The Interaction between Circulating Cell-Free Mitochondrial DNA and Inflammatory Cytokines in Predicting Human Mental Health Issue Risk in Adolescents: An Explorative Study

Ci a ion: Ala alo, A.; de Sousa
Maciel, I.; Kuchá iko á, N.; Chew, S.;
an Kamp, I.; Fo as e , M.; Jul ez, J.;
Kanninen, K.M. The In e ac ion
be ween Ci cula ing Cell-F ee
Mi ochond ial DNA and
In lamma o y Cy okines in
P edic ing Human Men al Heal h
Issue Risk in Adolescen s: An
Explo a i e S udy. Biomedicines 2023,
11, 818. h ps://doi.o g/10.3390/
biomedicines11030818
Academic Edi o : Shake A. Mousa
Recei ed: 30 Janua y 2023
Re ised: 23 Feb ua y 2023
Accep ed: 5 Ma ch 2023
Published: 7 Ma ch 2023
Copy igh : © 2023 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license (h ps://
c ea i ecommons.o g/licenses/by/
4.0/).
biomedicines
A icle
The In e ac ion be ween Ci cula ing Cell-F ee Mi ochond ial
DNA and In lamma o y Cy okines in P edic ing Human Men al
Heal h Issue Risk in Adolescen s: An Explo a i e S udy
A o Ala alo 1, Izaque de Sousa Maciel 1, Nina Kuchá iko á1, Sweelin Chew 1, I ene an Kamp 2,
Ma ia Fo as e 3,4,5,6, Jo di Jul ez 3,7 and Ka ja M. Kanninen 1,*
1A.I. Vi anen Ins i u e o Molecula Sciences, Uni e si y o Eas e n Finland, 70211 Kuopio, Finland
2Na ional Ins i u e o Public Heal h and he En i onmen , 3721 MA Bil ho en, The Ne he lands
3ISGlobal, 08036 Ba celona, Spain
4Uni e si a Pompeu Fab a (UPF), 08005 Ba celona, Spain
5CIBER Epidemiología y Salud Pública (CIBEREsp), 28029 Mad id, Spain
6PHAGEX Resea ch G oup, Blanque na School o Heal h Science, Uni e si a Ramon Llull (URL),
08025 Ba celona, Spain
7
Clinical and Epidemiological Neu oscience G oup (Neu oÈpia), Ins i u d’ In es igacióSani à ia Pe e Vi gili (IISPV),
43007 Reus, Spain
*Co espondence: [email p o ec ed]
Abs ac :
Adolescence is o en a challenging ime in which psychia ic issues ha e a s ong connec ion
o men al heal h diso de s la e in li e. The ea ly iden i ica ion o he p oblems can educe he bu den
o disease. To da e, he e ec i e iden i ica ion o adolescen s a isk o de eloping men al heal h
p oblems emains unde s udied. Al oge he , he in e ac ion be ween ci cula ing cell- ee m DNA
(cc -m DNA) and in lamma o y cy okines in adolescen s is insu icien ly unde s ood ega ding
expe ienced men al heal h di icul ies. Ou s udy selec ed he pa icipan s based on he S eng h
and Di icul y Ques ionnai e (SDQ) sco e using he cu -o poin s o 3 and 18 o he low and he
high sco e g oups, espec i ely. The answe s o he SDQ a he age o 12.2–15.7 yea s con ibu ed
o he in es iga ion o (i) whe he cc -m DNA uni s a e associa ed wi h cy okines, and (ii) i an
in e ac ion model o p edic ing isk o men al heal h issues is obse ed. We disco e ed a sex-speci ic
co ela ion be ween he sc eened ma ke s associa ed wi h men al heal h p oblems in he low and
high SDQ sco e g oups among he male pa icipan s and in he low SDQ sco e g oup among he
emale pa icipan s. The mi ochond ial MT-ND4 and MT-CO1 genes co ela ed signi ican ly wi h
in e leukin-12p70 (IL-12p70) in males and wi h monocy e chemoa ac an p o ein-1 (MCP-1) in
emales. Due o he na u e o he explo a i e s udy, he s udied ma ke s alone did no indica e
s a is ical signi icance o he p edic ion o men al heal h p oblems. Ou analysis p o ided new
insigh in o po en ial plasma-based bioma ke s o p edic men al heal h issues.
