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/
In e ac i e e ec s o aging and ae obic capaci y on ene gy me abolism– ela ed
me aboli es o se um, skele al muscle, and whi e adipose issue
© 2021 he Au ho s
Published e sion
Zhuang, Haihui; Ka inen, Si a; Tö mäkangas, Timo; Zhang, Xiaobo; Ojanen,
Xiaowei; Velagapudi, Vidya; Alen, Ma kku; B i on, S e en L.; Koch, Lau en G.;
Kainulainen, Heikki; Cheng, Sulin; Wiklund, Pe i
Zhuang, H., Ka inen, S., Tö mäkangas, T., Zhang, X., Ojanen, X., Velagapudi, V., Alen, M.,
B i on, S. L., Koch, L. G., Kainulainen, H., Cheng, S., & Wiklund, P. (2021). In e ac i e e ec s o
aging and ae obic capaci y on ene gy me abolism– ela ed me aboli es o se um, skele al
muscle, and whi e adipose issue. Ge oScience, 43(6), 2679-2691.
h ps://doi.o g/10.1007/s11357-021-00387-1
2021
ORIGINAL ARTICLE
In e ac i e e ec s o aging and ae obic capaci y on ene gy
me abolism– ela ed me aboli es o se um, skele al muscle,
and whi e adipose issue
Haihui Zhuang &Si a Ka inen &Timo Tö mäkangas &Xiaobo Zhang &
Xiaowei Ojanen &Vidya Velagapudi &Ma kku Alen &S e en L. B i on &
Lau en G. Koch &Heikki Kainulainen &Sulin Cheng &Pe i Wiklund
Recei ed: 21 July 2020 /Accep ed: 17 May 2021
#The Au ho (s) 2021
Abs ac Ae obic capaci y is a s ong p edic o o lon-
ge i y. Wi h aging, ae obic capaci y dec eases concom-
i an ly wi h changes in whole body me abolism leading
o inc eased disease isk. To add ess he ole o ae obic
capaci y, aging, and hei in e ac ion on me abolism, we
u ilized a models selec i ely b ed o low and high
in insic ae obic capaci y (LCRs/HCRs) and compa ed
he me abolomics o se um, muscle, and whi e adipose
issue (WAT) a wo ime poin s: Young a s we e
sac i iced a 9 mon hs o age, and old a s we e
sac i iced a 21 mon hs o age. Ta ge ed and semi-
quan i a i e me abolomics analysis was pe o med on
he ul a-p essu e liquid ch oma og aphy andem mass
spec ome y (UPLC-MS) pla o m. The e ec s o ae -
obic capaci y, aging, and hei in e ac ion we e s udied
ia eg ession analysis. Ou esul s showed ha high
ae obic capaci y is associa ed wi h an accumula ion o
iso ale ylca ni ine in muscle and se um a es , which is
Ge oScience
h ps://doi.o g/10.1007/s11357-021-00387-1
Haihui Zhuang, Si a Ka inen con ibu ed equally (as i s
au ho s) and Heikki Kainulainen, Sulin Cheng and Pe i Wiklund
con ibu ed equally (as senio au ho s).
H. Zhuang :X. Zhang :X. Ojanen :S. Cheng
School o Li e Sciences and Bio echnology, Shanghai Jiao Tong
Uni e si y, Shanghai, China
H. Zhuang :X. Zhang :S. Cheng :P. Wiklund
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
H. Zhuang :S. Ka inen (*):T. Tö mäkangas :
H. Kainulainen :S. Cheng
Facul y o Spo and Heal h Sciences, Uni e si y o Jy äskylä,
Jy äskylä, Finland
e-mail: si a.m.ka inen@jyu. i
V. Velagapudi
Me abolomics Uni , Ins i u e o Molecula Medicine Finland
(FIMM), HiLIFE, Uni e si y o Helsinki, Helsinki, Finland
M. Alen
Depa men o Medical Rehabili a ion, Oulu Uni e si y Hospi al,
Oulu, Finland
S. L. B i on
Depa men o Anes hesiology, Uni e si y o Michigan, Ann
A bo , MI, USA
S. L. B i on
Molecula and In eg a i e Physiology, Uni e si y o Michigan,
Ann A bo , MI, USA
L. G. Koch
Depa men o Physiology and Pha macology, The Uni e si y o
Toledo College o Medicine and Li e Sciences, Toledo, OH, USA
P. Wiklund
Huawei Helsinki Resea ch Cen e , Huawei Technologies Oy
(Finland) Co. L d, Helsinki, Finland
likely due o mo e e icien leucine ca abolism in mus-
cle. Wi h aging, se e al amino acids we e down egula -
ed in muscle, indica ing mo e e icien amino acid me-
abolism, whe eas in WAT less e icien amino acid
me abolism and dec eased mi ochond ial β-oxida ion
we e obse ed. Ou esul s u he e ealed ha high
ae obic capaci y and aging in e ac i ely a ec lipid me-
abolism in muscle and WAT, possibly comba ing un-
a o able aging- ela ed changes in whole body me abo-
lism. Ou esul s highligh he signi ican ole o WAT
me abolism o heal hy aging.
