1
Scien i ic RepoR S | (2020) 10:3061 | h ps://doi.o g/10.1038/s41598-020-59898-0
www.na u e.com/scien i ic epo s
H- e i in is essen ial o
mac ophages’ capaci y o s o e o
de oxi y exogenously added i on
Gonçalo Mesqui a1,2,7, Tânia Sil a
1,2,7, Ana C. Gomes
1,2,3, Ped o F. oli ei a1,4,5,
Ma co G. Al es
4, Rui e nandes
1,2, Agos inho A. Almeida
6, Ana C. Mo ei a1,2,3,8* &
Ma ia Salomé Gomes
1,2,3,8*
Mac ophages a e cen al cells bo h in he immune esponse and in i on homeos asis. I on is bo h
essen ial and po en ially oxic. The e o e, i on acquisi ion, anspo , s o age, and elease a e igh ly
egula ed, by se e al impo an p o eins. Cy osolic e i in is an i on s o age p o ein composed o 24
subuni s o ei he he L- o he H- ype chains. H- e i in di e s om L- e i in in he capaci y o oxidize
e2+ o e3+. In his wo k, we in es iga ed he ole played by H- e i in in he mac ophages’ abili y o
espond o immune s imuli and o deal wi h exogenously added i on. We used mice wi h a condi ional
dele ion o he H- e i in gene in he myeloid lineage o ob ain bone ma ow-de i ed mac ophages.
These mac ophages had no mal iabili y and gene exp ession unde basal cul u e condi ions. Howe e ,
when ea ed wi h in e e on-gamma and lipopolysaccha ide hey had a lowe ac i a ion o Ni ic
Oxide Syn hase 2. Fu he mo e, H- e i in-de icien mac ophages had a highe sensi i i y o i on-
induced oxici y. This sensi i i y was associa ed wi h a lowe in acellula i on accumula ion bu a highe
p oduc ion o eac i e oxygen species. These da a indica e ha H- e i in modula es mac ophage
esponse o immune s imuli and ha i plays an essen ial ole in p o ec ion agains i on-induced
oxida i e s ess and cell dea h.
I on is essen ial o almos all li ing o ganisms, due o i s key oles in biological p ocesses such as DNA and RNA
syn hesis, mi ochond ial espi a o y chain, cell p oli e a ion, and di e en ia ion, among o he s1. On he o he
hand, i on needs o be s ic ly con olled and s o ed, due o i s po en ial o cause cell damage2. P o eins like e -
i in, ans e in, and e opo in a e essen ial o con ol he in acellula and in e cellula i on luxes and imbal-
ances in hei exp ession o unc ion lead o se e e heal h p oblems3–5. In pa icula , e i in is essen ial o sa ely
s o e i on inside cells, a oiding i s ha m ul e ec s. This p o ein is composed o a 24 subuni s complex ha o ms a
nanocage, inside which i on is apped6. Fe i in subuni s can be o he H, L (cy osol) o M (mi ochond ia) ypes7.
The cy osolic subuni s ha e dis inc unc ions and a e p esen in di e en p opo ions in di e en cell ypes8.
H- e i in (FTH1) is essen ial o mammalians’ de elopmen , as he ull knock-ou o i s gene leads o emb y-
onic dea h in mice9. FTH1 di e s om L- e i in (FTL) in i s e oxidase ac i i y, which is esponsible o he
oxida ion o i on Fe2+ o Fe3+, p e en ing he oxida i e damage ha could be caused by Fen on eac ions in ol -
ing Fe2+ 10. These eac ions can gene a e oxygen eac i e species ha damage p o eins, lipids, and o he cell
componen s.
Mac ophages a e cen al playe s in i on me abolism. They ecycle senescen ed blood cells and nu se he
o ma ion o new ones11–13, hus pa icipa ing in he majo sys emic i on luxes occu ing in highe e eb a es.
Mac ophages a e also a i s line o de ence agains pa hogens, and hey modula e i on a ailabili y as pa o hos
p o ec i e mechanisms12,14–16.
1i3S – Ins i u o de In es igação e Ino ação em Saúde, Uni e sidade do Po o, Po o, Po ugal. 2IBMC – Ins i u o de
Biologia Molecula e Celula , Uni e sidade do Po o, Po o, Po ugal. 3Depa amen o de Biologia Molecula , ICBAS –
Ins i u o de Ciências Biomédicas Abel Salaza , Uni e sidade do Po o, Po o, Po ugal. 4Depa men o Mic oscopy,
Labo a o y o Cell Biology and Uni o Mul idisciplina y Resea ch in Biomedicine (UMIB), ICBAS, Uni e sidade do
Po o, Po o, Po ugal. 5Depa men o Gene ics, Facul y o Medicine (FMUP), Uni e sidade do Po o, Po o, Po ugal.