Keywo ds: men al heal h; adolescence; mi ochond ial DNA; bioma ke s; oxida i e s ess; in lamma ion
1. In oduc ion
The de elopmen o he b ain is c i ical be ween he ages o 12 and 20 [1]. Due o he
apid de elopmen o he cen al ne ous sys em and co ical con ol egions, excessi e
isk- aking beha io and psychia ic issues expe ienced in ea ly adolescence ha e a s ong
link o men al heal h p oblems in adul li e [
2
]. P io o he COVID-19 pandemic, global
s a is ics es ima ed he gene al p e alence o diagnosed men al heal h diso de s o cause
dea h o disabili y in he case o 13.5% o 10-14-yea -old (ea ly) adolescen s a e diagno-
sis [
3
]. The pandemic u he inc eased men al heal h p oblems due o social and physical
es ic ions—a s udy by G aupenspe ge e al. indica ed signi ican inc eases in dep ession
symp oms (p< 0.01) and loneliness (p< 0.001) du ing he ini ial phase o he COVID-19
pandemic [
4
]. Ea ly iden i ica ion o men al heal h p oblems is cen al o educing he
Biomedicines 2023,11, 818. h ps://doi.o g/10.3390/biomedicines11030818 h ps://www.mdpi.com/jou nal/biomedicines
Biomedicines 2023,11, 818 2 o 14
numbe o indi iduals ha su e om men al illness, imp o ing he quali y o li e o
a ec ed indi iduals, and educing socioeconomic cos s. App oxima ely 50% o young
adul s wi h he long- e m NEET (no in educa ion, employmen , o aining) s a us ha e
been diagnosed wi h a psychia ic diso de in adolescence [
5
]. Ho mone le els and he
onse age o pube y ha e been widely s udied as con ibu ing ac o s o b ain de elopmen
be ween he sexes. Inc eased es os e one and dehyd oepiand os e one (DHEA) le els in
males and es adiol le els in emales a e associa ed wi h physical changes as an indica o o
pube y [
6
]. In pa icula , he de elopmen o psychia ic diso de s a ound he ime o he
inc eased sex ho mone le els may be sex-speci ically implica ed wi h b ain de elopmen .
Fu he mo e, biological, cogni i e, and emo ional di e ences can p ecede he exp ession o
cogni i e p oblems ha a e ound o a y be ween males and emales [7].
To da e, a model combining psychia ic and molecula me hods o he e ec i e
iden i ica ion o indi iduals a heigh ened isk o de eloping men al heal h p oblems in
adolescence does no exis . In psychia ic s udies, he SDQ is used as a ool o de e mine a
gene al sco e o p oblem beha io and o gene a e sepa a e sco es o emo ional symp oms,
conduc p oblems, hype ac i i y, pee ela ionship p oblems, and p osocial beha io [
8
,
9
].
The low and high SDQ sco e g oups can be ca ego ized using a summa y men al heal h
index ( he p- ac o ), which is desc ibed as a single- ac o indica o ha p edisposes people
o psychopa hology [
10
]. The SDQ o al scale is a commonly used gene al psychopa hology
measu e [
11
]. A high p- ac o sco e indica es la ge li e impai men , and poo e de el-
opmen al his o ies [
10
]. Fu he mo e, he in e nalizing and ex e nalizing sides o he
p- ac o measu e obse able symp oms o anxie y and dis up i e beha io s o conduc
diso de , espec i ely [
11
]. De ailed in o ma ion on he SDQ sco es is desc ibed in Ma e ials
and Me hods.
The immune-in lamma o y bioma ke al e a ions a e cons an ly associa ed wi h psy-
chia ic condi ions and nega i e clinical ou comes. Recen s udies epo ed blood-based
ma ke s, including mi ochond ial and cy okine con en s, o majo dep essi e diso de ,
bipola diso de , and schizoph enia in adul s [
12
,
13
]. In de ail, mi ochond ia egula e
ene gy p oduc ion, lipid me abolism, and edox s a us as a pa o main aining cellula
homeos asis. Du ing cellula s a es o dyshomeos asis, mi ochond ial espi a ion gene a es
inc eased le els o eac i e oxygen species (ROS), a phenomenon known o dis u b cellula
unc ion and pa hological condi ions [
14
,
15
]. Excessi e mi ochond ial ROS (m ROS) also
damages and agmen s mi ochond ial DNA (m DNA) [
15
,
16
]. m DNA has been desc ibed
as an agonis o he immune sys em, ia damage-associa ed molecula pa e ns (DAMPs)
whe e dying cells elease endogenous molecules in o he ex acellula en i onmen [
17
].
The m DNA agmen s om he damaged mi ochond ia a e eleased in o he cy osol by
au ophagy. Cell- ee mi ochond ial DNA (c -m DNA) is ecognized by he DAMP-speci ic
ecep o s, including Toll-like ecep o s (TLRs), leading o he ac i a ion o immune cells,
and he igge ing o an in lamma o y eac ion [
16
,
18
]. Comp ehensi ely, cy okines in
plasma ha e been b oadly iden i ied as he p ominen diagnos ic bioma ke s. Fo example,
ac i a ed mac ophages a e a majo sou ce o cy okines and in lamma o y media o s, o
which in e leukin-1 (IL-1) is p oduced ea ly in he ac i a ion phase [
19
,
20
]. Along wi h
he plasma samples, signi ican al e a ion o cy okines IL-1
β
, IL-6, IL-10, and IL-1RA was
de ec ed in he p e on al co ex o dep essed indi iduals who died by suicide compa ed
wi h nonpsychia ic con ols [
21
]. A co ela ion be ween c -m DNA le els and in e leukin-4
(IL-4) p oduced by T-helpe 2 (Th2) cells has p e iously been epo ed [13].
Biomedicines 2023,11, 818 3 o 14
C -m DNA becomes cc -m DNA when en e ing he ex acellula luids. S udies ocus-
ing on cc -m DNA uni al e a ions pos ula ed his phenomenon as a po en ial indica o
and diagnos ic ool in ea ly-s age sc eening and p ognosis o men al diso de s [
13
,
22
,
23
].