Keywo ds Ae obic capaci y .Aging .Me abolomics .
Me aboli es
In oduc ion
Epidemiological s udies ha e indica ed ha low ae obic
capaci y, exp essed as maximal oxygen up ake
(VO
2max
), is a majo isk ac o o ca dio-me abolic
diseases [1], whe eas high ae obic capaci y is associa ed
wi h a mo e a o able ca dio-me abolic heal h p o ile
[2,3], e en in he p esence o excess adiposi y [4].
These obse a ions a e complemen ed by mul iple an-
domized con olled ials, which ha e shown ha exe -
cise aining signi ican ly imp o es ca dio espi a o y
i ness and adiposi y- ela ed ca dio-me abolic bio-
ma ke s [5]. Fu he mo e, ae obic capaci y has been
shown o be a s ong independen p edic o o ca dio-
ascula and all-cause mo ali y in many popula ion
coho s [6]. These s udies sugges he e is a undamen al
connec ion be ween impai ed ae obic me abolism, com-
plex disease isk, and educed longe i y.
Epidemiological s udies ha e also shown ha ae obic
capaci y and physical ac i i y le el dec ease wi h age [7,
8] concomi an ly wi h quan i a i e and s uc u al chang-
es in body composi ion [9]. These changes a e associa -
ed wi h al e a ions in whole body me abolism, insulin
esis ance, low-g ade in lamma ion, and dyslipidemia,
which con ibu e o he de elopmen o ca dio-
me abolic disease wi h age [10,11]. Howe e , i is no
well unde s ood o wha ex en he aging- ela ed chang-
es in me abolic p o iles a e a ibu able o ei he in insic
ae obic capaci y o dec eased physical ac i i y.
To enable he in es iga ion o in insic ae obic en-
du ance capaci y on me abolic p o ile, Koch and B i on
[12] de eloped he e ogeneous a lines by a i icial se-
lec ion o low and high inbo n exe cise capaci y. This
a model o low capaci y unne s (LCRs) and high
capaci y unne s (HCRs) p ospec i ely es s he associ-
a ion be ween ae obic exe cise capaci y and su i abil-
i y [13]. The pheno ype o HCRs is coinciden wi h a
hos o heal h bene i s [14], including a 28–40% in-
c eased li espan [15]. A ecen s udy in HCRs and LCRs
showed ha an enhancemen o ae obic capaci y could
mi iga e some o he changes in he plasma me abolic
p o ile, which we e associa ed wi h aging [16]. Con-
comi an ly wi h highe ae obic capaci y, muscle mi o-
chond ial unc ion and oxida i e ene gy me abolism a e
enhanced in HCRs compa ed wi h hose in LCRs [17,
18]. In addi ion o skele al muscle, whi e adipose issue
(WAT) sha es an impo an ole in he egula ion o
whole body me abolism [19] and WAT- o-muscle com-
munica ion is c i ical in o e all me abolic heal h [20].
Mos p e ious s udies ha e examined he e ec s o
ae obic capaci y on me abolism only in one issue [21,
22] and using only young animals (3-4 mon hs o age)
[17]. Thus, possible in e ac i e e ec s o aging and
ae obic capaci y on issue me abolism a e ye o be
esol ed. He e, we assessed me abolic p o iles in se um,
muscle, and WAT in bo h young (9 mon hs o age) and
old (21 mon hs o age) HCR and LCR g oups by using
a ge ed and semi-quan i a i e me abolomics analysis
which was pe o med on he iple quad upole andem
mass spec ome y coupled o he ul a-p essu e liquid
ch oma og aphy (UPLC-MS) pla o m. We p opose o
de ine he mos p ominen di e ences o ae obic ca-
paci y in he me abolic p o ile o skele al muscle [17,
18,23,24]. Also, we expec ha ae obic capaci y and
aging in e ac i ely a ec se e al me aboli es in muscle
and WAT leading o changes in se um me aboli e le els,
as LCRs a e known o ha e highe me abolic disease
isk and sho e li espan [13,14].