6LAQV/REQUIMTE, Depa amen o de Ciências Químicas, Faculdade de Fa mácia, Uni e sidade do Po o, Po o,
Po ugal. 7These au ho s con ibu ed equally: Gonçalo Mesqui a and Tânia Sil a. 8These au ho s join ly supe ised his
wo k: Ana C. Mo ei a and Ma ia Salomé Gomes. *email: [email p o ec ed]; [email p o ec ed]
open
2
Scien i ic RepoR S | (2020) 10:3061 | h ps://doi.o g/10.1038/s41598-020-59898-0
www.na u e.com/scien i ic epo s
www.na u e.com/scien i ic epo s/
Since mac ophages and i on me abolism a e igh ly connec ed and H- e i in is c ucial o i on s o age inside
he cells, we hypo hesized ha H- e i in has a key ole in mac ophages iabili y, de elopmen , ac i a ion, and
i on handling. To es his hypo hesis, we used mice wi h a condi ional dele ion o he H- e i in gene (F h1) in
he myeloid lineage, gene a ed om he H- e i in-loxP mice c ea ed by Lukas Kuhn and collabo a o s17. We
ob ained bone ma ow-de i ed mac ophages (BMDM) om hese mice and ound ha al hough in i o mac-
ophage di e en ia ion p oceeded no mally, H- e i in-de icien BMDM had sub le al e a ions in hei esponse
o immune s imula ion and a ma ked inc ease in suscep ibili y o oxida i e s ess and cell dea h induced by
exogenously added i on.
Resul s
H- e i in is no necessa y o in i o di e en ia ion o bone ma ow-de i ed mac-
ophages. Since H- e i in is essen ial o mouse de elopmen 9, we s a ed by e alua ing whe he i was also
necessa y o in i o di e en ia ion o mac ophages om hei bone ma ow (BM) p ecu so s. Bone ma ow-de-
i ed mac ophages (BMDM) we e ob ained om F h1Fl/Fl; Lyz2+/+ (F h1+/+, H- e i in-su icien ) and F h1Fl/
Fl; Lyz2c e/+ (F h1−/−, H- e i in-de icien ) mice. The absence o H- e i in in F h1−/− BMDM was con i med by
wes e n-blo (Fig.1a) and also by gene exp ession (Table1). Al hough i ended o be inc eased, he le el o
L- e i in was no signi ican ly di e en be ween F h1−/− and F h1+/+ mac ophages, ei he a he p o ein (Fig.1a)
o a he RNA le el (Table1). Th oughou he cul u e, cells wi h ei he geno ype we e indis inguishable in e ms
o mo phology, as well as in e ms o iabili y (measu ed by a esazu in educ ion assay) and in he acquisi ion o
mac ophage di e en ia ion ma ke s (e alua ed by low cy ome y) (Fig.1b–d). These da a also showed ha o
bo h geno ypes, a he 7 h and 10 h day o cul u e, nea ly 90% o he cells had acqui ed he mac ophage pheno ypic
ma ke s (Fig.1d). Thus, subsequen expe imen s we e pe o med 7 days a e he beginning o he cul u e.
In addi ion, we asked whe he he absence o H- e i in exp ession would impac he basal i on and in lam-
ma o y s a us o he mac ophages. Fo ha , we measu ed he exp ession o genes in ol ed in he cellula handling
o i on: ans e in ecep o (T c, i on up ake), e opo in (Slc40a1, i on expo ), heme oxygenase-1 (Hmox1,
heme deg ada ion), inducible ni ic oxide syn hase (Nos2a), a ginase (A g1) and umo nec osis ac o (Tn )
(mac ophage ac i a ion), and we did no ind any signi ican di e ences be ween he wo geno ypes (Table1).
H- e i in modula es mac ophage esponse o immune ac i a ion. In o de o in es iga e he
po en ial ole o H- e i in in mac ophage ac i a ion, in e e on gamma and lipopolysaccha ide (IFNG + LPS)
we e added o he BMDM on he 7 h day o cul u e, as desc ibed in Ma e ials and Me hods. No di e ences in cell
iabili y we e obse ed du ing he ea men wi h IFNG + LPS in cells o ei he geno ype (da a no shown). The
e ec s o ea men on mac ophage gene exp ession we e e alua ed 12, 24 and 72 hou s la e (Fig.2 and TableS1).