Kageyama e al. showed cc -m DNA le els o be signi ican ly educed in adul s su e ing
om majo dep essi e diso de (MDD) and bipola diso de (BD) [
13
]. Simila ly, Gonçal es
e al. epo ed highe le els o plasma cc -m DNA a e associa ed wi h la e -li e dep ession
in indi iduals o e he age o 60 [
24
]. Con e sely, li le in o ma ion is known abou he
cc -m DNA le els in young indi iduals su e ing om a men al diso de o psychological
s ess. Jeong e al. epo ed an inc eased bu no signi ican di e ence in he cc -m DNA
le els be ween he diagnos ic g oups o BD and he heal hy con ols a a mean age o
17.0 and 15.5 yea s old, espec i ely [
25
]. The e o e, he consensus on he po en ial o
cc -m DNA as a p edic i e bioma ke o men al heal h dys unc ion is con o e sial. Fu -
he mo e, exis ing s udies ha e no assessed how he sex o pa icipan s in luences he
cc -m DNA le el.
Based on he epo s ha (i) childhood auma is associa ed wi h ele a ed le els o
p o-in lamma o y cy okines [
26
], (ii) in e leukin-10 (IL-10) is dec eased in symp oma ic
adolescen s wi h BD compa ed o heal hy con ols [
27
], and (iii) he epo ed nominal
co ela ion be ween IL-4 and cc -m DNA le el in adul s wi h MDD [
13
], we hypo hesized
ha he e is a connec ion be ween cc -m DNA and in lamma o y cy okines in adolescen s
expe iencing in e nalized o ex e nalized men al heal h ou comes. We in es iga ed whe he
cc -m DNA le els a e associa ed wi h in lamma o y cy okines and can be used as bioma k-
e s o p edic ing men al heal h issues in adolescen s. Plasma le els o cc -m DNA and
cy okines we e measu ed in adolescen s in he low and high SDQ g oups. He e we epo
a sex-dependen co ela ion o he cc -m DNA uni wi h in lamma o y cy okine le els
in ela ion o he isk o men al heal h dys unc ion. We summa ize ha he cc -m DNA
uni is connec ed o IL-12p70 le els in males and MCP-1 le els in emales. Toge he , hese
indings indica e no el bioma ke panels o men al heal h isks in adolescen s.
2. Ma e ials and Me hods
2.1. Coho and Samples
Walnu s is a egional, con olled, andomized clinical Spanish coho o iginally used
o s udies on he ole o selec ed a y acids om walnu in ake associa ed wi h neu opsy-
chological and physical heal h. Fo he cu en s udy, he SDQ-based subsamples we e
selec ed om he baseline popula ion be o e he walnu in e en ion, as desc ibed in he
nex sec ion. The s udy pa icipan s we e om Ba celona, Spain, and none o hem we e
aking medica ion a he ime o pa icipa ing in he s udy. The pe iphe al blood collec ion
om all he pa icipan s was pe o med by ollowing a s anda dized p o ocol du ing he
school day in 2016–2018 [
28
]. Fas ing samples we e no included in he s udy. Blood was
collec ed o EDTA Plus ubes (BD Biosciences, San Jose, CA, USA), in e ed 6 imes, and
cen i uged a 2500 c o 20 min a 4
◦
C. The plasma laye was sepa a ed om he ed
blood and he bu y coa laye s and collec ed o s e ile ubes o s o age a
−
80
◦
C p io
o analyses.
2.2. S eng hs and Di icul ies Ques ionnai e and Eligibili y C i e ia
App oxima ely coinciden ally wi h he collec ion o blood (Table 1), he pa icipan s
illed ou a sel - epo ed e sion o he SDQ es . The ques ionnai e measu ed a gene al
sco e o p oblem beha io and i e subscales aimed o assess emo ional symp oms, conduc
p oblems, hype ac i i y, pee ela ionship p oblems, and p osocial beha io [
9
]. The cu -o
poin o 18 o he high SDQ sco e g oup ollowed he ecoded ca ego iza ion o he SDQ
sco e [
8
]. In he case o he low SDQ g oup, he cu -o poin o 3 was selec ed manually o
balance he size o each g oup (Table 1). The p opo ions o speci ic SDQ sco es in he low
and high g oups a e shown in Figu e 1. The o al SDQ sco e was calcula ed as a measu e
psychopa hology index, summing hype ac i i y, emo ional symp oms, conduc p oblems,
Biomedicines 2023,11, 818 4 o 14
and pee p oblems [
8
,
11
]. SDQ is a well-es ablished and widely used index o p o ide a
sco e anging om 0–40 (0–14 = low, 15–17 = bo de line, 18–40 = high) [8,29].
Biomedicines2023,11,xFORPEERREVIEW4o 14

calcula edasameasu epsychopa hologyindex,summinghype ac i i y,emo ional
symp oms,conduc p oblems,andpee p oblems[8,11].SDQisawell‐es ablishedand
widelyusedindex op o ideasco e anging om0–40(0–14=low,15–17=bo de line,
18–40=high)[8,29].

Figu e1.F ac iono SDQsco esamong heWalnu spa icipan samong(a)lowSDQsco es( alues
o 0–3)and(b)highSDQsco es( alueso 18–25).
Table1.Demog aphicin o ma ionand he angeo  heSDQsco e, heexecu ionageo SDQand
blood es s,and hesexdis ibu iono s udysubjec s.