Ma e ials and me hods
Animal model
The HCR/LCR a model was de i ed om a gene ical-
ly he e ogeneous ounde popula ion (N:NIH s ock) and
b ed wi h wo-way a i icial selec ion [12]. Animals
we e e alua ed a 11 weeks o age o maximal unning
capaci y a he Uni e si y o Michigan (Ann A bo , MI,
USA) wi h a speed- amped eadmill unning es (15°
slope, ini ial eloci y o 10 m/min, inc eased 1 m/min
e e y 2 min). In his s udy, 34 emale a s (16 HCRs and
Ge oScience
18 LCRs) om gene a ions 23–27 o selec ion we e
used. All a s we e kep in an en i onmen ally con-
olled acili y wi h ligh /da k cycle o 12/12 h and had
ee access o ood and ap wa e (R36, Lab o , S ock-
holm, Sweden).
Tes ing p ocedu e
A e a i ing o Finland, a s we e es ed o maximal
unning capaci y a he age o 9 mon hs wi h he same
speed- amped unning es as desc ibed p e iously [18].
Subsequen ly, bo h HCRs and LCRs we e di ided in o
weigh and maximal unning capaci y–ma ched sub-
g oups: HCR-Y o LCR-Y (young, n= 10 in each g oup),
and HCR-O (old, n=6)o LCR-O(old,n=8).A e
maximal ae obic capaci y es ing, HCR-Y and LCR-Y
we e weighed and sac i iced and hindlimb skele al mus-
cles, WAT, and blood samples we e collec ed. HCR-O
and LCR-O we e con inued o be housed indi idually in a
s anda d cage un il he age o 21 mon hs. Maximal ae obic
capaci y was assessed again a he age o 21 mon hs. A e
es ing, animals we e weighed and sac i iced and skele al
muscle, WAT, and se um samples we e ha es ed.
Tissue collec ion
Soleus, ex enso digi o um longus (EDL), plan a is,
gas ocnemius, quad iceps emo is muscle, adipose is-
sue a ound he o a ies, isce al adipose issue, and
e ope i oneal adipose issue we e excised and
weighed. Tissue samples we e snap ozen in liquid
ni ogen and hen s o ed in −80°C un il analyses. Gas-
ocnemius, e ope i oneal adipose issue, and se um
we e used o me abolomics analyses. Skele al muscle
mass was calcula ed as he sum weigh o soleus, EDL,
plan a is, gas ocnemius, and quad iceps emo is.
Me abolomics analyses
Ta ge ed and semi-quan i a i e me abolomics analyses
we e pe o med on a Wa e s Xe o TQ-S iple quad u-
pole andem mass spec ome e coupled o he ul a-
p essu e liquid ch oma og aphy (UPLC-MS) pla o m
using he p e iously published p o ocol in FIMM (In-
s i u e o Molecula Medicine Finland) [25]. B ie ly,
me aboli es we e ex ac ed om 100 μLse um,20mg
muscle, and 20 mg WAT samples espec i ely using
p o ein p ecipi a ion by adding ace oni ile +1% o mic
acid. The collec ed ex ac s we e dispensed in Os o 96-
well pla es (Wa e s Co po a ion, Mil o d, USA) and
il e ed by applying a acuum a a del a p essu e o
300–400 mba o 2.5 min on obo ’s acuums a ion.
Fil e ed sample ex ac (5 μL) was injec ed in an
Acqui y UPLC sys em coupled o a Xe o TQ-S iple
quad upole mass spec ome e (Wa e s Co po a ion,
Mil o d, MA, USA) which was ope a ed in bo h posi-
i e and nega i e pola i ies wi h swi ching ime o 20
ms. Mul iple Reac ion Moni o ing (MRM) acquisi ion
mode was selec ed o he quan i ica ion o me aboli es.
MassLynx 4.1 so wa e was used o da a acquisi ion,
da a handling, and ins umen con ol. Da a p ocessing
was done using Ta ge Lynx 4.1 so wa e. Eigh y-nine
me aboli es in se um, 71 in muscle, and 71 in WAT
we e iden i ied in whole samples.
