec i e sleep onse la ency (SOL) and imp o ed subjec i e sleep quali y in o e weigh and obese men wi h insomnia symp oms, independen o weigh loss. 11,12 Howe e , hese s udies did no in es iga e he me abolic changes associa ed wi h imp o ed sleep quali y. The esul s o die a y in e en ions a ge ing pa ien s wi h sleep diso de s a e conflic ing, 13 wi h wo s udies showing ha isocalo ic high a in ake is associa ed wi h significan ly be e sleep han o he die s in humans. 11,14 A la ge c oss-sec ional s udy om China epo ed an asso- cia ion be ween dec eased sleep du a ion and inc eased a in ake in 2828 adul s. 15 Howe e , he biological mechanisms unde lying he as- socia ions be ween exe cise/die and sleep diso de s la gely emain unclea . Se um high- h oughpu me abolomics p ofiling is a easible me hod o s udy he cha ac e is ic changes in small molecula me aboli es in a Abb e ia ions: SOL, sleep onse la ency; BMI, body mass index; TST, o al sleep ime; SE, sleep e ficiency; GC, gas ch oma og aphy; MS, mass spec ome y; NIST, Na ional Ins i u e o S anda ds and Technology; (O)PLS-DA, (o hogonal) pa ial leas -squa es-disc iminan analysis; VIP, a iable impo ance in he p ojec ion; FM, a mass; KIC, alpha-ke oisocap oic acid; 3-HB, 3-hyd oxybu ic acid; BCAA, b anched chain amino acid; Q2, qua ile 2. * Co esponding au ho . 800 Dongchuan RD. Minhang Dis ic , Shanghai, 200240, China. E-mail add esses: sulin.cheng@jyu.fi,shulin.cheng@jyu.fi(S. Cheng). Con en s lis s a ailable a ScienceDi ec Spo s Medicine and Heal h Science jou nal homepage: www.keaipublishing.com/en/jou nals/spo s-medicine-and-heal h-science/ h ps://doi.o g/10.1016/j.smhs.2020.05.002 Recei ed 14 Ap il 2020; Recei ed in e ised o m 19 May 2020; Accep ed 19 May 2020 A ailable online 22 May 2020 2666-3376/©2020 Chengdu Spo Uni e si y. P oduc ion and hos ing by Else ie B.V. on behal o KeAi. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/). Spo s Medicine and Heal h Science 2 (2020) 95–101 pa hophysiological s a e. S udies ha e shown ha le els o glucose, amino acids, a y acids and ica boxylic acid cycle in e media es di e ed in indi iduals wi h sleep diso de s compa ed wi h heal hy people 16–18 and in animal s udies. 19,20 Howe e , no s udy has examined he e ec o ei he exe cise o die in e en ion on ci cula ing me ab- olomics in pa ien s wi h insomnia symp oms. We hypo hesize ha ex- e cise and die a y in e en ions a ec he sleep quali y o pa ien s wi h insomnia symp oms h ough di e en me abolic pa hways. The e o e, we conduc ed he p esen s udy o assess he associa ions be ween se um me aboli es and sleep quali y by conduc ing andomized con olled 6-mon h exe cise o die in e en ion in men wi h insomnia symp oms. Ma e ial and me hods E hics s a emen s The s udy was app o ed by he E hics Commi ee o he Cen al Finland Heal h Ca e Dis ic (7/2011 OTE) and egis e ed a www.con o lled- ials.com (ISRCTN77172005). W i en in o med consen was ob- ained om all pa icipan s be o e he baseline measu emen s, and copies o he signed consen o ms we e a chi ed. S udy design and popula ion This s udy was pa o a 6-mon h andomized li es yle in e en ion in middle-aged Finnish men wi h sleep diso de s (including insomnia and sleep apnea). 21 Fo his epo , we included 72 men aged 30–65 yea s who sel - epo ed su e ing om ch onic insomnia las ing 3 mon hs o longe and comple ed he 6-mon h in e en ion. 21 The pa icipan s u he comple ed a Vi almed sleep ques ionnai e, which consis s o all he ques ions in he Basic No dic Sleep Ques ionnai e and Epwo h Sleepiness Scale, and a physician e iewed hei esponses. O he qual- ified pa icipan s who comple ed he in e en ion, 24 pa icipan s we e andomized o he exe cise g oup, 27 o he die g oup, and 21 o he con ol g oup. Exe cise in e en ion Exe cise was p esc ibed acco ding o he ecommenda ions o he Ame ican College o Spo s Medicine. 22 No dic walking combined wi h o he ae obic exe cises was pe o med o 30–60 min pe session, 3 o 5 sessions pe week, a he in ensi y le el o 60–75% o he es ima ed maximum hea a e. Two expe ienced exe cise aine s we e esponsible o he ins uc ion and supe ision o exe cise aining once a week and in he o he sessions, he pa icipan s exe cised independen ly. A he beginning o each week, he aine in o med he pa icipan o he du a ion and in ensi y (hea a e in e al) o each exe cise session. 12 Die in e en ion A specific indi idualized die p og am was de eloped a e baseline assessmen s o each pa icipan 's cu en die a y in akes (based on a 3- day ood dia y) and body weigh we e pe o med. The sugges ed die con ained he ollowing ela i e mac onu ien composi ion o ene gy in ake: 40% ca bohyd a es wi h <5% suc ose, 40% a (sa u a ed a y acid 10%, monounsa u a ed a y acid 15–20%, and polyunsa u a ed a y acid 10%), and 20% p o ein. O e weigh /obese pa icipan s we e ad ised o mode a ely educe hei o al ene gy in ake (by 300–500 kcal pe day o he fi s 3 mon hs) wi h guidance on he p opo ion o mac onu ien s o be consumed. The a ge was o educe hei body weigh by 3 kg in he fi s 3 mon hs o he in e en ion. De ailed in o - ma ion was gi en in p e ious pape . 