G oupSco eAgeo SDQTes 
(Yea s)
Ageo BloodTes 
(Yea s)Samples
MeanSDMeanSDnMaleFemale
LowSDQ0–313.480.8013.510.8119109
HighSDQ18–2514.111.0714.151.1520812
2.3.Ex ac ionandQuan i ica iono Plasmacc ‐m DNA
Ci cula ingcell‐ eegenomicDNAwasex ac ed om200μLo plasmabyusing he
QIAampMinElu eVi usSpinKi (Qiagen,Hilden,Ge many)acco ding o hemanu ac‐
u e ’sp o ocol.The inal olumeo  heelu edcc ‐m DNAwas50μL, omwhich he
DNAconcen a ionwasmeasu edwi h heNanoD opND‐1000spec opho ome e 
(The moFishe Scien i ic,Wal ham,МА,USA).We ocusedon womi ochond ialgenes
ha code o sepa a ecomplexeso  heelec on anspo chain.ND4codes o complexI
andCO1codes o complexIV.TaqManassays(The moFishe ,Ca lsbad,CA,USA)we e
used oquan i y wo egionso humanm DNAin wo eplica e unso qPCR.TheND4
andCO1geneswe equan i iedwi hTaqMan(R)GeneExp essionassaysHs02596876_g1
andHs02596864_g1, espec i ely(TableS1).The o al eac ion olumewas10μLcon ain‐
ing5.0μLo MaximaP obe/ROXqPCRMas e Mix(The moScien i ic,Ca lsbad,CA,
USA),2.9μLo  heDNA empla e,0.5μLo p ime ,and1.6μLo nuclease‐ eewa e 
(The moScien i ic,Ca lsbad,CA,USA).ThegenomicDNA empla ewasdilu ed oa
concen a iono 7.0ng/μLwi hs e ilewa e (Bax e ,Mississauga,ON,Canada).
Thegenequan i ica ionwaspe o medusing heS epOnePlus™Real‐TimePCR
Sys em(AppliedBiosys ems™,Ca lsbad,CA,USA) o p epa ing hes anda dcu es
and unning heex ac edgDNAsamples.Thes anda dp o ocol o quan i ica ionin‐
cludedini ialdena u a ionandholdings epsa 50°C o 2.0minand95°C o 10min
ollowedbycyclings epsa 95°C o 15sand60°C o 1.0min o a o alleng ho 40
cycles.Alls epswe epe o medas ecommendedby hemanu ac u e .
As anda dcu ewasp epa ed ocalcula eand alida e heuni o eachmi ochon‐
d ialgenebyusingcomme ciallymanu ac u edm DNAplasmidsacco ding o hep o‐
ocolo AppliedBiosys ems[30].Theplasmids o  hes anda dcu ewe esyn hesized
comme cially(Azen aLi eSciences,Suzhou,China).Theinse so ND4(233n )andCO1
Figu e 1.
F ac ion o SDQ sco es among he Walnu s pa icipan s among (
a
) low SDQ sco es ( alues
o 0–3) and (b) high SDQ sco es ( alues o 18–25).
Table 1.
Demog aphic in o ma ion and he ange o he SDQ sco e, he execu ion age o SDQ and
blood es s, and he sex dis ibu ion o s udy subjec s.
G oup Sco e Age o SDQ Tes
(Yea s)
Age o Blood Tes
(Yea s) Samples
Mean SD Mean SD nMale Female
Low SDQ 0–3 13.48 0.80 13.51 0.81 19 10 9
High SDQ 18–25 14.11 1.07 14.15 1.15 20 8 12
2.3. Ex ac ion and Quan i ica ion o Plasma cc -m DNA
Ci cula ing cell- ee genomic DNA was ex ac ed om 200
µ
L o plasma by using
he QIAamp MinElu e Vi us Spin Ki (Qiagen, Hilden, Ge many) acco ding o he man-
u ac u e ’s p o ocol. The inal olume o he elu ed cc -m DNA was 50
µ
L, om which
he DNA concen a ion was measu ed wi h he NanoD op ND-1000 spec opho ome e
(The mo Fishe Scien i ic, Wal ham, MA, USA). We ocused on wo mi ochond ial genes
ha code o sepa a e complexes o he elec on anspo chain. ND4 codes o complex I
and CO1 codes o complex IV. TaqMan assays (The mo Fishe , Ca lsbad, CA, USA) we e
used o quan i y wo egions o human m DNA in wo eplica e uns o qPCR. The ND4 and
CO1 genes we e quan i ied wi h TaqMan(R) Gene Exp ession assays Hs02596876_g1 and
Hs02596864_g1, espec i ely (Table S1). The o al eac ion olume was 10
µ
L con aining
5.0
µ
L o Maxima P obe/ROX qPCR Mas e Mix (The mo Scien i ic, Ca lsbad, CA, USA),
2.9
µ
L o he DNA empla e, 0.5
µ
L o p ime , and 1.6
µ
L o nuclease- ee wa e (The mo
Scien i ic, Ca lsbad, CA, USA). The genomic DNA empla e was dilu ed o a concen a ion
o 7.0 ng/µL wi h s e ile wa e (Bax e , Mississauga, ON, Canada).