S a is ical analysis
The desc ip i e cha ac e is ics o a s we e analyzed by
using IBM SPSS S a is ics 24.0 (SPSS, Chicago, USA).
Da a we e checked o no mali y wi h he Shapi o-Wilk
es . As mos o he a iables we e no no mally dis ib-
u ed, nonpa ame ic es s we e chosen o g oup com-
pa isons, and s a is ical signi icance was se a p<0.05.
The me abolomics analysis was same as in ou p e ious
s udy wi h se um and muscle samples [26].
All me abolomics da a we e ans o med wi h a sui able
ans o ma ion om he Box-Cox amily and hen scaled
by di iding each a iable wi h i s s anda d de ia ion (uni -
a iance scaling, c i e ia o absolu e alue o he esiduals
o skew and ku osis < 2). We used pa ial-leas -squa es
disc iminan analysis (PLS-DA) o age and ae obic ca-
paci y g oup compa isons o assessing mul i a ia e me-
aboli e p o iles. Pe mu a ion es s (G = 200 pe model)
we e used o alida e he model and o a oid o e - i ing.
We also pe o med compa a i e uni a ia e analyses
using linea eg ession models o assess (1) age (old s.
young) and ae obic capaci y (high s. low) g oup di -
e ences on me aboli es, and (2) associa ions be ween
me aboli es and unning speed while accoun ing also o
he g ouping s uc u e (age and ae obic capaci y). In he
o me case, we compu ed he p alue o he join impac
o bo h g ouping ac o s (age and ae obic capaci y) and
hei in e ac ion. In he la e case, we used he p alue
o he join con ibu ion o he me aboli e- ela ed p e-
dic o s in he model (me aboli e, me aboli e × age, me-
aboli e × ae obic capaci y, me aboli e × age × ae obic
capaci y). Model equa ions a e shown in he Technical
Supplemen . In a PLS model, a iable impo ance in
Ge oScience
p ojec ion (VIP) is used as an index o he es ima ed
impo ance o a a iable as a con ibu o o g oup sep-
a a ion in he model. Among me aboli es, we selec ed
me aboli es based on VIP alues ≥1andp< 0.05 o he
model combined impac o he ac o s in he uni a ia e
linea model.
Fo unning speed-me aboli e associa ion, we consid-
e ed as in luen ial me aboli es hose wi h FDR-co ec ed p
alue o he associa ion wi h me aboli e < 0.05. Mul i a -
ia e and uni a ia e modeling we e pe o med using a
cus om sc ip in he R p og amming en i onmen , e sion
4.0.2, u ilizing packages psych ( e sion 2.0.8) o assess-
men o no mali y o esiduals and emmeans ( e sion
1.5.1, h ps://CRAN.R-p ojec .o g/package=emmeans.)
o compu ing es ima ed ma ginal coe icien s and olls
( e sion 1.20.0) o PLS models [27].
Pa hway analysis
Me aboAnalys ( 3.5) was used o pa hway en ichmen
analysis o map signi ican ly di e en ial me aboli es
ound in PLS-DA and eg ession models o hei co e-
sponding pa hways [28,29]. This web-based ool elies
on he knowledgebase o Kyo o Encyclopedia o Genes
and Genomes (KEGG) me abolic pa hway.
Resul s
Body mass, unning capaci y, and skele al muscle
mass/body mass a io
HCRs had lowe body mass and highe maximal unning
speed han LCRs bo h a 9 mon hs (p< 0.001) and 21
mon hs o age (p= 0.002, p= 0.006, espec i ely) (Fig.
1a, c). Hindlimb skele al muscle mass o body mass a io
was lowe in bo h HCRs and LCRs when compa ing 21-
mon h-old a s o 9-mon h-old (Fig. 1b). The maximal
unning speed was 25.5% lowe in 21-mon h-old HCRs
compa ed wi h 9-mon h-old HCRs (p< 0.001) whe eas in
LCRs, no signi ican di e ence in unning speed wi h age
was obse ed (p=0.360;Fig.1c).