11 Du ing he in e en ion, an online die and nu i ion counseling se - ice (Meal T acke ; Wellness Found y Holding L d., Helsinki, Finland) was u ilized o supe ise pa icipan s’die a y in ake and p o ide die sugges ions. Pa icipan s we e ins uc ed o pho og aph all daily die a y in akes (including d inks) by using a sma phone o a digi al came a, and upload all pho og aphs o he se e 1–3 days pe week du ing he in e en ion. The a e age o pho o uploading in die g oup was 26 im- ages pe mon h (37 o mon h 1, 35 o mon h 2, 26 o mon h 3, 21 o mon h 4, 24 o mon h 5 and 14 o mon h 6). Con ol g oup Pa icipan s in he con ol g oup we e ins uc ed o main ain hei habi ual, p e- ec ui men li es yle du ing he in e en ion. They we e in i ed o a lec u e ha explained he pu pose o hei g oup. A e a 6- mon h s udy pe iod, hey we e gi en an oppo uni y o pa icipa e in he exe cise plus die a y in e en ion p og am o 3 mon hs. Backg ound in o ma ion and an h opome ic measu emen s All li es yle and medical his o y in o ma ion was collec ed by sel - adminis e ed ques ionnai es a he Labo a o y o Spo and Heal h Sci- ences, Uni e si y o Jy € askyl€ a. An h opome ic measu emen s we e pe o med a e o e nigh as ing (12 h). Heigh was measu ed using a fixed wall scale de ice o he nea es 0.1 cm. Weigh was de e mined o he nea es 0.1 kg using an elec onic scale and calib a ed be o e each measu emen session. Body mass index (BMI) was calcula ed as weigh (kg) pe heigh 2 (m 2 ). Objec i e measu emen s o sleep Se en nigh sleep measu emen s be o e and a e he in e en ion we e pe o med o all pa icipan s by using a non-con ac sleep moni- o ing sys em a hei homes (Beddi sleep acke ; Beddi , Espoo, Finland) 23 and au oma ically analyzed using he Beddi se e ia he In e ne . The sys em included a piezoelec ic bed senso . Ballis o- ca diog aphic signals we e sampled by he piezoelec ic senso a 140 Hz and simul aneously uploaded o a web se e ia he In e ne , whe e sleep/wake s a us was classified in 30-s epochs based on hea a e a iabili y, espi a o y a e a iabili y, and a bina y ac ig am. To al sleep ime (TST), SOL (de e mined as he du a ion om being p esen in bed wi h ligh s ou o he fi s 5 min o consecu i e sleep), wake ulness a e sleep onse , and sleep e ficiency (SE) (commonly defined as he a io o a pe son's TST o ime spen in bed, which e e s o he amoun o ime a pe son is ac ually asleep du ing he ime spen ying o sleep) we e ob ained o each nigh . A e age alues o e he nigh s measu ed we e used in he analyses. The possible condi ions ha may a ec he mea- su emen , such as child en and pe s in he bed oom we e eco ded in he pa icipan 's sleep dia y. A esea ch assis an isi ed each pa icipan 's home o se up he sys em be o e measu emen s we e s a ed. The sleep da a we e alida ed by he piezoelec ic sys em agains he wo-nigh polysomnog aphy measu emen (31 pa icipan s wi h complain s o insomnia, mean age: 51.8 8.4 yea s, mean BMI:30.9 4.8 kg/m 2 ). 12 Measu emen o he maximal oxygen up ake The 2-km walk es (UKK Ins i u e, Tampe e, Finland) was used o de e mine each pa icipan 's fi ness le el. Pa icipan s we e ins uc ed o walk 2 km as as as possible a a s eady pace. The walk ime du a ion and hea a e immedia ely upon finishing he walk we e measu ed o es i- ma e he maximal ae obic powe (es ima ed maximal oxygen up ake). An exe cise wa ch wi h a hea a e moni o ing bel (M5; Suun o L d., Van aa, Finland) was used o de e mine pa icipan s' hea a e. The es s we e conduc ed a on an ou doo a hle ic field and conside ed/p o en sa e o he o e weigh and obese adul s who me ou inclusion c i e ia. 24 Se um sample collec ion Blood samples we e collec ed om all subjec s in he mo ning be- ween 7:00 and 9:00 a.m. a e o e nigh as ing a baseline and a e he X. Zhang e al. Spo s Medicine and Heal h Science 2 (2020) 95–101 96 6-mon h in e en ion. Se um was ex ac ed by cen i uga ion and s o ed immedia ely a 80 C un il analyzed. Me abolomics assessmen Sample p epa a ion The se um sample (100 μ L) was fi s hawed a oom empe a u e, hen mixed wi h a me hanol-chlo o o m sol en (300 μ L 3:1, / ) and L- 2-chlo ophenylalanine (10 μ L, 0.3 mg/mL in wa e ) o me aboli es ex ac ion, and kep a 20 C o 10 min. A e cen i uga ion a 15 000 g o 10 min, 300 μ L o supe na an was ob ained and d ied comple ely unde ni ogen. Nex , 80 μ L o me hoxyamine (15 mg/mL in py idine) was added o he ial, o exed