Biomedicines 2023,11, 818 5 o 14
The gene quan i ica ion was pe o med using he S epOnePlus
™
Real-Time PCR
Sys em (Applied Biosys ems
™
, Ca lsbad, CA, USA) o p epa ing he s anda d cu es and
unning he ex ac ed gDNA samples. The s anda d p o ocol o quan i ica ion included
ini ial dena u a ion and holding s eps a 50
◦
C o 2.0 min and 95
◦
C o 10 min ollowed
by cycling s eps a 95
◦
C o 15 s and 60
◦
C o 1.0 min o a o al leng h o 40 cycles. All
s eps we e pe o med as ecommended by he manu ac u e .
A s anda d cu e was p epa ed o calcula e and alida e he uni o each mi ochond ial
gene by using comme cially manu ac u ed m DNA plasmids acco ding o he p o ocol
o Applied Biosys ems [
30
]. The plasmids o he s anda d cu e we e syn hesized com-
me cially (Azen a Li e Sciences, Suzhou, China). The inse s o ND4 (233 n ) and CO1
(177 n ) genes we e liga ed in o he pUC-GW-Amp plasmids du ing he manu ac u ing
p ocess (Figu e S1). The concen a ion ange o he s anda d cu e was 3.0
×
10
8
o
3.0
×
10
4
uni /
µ
L. The coe icien o de e mina ion had o be abo e 0.99 be o e he cu e
was app o ed. The uni le el o plasma cc -m DNA was based on he s anda d cu es
o each quan i ied m DNA egion. Values a e exp essed in plasmid uni s pe mic oli e
o plasma.
2.4. Cy okine Bead A ay
The sec e ed le els o cy okines in human plasma samples we e measu ed using he
Cy ome ic Bead A ay (BD Biosciences, San Jose, CA, USA) along wi h Human Soluble
P o ein Mas e Bu e Ki (BD Biosciences, San Jose, CA, USA) acco ding o he man-
u ac u e
´
s p o ocol. IL-12p70 and monocy e chemoa ac an p o ein-1 (MCP-1) we e
selec ed om he po en ial cy okines a e sc eening. Cy okine measu emen was pe -
o med wi h he Cy oFLEX S Flow Cy ome e (Beckman Coul e , Indianapolis, IN, USA).
Acqui ed cy okine da a we e analyzed in he FCAP A ayTM 2.0.1 so wa e (So low Inc.,
New B igh on, MN, USA).
2.5. S a is ical Analyses
The cc -m DNA uni s we e epo ed as mean
±
s anda d de ia ion. The - es along wi h
in e -assay coe icien s o a ia ion was used o analyze di e ences be ween ND4 and CO1 uni
le els. Unpai ed Mann–Whi ney U- es was used o compa e means be ween he SDQ-based
g oups. All es s we e wo- ailed and a p- alue
≤
0.05 indica ed a signi ican di e ence in
means. A nonpa ame ic Spea man’s Rho was used o analyze he co ela ion be ween SDQ
sco es, cy okine le els, and cc -m DNA uni . The cc -m DNA uni da a we e es ed o con i m
he usage o Spea man’s Rho by quan ile–quan ile plo (QQ plo ), which indica ed ha he
da a a e no ollowing a Gaussian dis ibu ion. In he co ela ion analysis, 0.0
≤
| |
≤
0.2
e e ed o no co ela ion, 0.2 < | |
≤
0.4 o he low co ela ion, 0.4 < | |
≤
0.6 o he mode a e
co ela ion, 0.6 < | |
≤
0.8 o he high co ela ion, and 0.8 < | |
≤
1.0 o he e y high
co ela ion. The absolu e alue o e ec size was calcula ed using Cohen’s d. The alues
we e di ided in o a small e ec (d > 0.2), a medium e ec (d > 0.5), and a la ge e ec size
(d > 0.8) [
12
]. All analyses we e ca ied ou in G aphPad P ism 9.1.1 (G aphPad So wa e,
San Diego, CA, USA) o Mic oso Excel e sion 2122 (Mic oso , Redmond, WA, USA)
3. Resul s
3.1. Cc -m DNA Uni Le el Does No Co ela e wi h SDQ Sco es o Sex in Adolescen s
We i s es ed he di e ence in he uni le els be ween he mi ochond ial CO1 and
ND4 genes and no signi ican di e ence was obse ed (p= 0.185). Addi ionally, o e i y a
need o wo cc -m DNA ma ke s, we calcula ed 33.0% and 37.8% in e -assay coe icien s
o a ia ion in he uni le els o ND4 and CO1 in he cases o he low SDQ sco e g oup and
he high SDQ sco e g oup, espec i ely.
Nex , we e alua ed he ela ionship o he sel - epo ed SDQ sco e and sex wi h he
exp ession o he mi ochond ial ND4 gene in plasma samples. The e was no signi ican
di e ence in ND4 le els be ween he low and high SDQ sco e g oups (Figu e 2a) o sexes
(Figu e 2b). The absolu e alue o e ec size d was 0.424 and 0.047, as shown in Figu e 2a,b,

Biomedicines 2023,11, 818 6 o 14
espec i ely. The co ela ion o ND4 uni le els and SDQ sco e did no indica e a signi ican
co ela ion (Figu e 2c; =
−
0.177, p= 0.302). Subsequen ly, we analyzed he e ec s o
he SDQ sco e and sex on he exp ession o he mi ochond ial CO1 gene. The e was no
signi ican di e ence in CO1 gene exp ession be ween he low and high SDQ g oups
(Figu e 2d) o sexes (Figu e 2e). The absolu e alue o e ec size d was 0.326 and 0.109,
as shown in Figu e 2d,e, espec i ely. CO1 uni le els and he SDQ sco e did no indica e
signi ican co ela ions (Figu e 2 ; = −0.252, p= 0.132).