E ec s o in insic ae obic capaci y and aging
on me abolic p o iles o se um, muscle, and whi e
adipose issue
We used PLS-DA o iden i y di e ences in he me a-
bolic p o iles be ween he a lines (LCR s. HCR) and
age g oups (young, Y s. old, O). The e we e clea
sepa a ions in se um (Q
2
=0.443,Q
2
in e cep =
−0.273), muscle (Q
2
=0.350,Q
2
in e cep = −0.276),
and WAT (Q
2
=0.325,Q
2
in e cep = −0.405) me abolic
p o iles (Fig. 2; Table S1). Howe e , a la ge di e ence
be ween R
2
and Q
2
was obse ed in WAT issue, sug-
ges ing a poo p edic i e pe o mance o he model in
WAT (Table S1). The co esponding associa ions be-
ween unning speed and me aboli es in g oups o ae -
obic capaci y and age a e shown in Supplemen a y
Table S4. When exp essing he da a ia unning speed,
he me abolic p o iles o se um and muscle a e simila as
wi h ae obic capaci y, bu he p o ile o WAT sepa a ed
also a young age (Fig. S1).
E ec s o in insic ae obic capaci y and aging on single
me aboli es o se um, muscle, and whi e adipose issue
Nex , we compa ed he di e ences a he le el o single
me aboli es be ween HCRs and LCRs wi h uni a ia e
analysis. The Venn diag ams o single me aboli es
changed by ae obic capaci y, aging, and hei in e ac ion
in se um, muscle, and WAT a e shown in Figs. 3,4,and
5. The co esponding uni a ia e analyses wi h eg es-
sion coe icien s a e shown in Supplemen a y Table S2
and design-con olled skew and ku osis es ima es o
me aboli es a e mos op imal ans o ma ion in Sup-
plemen a y Table S5. Supplemen a y Table S3 shows
he pa hway analysis o he signi ican ly changed
me aboli es.
E ec o in insic ae obic capaci y on se um, muscle,
and WAT me aboli es
In se um, iso ale ylca ni ine, inosine, and
hexanoylca ni ine we e up egula ed wi h high ae obic
capaci y, whe eas in muscle, only iso ale ylca ni ine
was up egula ed (Figs 3and 4; Table S2). In WAT, high
ae obic capaci y up egula ed he le el o
dime hylglycine and down egula ed he le els o be a-
ine and oc anoylca ni ine (Fig. 5;TableS2). The p e-
dominan pa hways ha we e a ec ed by ae obic capac-
i y we e nucleo ide me abolism in se um and amino acid
me abolism in WAT (p<0.050; Table S3).
E ec o aging on se um, muscle, and WAT me aboli es
In se um, aging was associa ed wi h highe le els o
se e al me aboli es, ha ing he mos p ominen e ec
Ge oScience

on amino acid me aboli es (Fig. 3; Table S2). The le els
o isoleucine, p oline, o ni hine, and glu amine we e
down egula ed, while alanine glycine ci ulline, hy-
d oxyp oline, and aspa a e we e up egula ed in se um
(Fig. 3). In muscle, se e al amino acids we e down eg-
ula ed including b anched-chain amino acids (BCAAs;
aline, leucine, and isoleucine) (Fig. 4;TableS2). In
WAT, alanine and leucine le els we e up egula ed and
ca ni ine and se e al acylca ni ines (isobu y ylca ni ine,
p opionylca ni ine, and iso ale ylca ni ine) we e down-
egula ed (Fig. 5; Table S2). The p edominan pa hways
ha we e a ec ed by aging we e linked o amino acid
me abolism in se um, muscle, and WAT (p≤0.050;
Table S3).
E ec o he in e ac ion o ae obic capaci y and aging
on se um, muscle, and WAT me aboli es
We ound ha high ae obic capaci y and aging in e ac-
i ely up egula ed he le els o leucine, aline, a ginine,
isobu y ylca ni ine, and phosphoe hanolamine, and
down egula ed succina e and u acil in se um (Fig. 3;
Table S2). In muscle, ace ylca ni ine and au ine we e
up egula ed, whe eas alanine, hyd oxyp oline,
ace oace ic acid, succina e, o ni hine, yp ophan,
pan o henic acid, hippu ic acid, olic acid, suc ose, and
spe midine we e down egula ed (Fig. 4;TableS2). In
WAT, ae obic capaci y and aging in e ac i ely down-
egula ed ace oace ic acid, ci ulline, GABA, and AMP
(Fig. 5;TableS2). The p edominan pa hways ha we e
a ec ed by he in e ac ion o ae obic capaci y and aging
we e linked o amino acid me abolism in se um and
muscle and lipid me abolism in WAT (p<0.050;
Table S3).