o 30 s and kep a 37 C o 90 min ollowed by he addi ion o 80 μ L N,O-Bis( ime hylsilyl) i- fluo oace amide (1% ime hylchlo osilane) and 20 μ L o n-hexane a 70 C o 60 min. Quali y con ol samples o se um we e p epa ed by pooling equal aliquo s o each sample and hen hey we e p e ea ed as se um samples. Pegasus 4-dimensional gas ch oma og aphy ime-o -fligh mass spec ome y analysis Each 1 μ L aliquo o he de i a ized solu ion was injec ed in spli less mode in o he Pegasus 4-dimensional gas ch oma og aphy (GC) ime-o - fligh mass spec ome y (MS) sys em (LECO Co po a ion, S Joseph, MI, USA). Sepa a ion was pe o med on a non-pola DB-5 capilla y column (30 m 250 μ m I.D., J&W Scien ific, Folsom, CA, USA), wi h high pu i y helium as he ca ie gas a a cons an flow a e o 1.0 mL/min. The GC empe a u e p og amming was se o begin a 90 C and hold o 0.2 min, ollowed by 10 C/min o en empe a u e amps o 180 C, 30 C/min o 240 C, 20 C/min o 280 C, and a final 11-min main enance a 280 C. The empe a u es o he injec ion and ion sou ce was se o 280 C and 220 C, espec i ely. Elec on impac ioniza ion (70 eV) a ull scan mode (m/z30–600) was used (sol en delay: 7.6 min), wi h an acquisi- ion a e o 100 spec um/second in he MS se ing. The acqui ed MS da a om GCMS we e analyzed using he Ch o- maTOF so wa e ( e sion 4.34, LECO, S Joseph) combined wi h Na ional Ins i u e o S anda ds and Technology (NIST) mass spec al lib a ies (NIST 14), LECO/Fiehn Me aboli e mass spec al lib a y ( e sion 1.00) and ou in-house lib a y. Based on he e en ion ime o a se ies o a y acid me hyl es e s (C6–C24), he e en ion index was calcula ed by he Re en ion Index Me hod unc ion o he Ch omaTOF so wa e. 25 Da a p ocessing and s a is ics SPSS s a is ical so wa e e sion 24 (IBM Co p., A monk, NY, USA) was used o pe o m s a is ical analysis. All da a we e checked o no mali y by using he Shapi o-Wilk W- es and o homogenei y by using he Le ene es be o e each analysis. Na u al loga i hm ans- o ma ions we e pe o med on non-no mally dis ibu ed da a. Spea man co ela ion analysis was pe o med wi h he change pe cen age o me- aboli es and change pe cen age o pa ame e s ela ed o sleep. Be ween- g oup di e ences we e e alua ed by epea ed measu es analysis o a iance, ollowed by Sidak co ec ions o mul iple compa isons. All es s we e wo-sided wi h a significance le el se a p<0.05. To accoun o he e oscedas ici y, da a we e no malized by log 2 ans o ma ion. Mul i a ia e s a is ical analysis and uni a ia e analysis we e used o me abolomics da a analysis. Da a se s esul ing om GC–MS we e sepa a ely impo ed in o he SIMCA-Pþ14.0 so wa e package (Ume ics, Umea, Sweden). P inciple componen analysis and (o hogonal) pa ial leas -squa es-disc iminan analysis ((O)PLS-DA) we e pe o med wi h SIMCA-P, and di e en ial me aboli es we e based on a iable impo ance in he p ojec ion (VIP) > 1.0 and p <0.05 acco ding o he S uden - es . In he p esen s udy, a 7- old c oss alida ion p ocedu e and 200 andom pe mu a ions es s we e pe o med o a oid o e fi ing o he supe ised (O)PLS-DA models. The in ensi y o each me aboli e was used as he final esul . Resul s Cha ac e is ics o he s udy pa icipan s a e shown in Table 1. Du ing he 6-mon h in e en ion pe iod, body weigh and a mass (FM) we e significan ly inc eased in he con ol g oup, while no significan changes in body weigh and FM we e obse ed in he in e en ion g oups. Howe e , compa ed o he con ol g oup, he die g oup had significan ly low body weigh and FM (p¼0.005 and p¼0.027, espec i ely), whe eas in he exe cise g oup did no show such di e ences. O he sleep quali y ela ed a iables, SOL dec eased (p<0.05) a e he exe cise in e en ion, and SE inc eased (p<0.05) a e he die in e en ion. No significan changes in sleep- ela ed pa ame e s we e ound in he con ol g oup (Fig. 1). Two hund ed en known me aboli es in se um we e obse ed among he di e en g oups. The esul s showed a clea sepa a ion be ween he baseline and ollow-up findings (exe cise g oup R2Y ¼0.959, qua ile 2 [Q2] ¼0.675; die g oup R2Y ¼0.969, Q2 ¼0.823; and con ol g oup R2Y ¼0.981, Q2 ¼0.828) (Supplemen a y Fig. S1). Figs. S1B and S1D, and S1F show ha he calcula ed R2 and Q2 alues in pe mu a ion es s we e lowe han he o iginal ones, and he Q2 in e cep on he e ical axis was <0. The e o e, he model was conside ed alid. The significan me aboli es in se um we e sc eened acco ding o he VIP alue la ge han 1 and p- alue (p<0.05) om he S uden - es adjus ed by he alse disco e y a e. Twen y-one me aboli es we e a ec ed by he exe cise in e en ion, 33 me aboli es we e a ec ed by he die in e en ion, and 5 we e a ec ed in he con ol g oup (VIP >1, p<0.05). All me aboli es, excep ace anilide in he exe cise and con ol g oups, showed g ea e han 2- old change