Biomedicines2023,11,xFORPEERREVIEW6o 14


Figu e2.Compa isono plasmacc ‐m DNAuni le elsbe weenlowandhighSDQg oups,includ‐
ingsex‐speci icdi e ences.(a)ND4uni le elinlow(n=18)andhigh(n=19)SDQg oups,(b)ND4
uni le elinmales(n=16)and emales(n=20),(c)in e ac iono  heSDQsco ewi h hele elo 
ND4uni s,(d)CO1uni le elin helowandhighSDQg oups(n=18/g oup),(e)CO1uni le elin
males(n=16)and emales(n=21),( )in e ac iono  heSDQsco ewi h hele elo CO1uni .No e:
ou lie swe e emo edsepa a ely omeachg aphbasedon heusedme hod.ns:non‐signi ican .
3.2.Cy okineLe elsA eNo Rela ed oSDQSco eso SexinAdolescen s
Toassesswhe he cy okineswe eapo en ialindica o o  heinc eased isk o 
men alheal hdiso de sinadolescen s,weanalyzedplasmale elso IL‐12p70andMCP‐
1.Bo hcy okineswe edesc ibedas hedowns eamcy okineso IL‐1β inaDAMP‐
associa edin lamma ion[31,32].
Thee ec o  hesel ‐ epo edSDQsco eandsexon hecy okinele elo IL‐12p70was
s udiedusing hecy okinebeada ay.The ewasnosigni ican di e enceinIL‐12p70le els
be ween helowandhighSDQsco eg oups(Figu e3a),o be weenmalesand emales
(Figu e3b).Theabsolu e alueo e ec sizedwas0.624and0.219,asshowninFigu es3a
and3b, espec i ely.The ewasnosigni ican co ela iono IL‐12p70concen a ionand
SDQsco es(Figu e3c; =0.213,p=0.243).Simila ly,MCP‐1le elswe eno a ec edby he
SDQsco e(Figu e3d)o sex(Figu e3e).Theabsolu e alueo e ec sizedwas0.237and
0.160,asshowninFigu es3eand3 , espec i ely.Theco ela iono MCP‐1concen a ion
and heSDQsco ewasno signi ican (Figu e3 ; =−0.014,p=0.934).
Figu e 2.
Compa ison o plasma cc -m DNA uni le els be ween low and high SDQ g oups, including
sex-speci ic di e ences. (
a
)ND4 uni le el in low (n= 18) and high (n= 19) SDQ g oups, (
b
)ND4
uni le el in males (n= 16) and emales (n= 20), (
c
) in e ac ion o he SDQ sco e wi h he le el o
ND4 uni s, (
d
)CO1 uni le el in he low and high SDQ g oups (n= 18/g oup), (
e
)CO1 uni le el in
males (n= 16) and emales (n= 21), (
) in e ac ion o he SDQ sco e wi h he le el o CO1 uni . No e:
ou lie s we e emo ed sepa a ely om each g aph based on he used me hod. ns: non-signi ican .
3.2. Cy okine Le els A e No Rela ed o SDQ Sco es o Sex in Adolescen s
To assess whe he cy okines we e a po en ial indica o o he inc eased isk o men al
heal h diso de s in adolescen s, we analyzed plasma le els o IL-12p70 and MCP-1. Bo h
cy okines we e desc ibed as he downs eam cy okines o IL-1
β
in a DAMP-associa ed
in lamma ion [31,32].
The e ec o he sel - epo ed SDQ sco e and sex on he cy okine le el o IL-12p70
was s udied using he cy okine bead a ay. The e was no signi ican di e ence in IL-12p70
le els be ween he low and high SDQ sco e g oups (Figu e 3a), o be ween males and
emales (Figu e 3b). The absolu e alue o e ec size d was 0.624 and 0.219, as shown in
Figu e 3a,b, espec i ely. The e was no signi ican co ela ion o IL-12p70 concen a ion
and SDQ sco es (Figu e 3c; = 0.213, p= 0.243). Simila ly, MCP-1 le els we e no a ec ed
by he SDQ sco e (Figu e 3d) o sex (Figu e 3e). The absolu e alue o e ec size d was 0.237
and 0.160, as shown in Figu e 3e, , espec i ely. The co ela ion o MCP-1 concen a ion
and he SDQ sco e was no signi ican (Figu e 3 ; = −0.014, p= 0.934).