Discussion
The p esen s udy sugges s ha ae obic capaci y and
aging ha e di e en impac s on he me abolic p o iles
o se um, skele al muscle, and whi e adipose issue
(WAT). Con a y o ou hypo hesis, only
Fig. 1 The e ec s o ae obic capaci y and aging on muscle mass
o body mass a io and maximal unning speed in a s. Body mass
(a), muscle mass o body mass a io (b), and maximal unning
speed (c) in he s udied a g oups. HCR-Y, high capaci y unne ,
young; LCR-Y, low capaci y unne , young; HCR-O, high capac-
i y unne , old; LCR-O, low capaci y unne , old. Da a is p esen ed
as mean ± SD. *p<0.05, **p<0.01, and ***p<0.001
Ge oScience
iso ale ylca ni ine was up egula ed by high ae obic ca-
paci y in muscle, whe eas aging changed a la ge amoun
o me aboli es p edominan ly linked o amino acid me-
abolism in se um, muscle, and WAT. Ae obic capaci y
and aging in e ac i ely a ec ed se e al me aboli es, he
mos p edominan pa hways being amino acid me abo-
lism in se um and muscle and lipid me abolism in WAT.
Ae obic capaci y and aging ha e dis inc e ec s
on me abolic p o iles o muscle and WAT
We obse ed a clea sepa a ion o he me abolic p o iles
(PLS-DA) o se um and muscle in HCRs and LCRs a
young age, whe eas a old age, he me abolic p o iles
o e lapped (HCR-Y s. LCR-Y and HCR-O s. LCR-
O; Fig. 2). In e es ingly, he opposi e was obse ed o
WAT, whe e he me abolic p o iles o e lapped a young
age and showed a clea sepa a ion be ween HCRs and
LCRs a old age (Fig. 2e). A p e ious s udy by Falegan
e al. examined he me abolic p o ile o plasma o young
(13 mon hs) and old (26 mon hs) male HCR and LCR
a s and obse ed a s ong p o ile sepa a ion in old and
LCRs, whe eas young and HCRs we e ound less p e-
dic i e [16]. Hence, hey concluded ha in plasma,
me abolomics analysis be e p edic s age a he han
ae obic capaci y. Ye ou esul s om se um sugges
ha a old age, he me abolic p o iles o HCRs and
LCRs a e mo e simila . The eason o he di e gence
in obse a ions may be due o di e ences in sex, age,
and gene a ion (Falegan e al. used a s om gene a ions
17 and 19 o selec ion), as ing du a ion, and eu hanasia
me hod. Di e en me abolomics pla o ms may also
accoun o he disc epan esul s as Falegan e al. used
NMR p o on spec ome y, while in he p esen s udy
Fig. 2 Sco e plo o PLS-DA in se um (a), muscle (c), and WAT
(e) wi h ae obic capaci y and age and linea eg ession model wi h
200 imes pe mu a ion es o he models o se um (b), muscle (d),
and WAT ( ). LCR-O, low capaci y unne , old; HCR-O, high
capaci y unne , old; LCR-Y, low capaci y unne , young; HCR-
Y, high capaci y unne , young
Ge oScience
we used UPLC-MS in a ge ed me aboli e p o iling.
Ne e heless, ou esul s highligh he ole o WAT
me abolism in heal hy aging.
We u he obse ed ha he me abolic p o iles o
se um and muscle we e simila i espec i e o whe he
he da a was exp essed using he a lines and age g oups
(Fig. 2a, c) o ia unning speed (Fig. S1a, c). Ou esul
sugges s ha he sepa a ion o he a lines h ough
maximal unning capaci y has d i en he changes in
he me abolism o hese a s, as high unning speed is
he p ac ical mani es a ion o he gene ic b eeding o
HCR a s and LCR a s. In e es ingly, unlike in he line
s. age compa ison, he me abolic p o iles o young a s
appea ed di e en ially exp essed when shown ia un-
ning speed (Fig. S1e). This obse a ion may be due o a
la ge di e ence in he unning speed be ween he a
lines a young age (HCR-Y s. LCR-Y), bu ha is less
p ominen a old age (Fig. 1c). These obse a ions sug-
ges ha ae obic capaci y enhances he me abolism o
adipose issue and suppo s ea lie s udies which ha e
shown ha exe cise aining enhances whi e adipose
issue me abolism in HCR and LCR a s [30], and ha
adipose issue may media e some o he heal h bene i s
o ae obic exe cise aining [31,32].