be ween baseline alues a and 6 mon hs (Table S1). The significan ly al e ed me aboli es in all g oups a e shown in Fig. 2A, B and 2C. Fig. 2D shows 2 unique me ab- oli es ha we e al e ed in he exe cise g oup and 14 me aboli es ha we e al e ed in he die g oup. Nex , we assessed whe he hese me aboli es we e associa ed wi h changes in sleep pa ame e s (Fig. 3 and Table 2). We ound ha dec eased alpha-ke oisocap oic acid (KIC) was co ela ed wi h imp o ed SE in he exe cise g oup ( ¼0.52, p¼0.026). Addi ionally, 67% o pa icipan s showed inc eased phospha e le els a e he exe cise in e - en ion, al hough he associa ion be ween phospha e and SE was no s a is ically significan . In he die g oup, change o 3-hyd oxybu ic acid (3-HB) ( ¼0.47, p¼0.025) and D-glucopy anose ( ¼0.54, p¼0.006) co ela ed nega i ely wi h SE. Howe e , in he same g oup, oxalic acid ( ¼0.49, p¼0.021), D-glucopy anose ( ¼0.43, p¼0.048), 4-deoxye y h onic acid ( ¼0.60, p¼0.004) and aga ose ( ¼0.51, p¼0.016) co ela ed posi i ely wi h change o SOL, and 2-ke o-iso ale- ic acid ( ¼0.45, p¼0.029) co ela ed wi h TST. Discussion In his s udy, 21 me aboli es we e al e ed significan ly a e he ex- e cise in e en ion and 33 me aboli es we e al e ed a e he die a y in e en ion. Fi e me aboli es changed significan ly in he con ol g oup du ing he same pe iod. SE is commonly defined as he a io o a pe son's TST o ime spen in bed. 26 Ea lie s udies using sel -adminis e ed ques ionnai es showed ha sleep iming is associa ed wi h me abolic al e a ions in di e en pa h- ways. 27 He ein, we used an objec i e measu e o sleep and a non-con ac sleep moni o ing sys em a pa icipan s' homes. We ound ha he significan ly al e ed me aboli es a e he exe cise in e en ion we e mainly ela ed o amino acids me abolism (isoleucine, L-allo h eonine, and glu amine), ca bohyd a es me abolism (4-deoxye y h onic acid, e y h ose, aga ose, ucose, D-glucopy anose, and ehalose-6-phospha e) and lipids me abolism (bu y ic acid, 2-ke o-iso ale ic acid, monoolein, be a-glyce ophospho ic acid, and 2-monopalmi in). We ound ha dec eased KIC was co ela ed wi h imp o ed SE. KIC is eadily amina ed o leucine ia leucine dehyd ogenase and/o b anched chain amino acid (BCAA) ans e ase a he expense o NH3 in a ious issues including X. Zhang e al. Spo s Medicine and Heal h Science 2 (2020) 95–101 97 skele al muscle, 28 which may ul ima ely dec ease a igue and p ese e skele al muscle o ce p oduc ion du ing in ense exe cise. Howe e , o ou knowledge no p e ious s udy has linked exe cise and KIC wi h sleep quali y. 29 BCAAs con ibu e o ene gy me abolism du ing exe cise as ene gy sou ces and subs a es o expand he pool o ci ic acid-cycle in- e media es, hus allowing mo e e ficien mi ochond ial ene gy p oduc- ion om lipids and he eby be e me abolic heal h. 30 A s udy conduc ed in o e weigh dysglycemic and no mal weigh no moglycemic men ound ha insulin sensi i i y inc eased simila ly in dysglycemic and no moglycemic men a e 12 weeks o exe cise, in pa allel o simila inc eases in he concen a ion o plasma glu amine. 31 Howe e , o he p e ious s udies showed ha a dec ease in he mola sum o ci cula ing BCAAs was he bes p edic o o imp o emen in insulin sensi i i y induced by a combined esis ance and ae obic exe cise in e en ion in o e weigh indi iduals. 32 Long- e m leisu e- ime physical ac i i y was associa ed wi h low ci cula ing BCAAs concen a ions. 33 Because o he disc epancy be ween ou s udy esul s and he esul s o o he s udies, mo e s udies a e needed o be e unde s and how exe cise a ec s amino acid me abolism ela ed o insomnia symp oms. On he o he hand, he e ec o exe cise and sleep on ca bohyd a es me abolism 34,35 and lipids me abolism 36–38 ha e been epo ed p e iously. Exe cise appea s o enhance he abili y o skele al muscles o u ilize lipids as opposed o glycogen, hus educing plasma lipid le els. 39 Se e al s udies also showed ha poo e o longe sleep du a ion we e associa ed wi h highe isks o abno mal se um lipid p ofiles. 40–42 Fu he s udies a e needed o explo e whe he he al e a ions o ene gy- ela ed amino acids, ca bo- hyd a es, and lipids me abolisms associa ed wi h exe cise can ansla e o imp o ed sleep quali y. The die a y in e en ion was associa ed wi h al e ed concen a ions o 33 me aboli es. Among hose me aboli es, 7 we e co ela ed wi h imp o ed sleep quali y. These me aboli es a e mainly ela ed o ca bo- hyd a es me abolism (4-deoxye y h onic acid, aga ose, and D-gluco- py anose), lipids me abolism (2-ke o-iso ale ic acid) and o ganic acids me abolism (3-HB, and oxalic acid). Inc eased oxalic acid, D-glucopy - anose, 4-deoxye y h onic acid and aga ose we e associa ed wi h imp o ed SOL. Oxalic acid (oxala e) is de i ed om die , he deg ada ion o asco ba e, and p oduced by he li e and e y h ocy es. Oxalic acid has been shown o be educed in a s and humans ollowing sleep es ic ion, and eco e o nea baseline le els a e sleep es ic ion. 43 Mo eo e , he 3-HB concen a ion inc eased a e he die a y in e en ion. Se um 3-HB, which is a ke one body, 44,45 has been shown o inc ease in subcu aneous adipose a e sleep loss in humans. 