In wild- ype cells, F h1 gene exp ession inc eased upon ea men wi h IFNG + LPS, especially a 24 h (Fig.2a), in
acco dance wi h p e ious epo s14. F l exp ession sligh ly inc eases wi h IFNG + LPS ea men a 24 h, howe e ,
no di e ences we e obse ed be ween he wo geno ypes (Fig.2b). In e es ingly, he le els o Slc40a1, which codes
o he i on expo e e opo in, inc eased du ing he i s 24 h o ea men and dec eased a 72 h (Fig.2c). O
no e, a 24 h pos - ea men , he exp ession o Slc40a1 was signi ican ly highe in F h1−/− mac ophages han in
wild- ype cells. The exp ession o Hmox1 also inc eased wi h ea men , wi h F h1−/− cells ha ing a signi ican ly
highe exp ession han wild- ype cells a 24 h pos - ea men (Fig.2d). The exp ession o T c in F h1+/+ cells
inc eased wi h ime, whe eas in F h1−/− cells i s exp ession was signi ican ly lowe a all he e alua ed ime-poin s
(Fig.2e). The gene showing he highes induc ion wi h ea men was Nos2a, a hallma k o mac ophage ac i a-
ion18 (Fig.2 ). Howe e , he induc ion o Nos2a upon ea men wi h IFNG + LPS was signi ican ly lowe in
F h1−/− mac ophages when compa ed o F h1+/+ cells (Fig.2 ).
Mac ophage ac i a ion was also e alua ed by measu ing he elease o ni i es and umo nec osis ac o al a
(TNFa) o he cul u e supe na an s. In ag eemen wi h he Nos2a gene exp ession da a, signi ican ly lowe le els o
ni i es we e ound in he supe na an s o F h1−/− han in F h1+/+ mac ophage cul u es ea ed wi h IFNG + LPS
(Fig.3a). Howe e , he elease o TNFa was he same, independen ly o he exp ession o F h1 (Fig.3b).
H- e i in suppo s mac ophage su i al in he p esence o exogenous i on. Al hough F h1−/−
mac ophages had no mal basal i on- ela ed gene exp ession (Table1), we hypo hesized ha hese cells could ha e
a hampe ed capaci y o deal wi h exogenously added i on. To assess ha , cells we e ea ed wi h di e en con-
cen a ions o e ic ammonium ci a e (FAC) o hemin. Cellula iabili y was e alua ed by esazu in educ ion 3
days a e ea men . As expec ed, FAC and hemin caused a concen a ion-dependen educ ion o cell iabili y,
bu mo e in e es ingly, F h1−/− mac ophages we e mo e suscep ible o i on oxici y han F h1+/+ mac ophages
(Fig.4a,c and Table2). Mo eo e , FAC was mo e oxic han hemin, o bo h cell ypes (Table2).
In o de o con i m hese esul s, we e alua ed he impac o he same compounds in mac ophage cell dea h
using he memb ane-impe meable dye SYTOX G eenTM. Bo h FAC and hemin caused a dose-dependen inc ease
in cell dea h, which was again mo e ma ked in F h1−/− cells compa ed o F h1+/+ cells (Fig.4b,d). The damage
induced by exogenous i on was also e iden in mo phological al e a ions de ec ed by T ansmission Elec on
Mic oscopy (TEM) (Fig.4e). In cells ea ed wi h FAC, he main al e a ion obse ed was an inc ease in he num-
be o esicles in he cy oplasm. Some cells also had nuclea al e a ions sugges i e o nec osis. Wi h he hemin
ea men , cells we e la ge and he e we e signs o memb ane damage (Fig.4e, a ows). These al e a ions we e
mo e equen and ma ked in F h1−/− cells (Fig.4e).
The da a p esen ed he e indica e ha F h1−/− cells a e less capable o handling exogenous i on, being mo e
sensi i e o i on-induced cell dea h.