High ae obic capaci y is associa ed wi h e icien
leucine ca abolism in muscle and al e ed amino acid
and a y acid me abolism in WAT
Skele al muscle is an impo an issue con ibu ing bo h
o ae obic capaci y and whole body me abolism [33].
Hence, i was unexpec ed ha high ae obic capaci y was
associa ed only wi h up egula ion o iso ale ylca ni ine
in muscle (Fig. 3). Iso ale ylca ni ine is p oduced du -
ing leucine ca abolism and accumula ion o
iso ale ylca ni ine in se um migh indica e a de ec in
leucine ca abolism [34,35]. In humans, inc eased con-
cen a ions o iso ale ylca ni ine and hexanoylca ni ine
ha e been p e iously epo ed in ype 2 diabe es
[36–38] as well as in a model o ype 2 diabe es-
Zucke diabe ic a y a s [39]. Howe e , e icien leu-
cine ca abolism may also lead o accumula ion o
Fig. 3 Venn diag am o di e en ially exp essed me aboli es wi h
ae obic capaci y, age, and hei in e ac ion in se um. A ows
ep esen he di ec ion o eg ession coe icien ; ha in he
o e lapping egion (ae obic capaci y * age) may o igina e om
ae obic capaci y (1s a ow) and age (2nd a ow) o hei in e ac-
ion (3 d a ow)
Ge oScience
iso ale ylca ni ine when ene gy demand dec eases.
HCRs may ca abolize leucine mo e ac i ely du ing
mo emen o supply ace yl-coenzyme A o ica boxylic
acid cycle, whe eas a es he p oduced iso ale yl-
coenzyme A is no needed, and is u he me abolized
o iso ale ylca ni ine. Suppo ing his hypo hesis, ou
p e ious s udies ha e shown inc eased exp ession o he
enzymes o BCAA ca abolism in skele al muscle o
HCRs [24] and physically ac i e humans [40]. Also, a
p e ious s udy by O e mye e al. showed ha young
(3–4.5 mon hs o age) HCRs oxidize BCAAs (leucine,
isoleucine, and aline) mo e e icien ly compa ed o
LCRs o he same age [17]. Fu he mo e, hey ound
ha muscle BCAAs we e lowe in HCRs han in LCRs
a e 10 min o speed- amped eadmill unning. These
changes we e pa alleled by a all in plasma BCAAs in
HCRs, indica ing inc eased u iliza ion o BCAAs du -
ing exe cise. Acco ding o ou esul s, we specula e ha
du ing es ing a as ed s a e, HCRs may accumula e
iso ale ylca ni ine in o muscle and se um due o mo e
e icien leucine ca abolism in muscle.
Compa edwi h o he me abolically ac i e issues, he
oxida i e capaci y o WAT is ela i ely low. Fo ins ance,
O
2
consump ion pe kg we weigh in WAT is abou one-
en h o ha o he es ing skele al muscle [41]. In WAT
issue, high ae obic capaci y up egula ed dime hylglycine
and down egula ed be aine ( ime hylglycine) (Fig. 5)sug-
ges ing inc eased me hyla ion o homocys eine o me hio-
nine, which has a ious unc ions in me abolism [42].
Howe e , he possible ole o be aine in HCR WAT
emains o be shown. Oc anoylca ni ine was also down-
egula ed (Fig. 5) sugges ing mo e e icien a y acid
oxida ion wi h highe ae obic capaci y in WAT. We ob-
se ed a di e ence only in h ee me aboli es, indica ing
ela i ely small di e ences be ween HCR and LCR a s in
WAT me abolism.
Aging was associa ed wi h he mos p ominen changes
in se um, muscle, and WAT me aboli es
In ou s udy se up, aging was associa ed wi h he mos
p ominen changes in se um, muscle, and WAT (Figs. 3,4,
Fig. 4 Venn diag am o di e en ially exp essed me aboli es wi h
ae obic capaci y, age, and hei in e ac ion in muscle. A ows
ep esen he di ec ion o eg ession coe icien ; ha in he
o e lapping egion (ae obic capaci y * age) may o igina e om
ae obic capaci y (1s a ow) and age (2nd a ow) o hei in e ac-
ion (3 d a ow) o bo h
Ge oScience