46 We also ound ha D-glucopy anose inc eased a e he die in e en ion. D-glucopy anose (glucose) is a monosaccha ide con aining 6 ca bon a oms and an aldehyde g oup and is he e o e e e ed o as an aldohexose, which is a p ima y sou ce o Table 1 Cha ac e is ics o subjec s (mean SD). Exe cise (n¼24) Die (n¼27) Con ol (n¼21) Be ween-g oup Baseline 6 mon hs Baseline 6 mon hs Baseline 6 mon hs 1 s 3 2 s 3 Age (yea s) 51.2(9.7) –51.7(9.3) –50.7(9.3) – Weigh (kg) 92.3(14.6) 92.5(13.9) 93.8(11.8) 92.7(11.9) 93.1(17.2) c 94.4(17.8) 0.176 0.005 BMI 29.3(4.0) 29.4(3.8) 29.4(3.7) 29.0(3.7) 29.2(4.4) 29.4(4.6) 0.612 0.047 To al a mass (kg) 27.3(9.5) 27.2(9.0) 27.4(8.4) 26.8(8.6) 28.0(9.7) c 28.9(10.7) 0.180 0.027 T iglyce ides (mmol/L) 1.7(0.7) 1.8(0.9) 1.6(0.6) 1.5(0.6) 1.5(0.7) 1.4(0.6) 0.104 0.867 Ene gy in ake (kcal) 2285.2(561.6) 2043.1(643.6) 1981.2(372.1) 1877.1(487.9) 2223.5(410.1) c 1920.9(463.7) 0.588 0.606 VO2max 29.3(9.3) a 31.4(9.3) 27.2(9.6) b 29.6(10.8) 30.9(7.9) 30.4(8.0) 0.086 0.053 To al ene gy expendi u e (MET min/day) 2349.9(224.8) 2335.5(235.3) 2346.1(215.9) 2398.8(234.9) 2341.1(242.8) 2322.6(233.7) 0.887 0.282 a: exe cise 0 m compa e wi h exe cise 6 m; b: die 0 m compa e wi h die 6 m; c: con ol 0 m compa e wi h con ol 6 m, p<0.05. Be ween-g oup (1: exe cise, 2: die , 3: con ol) di e ences we e e alua ed by epea ed measu es ANOVA, ollowed wi h Sidak co ec ions o mul iple compa isons. Fig. 1. Changes o sleep pa ame e s a e he in e en ions. TST, o al sleep ime; SOL, sleep onse la ency; WASO, wake ulness a e sleep onse ; SE, sleep e ficiency. X. Zhang e al. Spo s Medicine and Heal h Science 2 (2020) 95–101 98 ene gy o li ing o ganisms. Abno mal glucose me abolism ha e been linked o dis u bances o di e en aspec s o sleep, including sleep du a ion and quali y, espi a o y unc ion du ing sleep, and ci cadian iming. 47 Sho and long sleep du a ions we e associa ed wi h wo se glucose con ol in people wi h diabe es. 48 Recen ly, Kay e al.showed ha good sleepe s, bu no pa ien s wi h insomnia, had lowe whole-b ain glucose me abolism du ing eco e y non- apid eye mo e- men sleep compa ed o baseline findings and indi iduals wi h insomnia. 49 We also ound inc eased aga ose and KIV a e he die in e en ion. Taga ose (D- aga ose) has been associa ed wi h imp o ed Fig. 2. The significan ly al e ed me aboli es a e he in e en ion and Venn figu e o hese me aboli es in he h ee g oups. A, significan ly al e ed me aboli es in exe cise g oup; B, significan ly al e ed me aboli es in die g oup; C, significan ly al e ed me aboli es in con ol g oup; D, common and di e en me aboli es in he h ee g oups. Y-axis, s anda d da a a e log 2 ans o ma ion. Fig. 3. The in ensi y o significan ly changed me aboli es in indi iduals be ween baseline and he 6-mon h ollow-up in he exe cise and die g oups. Each line ep esen s one indi idual. X. Zhang e al. Spo s Medicine and Heal h Science 2 (2020) 95–101 99 glucose con ol, weigh loss, and an ele a ed high-densi y lipop o ein choles e ol le el. 50 The obse ed esul s o he die a y in e en ion induced al e a ion o pu ine, glucose, and galac ose me abolisms, and esea ch is needed o whe he hese al e a ions a e a ibu ed o imp o ed sleep quali y. The die in e en ion g oup ailed o educe body weigh and FM significan ly du ing he s udy, and his was p obably because he die p og am ocused on he ela i e mac onu ien composi ion and some subjec s we e unable o educe hei o al ene gy in ake. These esul s sugges wo poin s. Fi s , in addi ion o sho sleep du a ion, 51 ch onic insomnia symp oms also con ibu e o weigh gain. 52 Second, he al e ed ene gy-balance is likely o play a ole in die induced sleep al e a ion. 51 Ne e heless, u he s udies a e needed o elucida e he unde lying mechanisms. Ou s udy has s eng hs and limi a ions. Wi h ega d o he s eng hs, fi s , he s udy pa icipan s we e andomly assigned o he in e en ion and con ol g oups, and sleep was measu ed using a non-con ac sleep moni o ing sys em. Second, he exe cise aining p og am was p esc ibed acco ding o he ecommenda ions o he Ame ican College o Spo s Medicine and was supe ised by expe ienced aine s. Las ly, he die in e en ion was indi idualized acco ding he pa icipan s’body weigh and calo ic in ake, and du a ions o he exe cise and die in e en ions we e ela i ely long. The limi a ions include he small numbe o pa - icipan