Biomedicines 2023,11, 818 7 o 14
Biomedicines2023,11,xFORPEERREVIEW7o 14


Figu e3.Compa isono plasmacy okinele elsbe weenlowandhighSDQg oups,includingsex‐
speci icchanges.(a)Di e enceinIL‐12p70exp essionle elbe ween helow(n=16)andhighSDQ
g oup(n=17),(b)di e enceinIL‐12p70exp essionle elbe weenmale(n=13)and emale(n=20),
(c)in e ac iono  heSDQsco ewi h hele elo IL‐12p70,(d)di e enceinMCP‐1exp essionle el
be ween helowandhighSDQg oup(n=18/g oup),(e)di e enceinMCP‐1exp essionle elbe ween
male(n=18)and emale(n=20),( )in e ac iono  heSDQsco ewi h hele elo MCP‐1.No e:ou lie s
we e emo edsepa a ely omeachg aphbasedon heusedme hod.ns:non‐signi ican .
3.3.Mi ochond ialGenesCo ela ewi hIL‐12p70o MCP‐1inaSex‐Speci icFashion
Wenex analyzed heco ela ionbe ween hecc ‐m DNAma ke s,ND4andCO1,
andin lamma o ycy okines,IL‐12p70andMCP‐1,sepa a edbysex.Inmales,IL‐12p70
co ela edmode a elywi hbo hND4(Figu e4a; =0.476,p=0.122)andwi hCO1(Figu e
4b; =0.518,p=0.089).Addi ionally, he ewasalowco ela ionbe weenMCP‐1andND4
(Figu e4c; =0.306,p=0.288)andCO1(Figu e4d; =0.306,p=0.288).In emales,onlya
lowco ela ionbe weenIL‐12p70andND4(Figu e4e; =−0.230,p=0.344)o CO1(Figu e
4 ; =−0.254,p=0.280)wasobse ed.Addi ionally,we oundMCP‐1co ela ed
mode a elywi hbo hND4(Figu e4g; =−0.593,p=0.008)andCO1(Figu e4h; =−0.522,
p=0.018)in emales.
Figu e 3.
Compa ison o plasma cy okine le els be ween low and high SDQ g oups, including
sex-speci ic changes. (
a
) Di e ence in IL-12p70 exp ession le el be ween he low (n= 16) and high
SDQ g oup (n= 17), (
b
) di e ence in IL-12p70 exp ession le el be ween male (n= 13) and emale
(n= 20), (
c
) in e ac ion o he SDQ sco e wi h he le el o IL-12p70, (
d
) di e ence in MCP-1 exp ession
le el be ween he low and high SDQ g oup (n= 18/g oup), (
e
) di e ence in MCP-1 exp ession
le el be ween male (n= 18) and emale (n= 20), (
) in e ac ion o he SDQ sco e wi h he le el
o MCP-1. No e: ou lie s we e emo ed sepa a ely om each g aph based on he used me hod.
ns: non-signi ican .
3.3. Mi ochond ial Genes Co ela e wi h IL-12p70 o MCP-1 in a Sex-Speci ic Fashion
We nex analyzed he co ela ion be ween he cc -m DNA ma ke s, ND4 and CO1,
and in lamma o y cy okines, IL-12p70 and MCP-1, sepa a ed by sex. In males, IL-12p70
co ela ed mode a ely wi h bo h ND4 (Figu e 4a; = 0.476, p= 0.122) and wi h CO1
(Figu e 4b; = 0.518, p= 0.089). Addi ionally, he e was a low co ela ion be ween MCP-1
and ND4 (Figu e 4c; = 0.306, p= 0.288) and CO1 (Figu e 4d; = 0.306, p= 0.288). In
emales, only a low co ela ion be ween IL-12p70 and ND4 (Figu e 4e; =
−
0.230, p= 0.344)
o CO1 (Figu e 4 ; =
−
0.254, p= 0.280) was obse ed. Addi ionally, we ound MCP-1
co ela ed mode a ely wi h bo h ND4 (Figu e 4g; =
−
0.593, p= 0.008) and CO1 (Figu e 4h;
= −0.522, p= 0.018) in emales.
Biomedicines 2023,11, 818 8 o 14
Biomedicines2023,11,xFORPEERREVIEW8o 14


Figu e4.Co ela ionanalysiso cy okinele elsandcc ‐m DNAuni s.(a–d)Co ela ionanalysiso 
plasmaIL‐12p70andMCP‐1le elswi h hoseo ND4andCO1uni sinmales.(e–h)Co ela ion
analysiso plasmaIL‐12p70andMCP‐1le elswi h hoseo ND4andCO1uni sin emales.(i)Clus‐
e inghea mapanalysis o co ela iono  hecy okinesIL‐12p70andMCP‐1and hegenesND4
andCO1uni sinall hesamples,andsepa a edbysex.(*Highco ela ion0.6<| |≤0.8,** e y
highco ela ion0.8<| |≤1.0.)
Figu e 4.
Co ela ion analysis o cy okine le els and cc -m DNA uni s. (
a
–
d
) Co ela ion analysis
o plasma IL-12p70 and MCP-1 le els wi h hose o ND4 and CO1 uni s in males. (
e
–
h
) Co ela-
ion analysis o plasma IL-12p70 and MCP-1 le els wi h hose o ND4 and CO1 uni s in emales.