s and a ela i ely high d op-ou a e in he con ol g oup, which may ha e a ec ed he s udy esul s. In addi ion, we should ha e added ano he g oup which is exe cise plus die ing g oup, his would be be e o dis inguish he e ec s o exe cise, die , and exe cise plus die . Conclusion In conclusion, he exe cise and die in e en ions we e associa ed wi h imp o ed sleep quali y and ci cula ing me aboli es ac oss di e en biochemical pa hways. The e ec o exe cise on sleep is mainly ela ed o amino acids, ca bohyd a es and lipids me abolisms, whe eas he e ec o die on sleep is ela ed o ca bohyd a es, lipids and o ganic acids me- abolisms. Al hough ou s udy esul s could no show whe he he al- e a ions o me abolic we e causally linked o li es yle in e en ions o imp o ed sleep quali y, ou findings may p o ide new insigh in o he biological mechanisms unde lying he me abolic e ec s o physical ac- i i y, die , and sleep quali y. Fu he s udies a e needed o be e un- de s and he ole o li es yle and sleep quali y in me abolic egula ion. Au ho s’con ibu ions The au ho s’con ibu ions a e as ollows: SC is he p incipal in es- iga o ; she designed he s udy and o e saw he implemen a ion o he p ojec , ained he esea che s, supe ised he doc o al s uden s, and pa icipa ed in he da a collec ion, analyses and in e p e a ion, and edi ing o he manusc ip . XZ, XW, and PW d a ed he manusc ip . SL, XO, and XT pa icipa ed in he da a analyses, and in e p e a ion and discussion o he manusc ip . Submission s a emen The manusc ip has no been published and is no unde conside - a ion o publica ion elsewhe e. E hical app o al The s udy was app o ed by he E hics Commi ee o he Cen al Finland Heal h Ca e Dis ic (7/2011 OTE) and egis e ed a www.con o lled- ials.com (ISRCTN77172005). W i en in o med consen was ob- ained om all pa icipan s be o e he baseline measu emen s we e pe o med, and copies o he signed consen o ms we e ob ained. Conflic o in e es The au ho s ha e no conflic s o in e es o epo . Acknowledgmen s The au ho s a e g a e ul o Shumei Cheng, Tuija M. Mikkola, Samu Ma inm€ aki, Ja kko Tenhunen, Sa u Pekkala, E eliinaMunukka, and Aki Rahikainen, o hei con ibu ions o he da a collec ion o exe cise in e en ion and die in e en ion. This s udy was suppo ed financially by he Finnish Funding Agency o Technology and Inno a ion (TEKES2206/31/2010), 111 P ojec (B17029) o Shanghai Jiao Tong Uni e si y, Shanghai Jiao Tong Uni- e si y Zhiyuan Founda ion (g an CP2014013), and China Pos doc Schola ship Council (201806230001). Appendix A. Supplemen a y da a Supplemen a y da a o his a icle can be ound online a h ps://do i.o g/10.1016/j.smhs.2020.05.002. Re e ences 1. Yazdi Z, Sadeghniia -Haghighi K, Loukzadeh Z, Elmizadeh K, Abbasi M. P e alence o sleep diso de s and hei impac s on occupa ional pe o mance: a compa ison be ween shi wo ke s and nonshi wo ke s. Sleep Diso d. 2014;2014:870320. h ps://doi.o g/10.1155/2014/870320. 2. Budhi aja R, Ro h T, Hudgel DW, Budhi aja P, D ake CL. P e alence and polysomnog aphic co ela es o insomnia como bid wi h medical diso de s. Sleep. 2011;34(7):859–867. h ps://doi.o g/10.5665/SLEEP.1114. 3. Cai XH, Xie YP, Li XC, e al. The p e alence and associa ed isk ac o s o sleep diso de - ela ed symp oms in p egnan women in China. Sleep B ea h. 2013;17(3): 951–956. h ps://doi.o g/10.1007/s11325-012-0783-2. 4. Shapi o CM, Wa en PM, T inde J, e al. Fi ness acili a es sleep. Eu J Appl Physiol Occup Physiol. 1984;53(1):1–4. h ps://doi.o g/10.1007/BF00964680. 5. Mein jes AF, D i e HS, Shapi o CM. Imp o ed physical fi ness ailed o al e he EEG pa e ns o sleep in young women. Eu J Appl Physiol Occup Physiol. 1989;59(1-2): 123–127. h ps://doi.o g/10.1007/BF02396589. 6. Mon gome y I, T inde J, Pax on SJ. Ene gy expendi u e and o al sleep ime: e ec o physical exe cise. Sleep. 1982;5(2):159–168. h ps://doi.o g/10.1093/sleep/ 5.2.159. 7. Edinge JD, Mo ey MC, Sulli an RJ, e al. Ae obic fi ness, acu e exe cise and sleep in olde men. Sleep. 1993;16(4):351–359. h ps://doi.o g/10.1093/sleep/16.4.351. Table 2 Co ela ions be ween significan ly changed me aboli es and sleep pa ame e s. G oup Me aboli es SE SOL TST p p p Exe cise alpha-ke oisocap oic acid 0.52 0.026 Die 3-hyd oxybu ic acid 0.47 0.025 D-glucopy anose 0.54 0.006 0.43 0.048 Taga ose 0.51 0.016 Oxalic acid 0.49 0.021 4-deoxye y h onic acid 0.60 0.004 2-ke o-iso ale ic acid 0.45 0.029 : co ela ion coe ficien ; p: p alue; SE: sleep e ficiency; SOL: sleep onse la ency; TST: o al sleep ime. X. Zhang e al. Spo s Medicine and Heal h Science 2 (2020) 95–101 100 8. Aiello KD, Caughey WG, Nellu i B, Sha ma A, Mookadam F, Mookadam M. E ec o exe cise aining on sleep apnea: a sys ema ic e iew and me a-analysis. Respi Med. 2016;116:85–92. h ps://doi.o g/10.1016/j. med.2016.05.015. 9. Kline CE, C owley EP, Ewing GB, e al. The e ec o exe cise aining on obs uc i e sleep apnea and sleep quali y: a andomized con olled ial. Sleep. 2011;34(12): 1631–1640. h ps://doi.o g/10.5665/sleep.1422. 10. D i e HS, Taylo SR. Exe cise and sleep. Sleep Med Re . 2000;4(4):387–402. 11. Tan X, Alen M, Wang K, e al. E ec o six-mon h die in e en ion on sleep among o e weigh and obese men wi h ch onic insomnia symp oms: a andomized con olled ial. Nu ien s. 2016;8(11). h ps://doi.o g/10.3390/nu8110751. 12. Tan X, Alen M, Wiklund P, Pa inen M, Cheng S. E ec s o ae obic exe cise on home- based sleep among o e weigh and obese men wi h ch onic insomnia symp oms: a andomized con olled ial. Sleep Med. 2016;25:113–121. h ps://doi.o g/10.1016/ j.sleep.2016.02.010. 13. Knowlden AP, Hackman CL, Sha ma M. Sys ema ic e iew o die a y in e en ions a ge ing sleep beha io . J Al e na i e Compl Med. 2016;22(5):349–362. h ps:// doi.o g/10.1089/acm.2015.0238. 14. Lindse h G, Mu ay A. Die a y mac onu ien s and sleep. Wes J Nu s Res. 2016;38(8): 938–958. h ps://doi.o g/10.1177/0193945916643712. 15. Shi Z, McE oy M, Luu J, A ia J. Die a y a and sleep du a ion in Chinese men and women. In J Obes. 2008;32(12):1835–1840. h ps://doi.o g/10.1038/ijo.2008.191. 16. Giskeodega d GF, Da ies SK, Re ell VL, Keun H, Skene DJ. Diu nal hy hms in he human u ine me abolome du ing sleep and o al sleep dep i a ion. Sci Rep. 2015;5: 14843. h ps://doi.o g/10.1038/s ep14843. 17. Bell LN, Kilkus JM, Boo h 3 d JN, B omley LE, Impe ial JG, Pene PD. E ec s o sleep es ic ion on he human plasma me abolome. Physiol Beha . 2013;122:25–31. h ps://doi.o g/10.1016/j.physbeh.2013.08.007. 18. Xu H, Zheng X, Jia W, Yin S. Ch oma og aphy/mass spec ome y-based bioma ke s in he field o obs uc i e sleep apnea. Medicine (Bal im). 2015;94(40), e1541. h ps://doi.o g/10.1097/MD.0000000000001541. 19. Zhan S, Wu Y, Sun P, Lin H, Zhu Y, Han X. Dec ease in ci cula ing a y acids is associa ed wi h isle dys unc ion in ch onically sleep- es ic ed a s. In J Mol Sci. 2016;17(12). h ps://doi.o g/10.3390/ijms17122102. 20. Sengup a A, Rhoades SD, Kim EJ, e al. Sleep es ic ion induced ene gy, me hyla ion and lipogenesis me abolic swi ches in a li e . In J Biochem Cell Biol. 2017;93: 129–135. h ps://doi.o g/10.1016/j.biocel.2017.08.014. 21. Tan X, Saa inen A, Mikkola TM, e al. E ec s o exe cise and die in e en ions on obesi y- ela ed sleep diso de s in men: s udy p o ocol o a andomized con olled ial. T ials. 2013;14:235. h ps://doi.o g/10.1186/1745-6215-14-235. 22. Ga be CE, Blissme B, Deschenes MR, e al. Ame ican College o Spo s Medicine posi ion s and. Quan i y and quali y o exe cise o de eloping and main aining ca dio espi a o y, musculoskele al, and neu omo o fi ness in appa en ly heal hy adul s: guidance o p esc ibing exe cise. Med Sci Spo s Exe c. 2011;43(7): 1334–1359. h ps://doi.o g/10.1249/MSS.0b013e318213 e b. 23. Paalasmaa J, Wa is M, Toi onen H, Leppako pi L, Pa inen M. Unob usi e online moni o ing o sleep a home. Con P oc IEEE Eng Med Biol Soc. 2012;2012: 3784–3788. h ps://doi.o g/10.1109/EMBC.2012.6346791. 24. Oja P, Laukkanen R, Pasanen M, Ty y T, Vuo i I. A 2-km walking es o assessing he ca dio espi a o y fi ness o heal hy adul s. In J Spo s Med. 1991 Aug;12(4): 356–362. h ps://doi.o g/10.1055/s-2007-1024694. 25. Kind T, Wohlgemu h G, Lee DY, e al. FiehnLib: mass spec al and e en ion index lib a ies o me abolomics based on quad upole and ime-o -fligh gas ch oma og aphy/mass spec ome y. Anal Chem. 2009;81(24):10038–10048. h ps://doi.o g/10.1021/ac9019522. 26. Reed DL, Sacco WP. Measu ing sleep e ficiency: wha should he denomina o be? J Clin Sleep Med. 2016;12(2):263–266. h ps://doi.o g/10.5664/jcsm.5498. 27. Xiao Q, De kach A, Moo e SC, e al. Habi ual Sleep and human plasma me abolomics. Me abolomics. 2017;13(5). h ps://doi.o g/10.1007/s11306-017-1205-z. 28. Ma hews DE, Bie DM, Rennie MJ, e al. Regula ion o leucine me abolism in man: a s able iso ope s udy. Science. 1981;214(4525):1129–1131. h ps://doi.o g/10.1126/ science.7302583. 29. Banis e EW, Came on BJ. Exe cise-induced hype ammonemia: pe iphe al and cen al e ec s. In J Spo s Med. 1990;11(Suppl 2):S129–S142. h ps://doi.o g/ 10.1055/s-2007-1024864. 30. Kainulainen H, Hulmi JJ, Kujala UM. Po en ial ole o b anched-chain amino acid ca abolism in egula ing a oxida ion. Exe c Spo Sci Re . 2013;41(4):194–200. h ps://doi.o g/10.1097/JES.0b013e3182a4e6b6. 31. Lee S, Olsen T, Vinknes KJ, e al. Plasma sulphu -con aining amino acids, physical exe cise and insulin sensi i i y in o e weigh dysglycemic and no mal weigh no moglycemic men. Nu ien s. 2018;11(1). h ps://doi.o g/10.3390/nu11010010. 