(
i
) Clus e ing hea map analysis o co ela ion o he cy okines IL-12p70 and MCP-1 and he genes
ND4 and CO1 uni s in all he samples, and sepa a ed by sex. (* High co ela ion 0.6 < | |
≤
0.8,
** e y high co ela ion 0.8 < | | ≤1.0.)
Subsequen ly, we es ed he in e ac ion o he mi ochond ial genes and in lamma o y
cy okines as desc ibed by Kageyama e al. [
13
]. The co ela ion be ween he cc -m DNA
Biomedicines 2023,11, 818 9 o 14
uni s and cy okine le els sepa a ed by sex we e assessed by he Spea man’s Rho (Figu e 4i).
In male subjec s, a e y high posi i e co ela ion o bo h mi ochond ial ND4 ( = 0.600,
p= 0.242) and CO1 ( = 0.829, p= 0.058) wi h IL-12p70 we e obse ed in he high SDQ
sco e g oup. Addi ionally, a high posi i e co ela ion o bo h mi ochond ial ND4 ( = 0.691,
p= 0.069) and CO1 ( = 0.762, p= 0.037) wi h hMCP-1 was obse ed in males o he low
SDQ sco e g oup. In emales, a e y high nega i e co ela ion o bo h mi ochond ial ND4
( =
−
0.917, p= 0.001) and CO1 ( =
−
0.800, p= 0.014) wi h MCP-1 was obse ed in
he low SDQ sco e g oup. No signi ican co ela ions be ween mi ochond ial genes and
cy okines we e obse ed in emales wi h a high SDQ sco e. Fu he mo e, IL-12p70 was no
signi ican ly co ela ed o he mi ochond ial genes o emales in ei he he low o high SDQ
sco e g oup.
4. Discussion
He e, we measu ed he co ela ion be ween cc -m DNA and cy okine ma ke s, im-
plica ing a po en ial connec ion whe e he ex acellula m DNA may igge an immune-
in lamma o y eac ion in subjec s a isk o psychopa hology. The esul s indica e, on he
whole, he exis ence o a speci ic co ela ion be ween he le els o cc -m DNA ma ke s
(ND4 and CO1) and in lamma o y ma ke s (IL-12p70 and MCP-1) o p edic an inc eased
isk o men al heal h issues compa ed wi h low- isk con ols. Toge he , hese indings
indica e p oposing no el bioma ke panels o men al heal h isks in adolescen s. Con-
e sely, ou s udy did no un eil he po en ial o using single cc -m DNAs o cy okines as
p edic i e ma ke s. Nex , i is impo an o comple e ou p edic i e model by analyzing
he ole o o he po en ial cy okine ma ke s in he in lamma o y pa hway. To he bes o
ou knowledge, his s udy epo s o he i s ime how he isk o men al heal h issues in
adolescen s is connec ed o plasma cc -m DNA uni le el, an indica o o oxida i e s ess,
and cy okine le els, indica o s o in lamma ion. Along wi h he analysis o he adolescen
samples, we also p o ided insigh in o he in lamma ion p ocess igge ed by cc -m DNA,
which is no a ec ed by aging o ch onic diseases. Ou esul s sugges ha in lamma o y
cy okine changes can po en ially be associa ed wi h cc -m DNA ma ke s in adolescen s,
bu he associa ion is sex-speci ic.
The genes o mi ochond ial complex I (ND1-ND6) a e he mos s udied genes o cc -
m DNA associa ed wi h men al heal h diso de s [
12
,
13
,
25
,
33
]. To da e, li le esea ch has
been conduc ed on he ole o he genes encoding o he mi ochond ial complexes. The e o e,
we decided o also pe o m quan i a i e analysis o he CO1 gene, pa o complex IV,
o disco e whe he he loca ion o he gene can a ec he esul s. The calcula ed in e -
assay coe icien o a ia ion in he cc -m DNA uni le el was abo e 30% o indica e he
impo ance o measu ing he uni le el o bo h genes. Con e sely, he - es did no indica e
signi ican di e ences be ween he uni le els o CO1 and ND4 (p= 0.185). Howe e , he
esul s om he coe icien o a ia ion es p o ided a eason o es he exp ession o bo h
genes in he plasma samples. Ou esul s did no indica e signi ican a ia ion in he uni
le els o ND4 and CO1 be ween sexes o be ween he SDQ sco e g oups (Figu e 2), which
emphasizes he eliabili y o he p ocess. I also sugges s ha endogenous di e ences do
no impac he quan i ica ion o he s udied cc -m DNA genes. Addi ionally, a ela i ely
small sample numbe a ec ed he e ec size o his s udy. The absolu e alue o e ec size
d was 0.33–0.42 be ween he low SDQ and high SDQ sco e g oups in he quan i ica ion o
m DNA uni le els.
P e iously, Lindq is e al. and Kageyama e al. epo ed cc -m DNA uni s among
adul s wi h a diagnosed men al heal h diso de when compa ed o heal hy con ols. How-
e e , he Kageyama e al. pape epo ed a dec ease, while he Lindq is e al. s udy showed
an inc ease in cc -m DNA le el in indi iduals su e ing om a men al diso de [
12
,
13
].
Mo eo e , Jeong e al., s udied se um cc -m DNA uni le els in adolescen s wi h a diagnosis
o BD wi hou inding signi ican di e ences when compa ed o heal hy indi iduals [
25
].
Thei indings on adolescen s wi h an ac ual men al heal h diso de a e in line wi h he
esul s o ou s udy, which indica ed ha cc -m DNA uni le els a e no signi ican ly al e ed