32. Glynn EL, Pine LW, Hu man KM, e al. Impac o combined esis ance and ae obic exe cise aining on b anched-chain amino acid u no e , glycine me abolism and insulin sensi i i y in o e weigh humans. Diabe ologia. 2015;58(10):2324–2335. h ps://doi.o g/10.1007/s00125-015-3705-6. 33. Kujala UM, Makinen VP, Heinonen I, e al. Long- e m leisu e- ime physical ac i i y and se um me abolome. Ci cula ion. 2013;127(3):340–348. h ps://doi.o g/ 10.1161/CIRCULATIONAHA.112.105551. 34. Mul JD, S an o d KI, Hi shman MF, Goodyea LJ. Exe cise and egula ion o ca bohyd a e me abolism. P og Mol Biol T ansl Sci. 2015;135:17–37. h ps://doi.o g/ 10.1016/bs.pmb s.2015.07.020. 35. Van Cau e E, Spiegel K, Tasali E, Lep oul R. Me abolic consequences o sleep and sleep loss. Sleep Med. 2008;9(Suppl 1):S23–S28. h ps://doi.o g/10.1016/S1389- 9457(08)70013-3. 36. Papand eou C, Camacho-Ba cia L, Ga cia-Ga ilan J, e al. Ci cula ing me aboli es associa ed wi h objec i ely measu ed sleep du a ion and sleep a iabili y in o e weigh /obese pa icipan s: a me abolomics app oach wi hin he SATIN s udy. Sleep. 2019;42(5). h ps://doi.o g/10.1093/sleep/zsz030. 37. Sp ie LL. New insigh s in o he in e ac ion o ca bohyd a e and a me abolism du ing exe cise. Spo s Med. 2014;44(Suppl 1):S87–S96. h ps://doi.o g/10.1007/ s40279-014-0154-1. 38. Seelig E, Kelle U, Kla ho e M, e al. Neu oendoc ine egula ion and me abolism o glucose and lipids in p ima y ch onic insomnia: a p ospec i e case-con ol s udy. PLoS One. 2013;8(4), e61780. h ps://doi.o g/10.1371/jou nal.pone.0061780. 39. Ea nes CP, A e o EG, Sui X, Lee DC, Chu ch TS, Blai SN. Maximal es ima ed ca dio espi a o y fi ness, ca diome abolic isk ac o s, and me abolic synd ome in he ae obics cen e longi udinal s udy. Mayo Clin P oc. 2013;88(3):259–270. h ps:// doi.o g/10.1016/j.mayocp.2012.11.006. 40. Kuula L, Pesonen AK, Kajan ie E, e al. Sleep and lipid p ofile du ing ansi ion om childhood o adolescence. J Pedia . 2016;177:173–178. h ps://doi.o g/10.1016/ j.jpeds.2016.06.026. e1. 41. Zhan Y, Chen R, Yu J. Sleep du a ion and abno mal se um lipids: he China Heal h and Nu i ion Su ey. Sleep Med. 2014;15(7):833–839. h ps://doi.o g/10.1016/ j.sleep.2014.02.006. 42. Ke kho s M, Boudjel ia KZ, S enui P, B ohee D, Cauchie P, Vanhae e beek M. Sleep es ic ion inc eases blood neu ophils, o al choles e ol and low densi y lipop o ein choles e ol in pos menopausal women: a p elimina y s udy. Ma u i as. 2007;56(2): 212–215. h ps://doi.o g/10.1016/j.ma u i as.2006.07.007. 43. Weljie AM, Mee lo P, Goel N, e al. Oxalic acid and diacylglyce ol 36:3 a e c oss- species ma ke s o sleep deb . P oc Na l Acad Sci U S A. 2015;112(8):2569–2574. h ps://doi.o g/10.1073/pnas.1417432112. 44. Zi in JA, Sna JF. An au oma ed colo ime ic me hod o he measu emen o 3- hyd oxybu y a e concen a ion. Anal Biochem. 1973;52(2):456–461. h ps://doi.o g/ 10.1016/0003-2697(73)90048-1. 45. B ashea A, Cook GA. A spec opho ome ic, enzyma ic assay o D-3- hyd oxybu y a e ha is no dependen on hyd azine. Anal Biochem. 1983;131(2): 478–482. h ps://doi.o g/10.1016/0003-2697(83)90201-4. 46. Cede naes J, Schonke M, Wes holm JO, e al. Acu e sleep loss esul s in issue-specific al e a ions in genome-wide DNA me hyla ion s a e and me abolic uel u iliza ion in humans. Sci Ad . 2018;4(8):eaa 8590. h ps://doi.o g/10.1126/sciad .aa 8590. 47. Reu akul S, Van Cau e E. In e ac ions be ween sleep, ci cadian unc ion, and glucose me abolism: implica ions o isk and se e i y o diabe es. Ann N Y Acad Sci. 2014;1311:151–173. h ps://doi.o g/10.1111/nyas.12355. 48. Alnaji A, Law GR, Sco EM. The ole o sleep du a ion in diabe es and glucose con ol. P oc Nu Soc. 2016;75(4):512–520. h ps://doi.o g/10.1017/ S002966511600063X. 49. Kay DB, Ka im HT, Hasle BP, e al. Impac o acu e sleep es ic ion on ce eb al glucose me abolism du ing eco e y non- apid eye mo emen sleep among indi iduals wi h p ima y insomnia and good sleepe con ols. Sleep Med. 2019;55: 81–91. h ps://doi.o g/10.1016/j.sleep.2018.12.007. 50. Lu Y, Le in GV, Donne TW. Taga ose, a new an idiabe ic and obesi y con ol d ug. Diabe es Obes Me abol. 2008;10(2):109–134. h ps://doi.o g/10.1111/j.1463- 1326.2007.00799.x. 51. Shech e A, O'Kee e M, Robe s AL, Zammi GK, RoyChoudhu y A, S -Onge MP. Al e a ions in sleep a chi ec u e in esponse o expe imen al sleep cu ailmen a e associa ed wi h signs o posi i e ene gy balance. Am J Physiol Regul In eg Comp Physiol. 2012;303(9):R883–R889. h ps://doi.o g/10.1152/ajp egu.00222.2012. 52. Swi DL, Johannsen NM, La ie CJ, Ea nes CP, Chu ch TS. The ole o exe cise and physical ac i i y in weigh loss and main enance. P og Ca dio asc Dis. 2014;56(4): 441–447. h ps://doi.o g/10.1016/j.pcad.2013.09.012. X. Zhang e al. Spo s Medicine and Heal h Science 2 (2020) 95–101 101