H- e i in-de icien mac ophages e ain less exogenously added i on. In o de o e alua e how
H- e i in de iciency impac ed i on in e naliza ion and e en ion by mac ophages, we used TEM equipped wi h
3
Scien i ic RepoR S | (2020) 10:3061 | h ps://doi.o g/10.1038/s41598-020-59898-0
www.na u e.com/scien i ic epo s
www.na u e.com/scien i ic epo s/
an Ene gy-Dispe si e X- ay Spec oscopy (EDS) sys em o ace in acellula i on. Cells we e analysed 30 min-
u es, 1 and 12 hou s a e i on ea men s. In all he ime-poin s, a clea endency was obse ed o lowe in a-
cellula i on le els in F h1−/− as compa ed o F h1+/+ cells (Fig.5a). This di e ence was s a is ically signi ican
a all ime-poin s wi h hemin, bu only a 30 minu es pos - ea men wi h FAC (Fig.5a). We complemen ed his
assessmen by measu ing in acellula i on le els 24 hou s a e i on addi ion, using a omic abso p ion spec om-
e y. We ound ha F h1−/− BMDM had a signi ican ly lowe amoun o in acellula i on han F h1+/+ cells a e
Figu e 1. H- e i in is no necessa y o in i o di e en ia ion o bone ma ow-de i ed mac ophages. (a)
Quan i ica ion o FTH1 and FTL by Wes e n blo in p o ein ex ac s om F h1+/+ (black) and F h1−/− (g ey)
BMDM, a 10 days o cul u e. Quan i ica ion was made by densi ome y analysis wi h ImageLabTM so wa e
o FTH1 o FTL bands no malized o β-ac in (ACTB) in samples om 2–4 independen cul u es and a e
ep esen ed as pe cen age ela i e o F h1+/+ BMDM ( es *p < 0.05). (b) Cell iabili y o F h1+/+ (black) and
F h1−/− (g ey) BMDM was measu ed a 4, 7 and 10 days o cul u e, by esazu in educ ion. The esul s ep esen
he mean + SD o a leas h ee independen cul u es. The g aphs depic he pe cen age o F h1−/− iable cells
ela i e o F h1+/+ cul u ed in pa allel. (c) Ligh mic oscopy images o BMDM a 7 (le panels) and 10 ( igh
panels) days o cul u e. The cells we e isualized and imaged in a Leica DMI6000 Time-lapse mic oscope.
Images a e ep esen a i e o he cul u es ob ained om ou animals pe geno ype. Scale ba : 50 μm. (d) Flow
cy ome y analysis o BMDM a 7 and 10 days o di e en ia ion. Cells we e s ained o he myeloid ma ke s
F4/80 and CD11b. CD11b+ F4/80+ cells we e conside ed comple ely di e en ia ed mac ophages. Le panels:
low cy ome y plo s ga ed o CD11b+ F4/80+ cells. Righ panels: his og am plo s o F h1+/+ (solid black line)
and F h1−/− (do ed g ey line) o F4/80, CD11b, and cell size (FSC).
4
Scien i ic RepoR S | (2020) 10:3061 | h ps://doi.o g/10.1038/s41598-020-59898-0
www.na u e.com/scien i ic epo s
www.na u e.com/scien i ic epo s/
ea men wi h FAC (Fig.5b). O e all, hese esul s indica e ha he lack o H- e i in hampe s mac ophages’
capaci y o e ain i on when his is exogenously added.
In o de o unde s and he mechanisms unde lying hese di e ences in cellula i on dynamics, we e alua ed
he al e a ions in gene exp ession occu ing upon i on ea men s (Tables3 and S2). In gene al, F h1−/− mac-
ophages exhibi ed simila ends as F h1+/+ mac ophages in he esponse o he i on ea men s, wi h up egula-
ion o genes in ol ed in i on s o age (F l) and i on expo (Slc40a1) as well as wi h a dec ease in he exp ession o
he i on in e naliza ion- ela ed gene T c in esponse o heme (Table3). Howe e , a endency was obse ed o an
inc eased exp ession o Slc40a1 in F h1−/− in compa ison wi h wild- ype cells. This endency was also obse ed
a he p o ein le el, al hough i only eached s a is ical signi icance in he case o he hemin ea men (Fig.6b).
We did no obse e s a is ically signi ican di e ences be ween F h1+/+ and F h1−/− mac ophages conce ning he
p o ein exp ession le els o FTL (Fig.6d) o HMOX1 (Fig.6c). O e all, hese da a indica e ha he absence o
H- e i in in mac ophages leads o an inc ease in i on expo h ough e opo in.
FTH1 p o ec s mac ophages agains i on-induced oxida i e s ess. Knowing ha F h1-de icien
mac ophages a e mo e p one o i on-induced dea h despi e lowe in acellula i on le els, we in es iga ed
whe he hei inc eased cell dea h was caused by oxida i e s ess. Fo ha , mac ophages we e incuba ed wi h i on
oge he wi h he an ioxidan N-ace yl-cys eine (NAC) and cell dea h was quan i ied by inco po a ion o SYTOX
G een. The oxic e ec o exogenous i on was pa ially p e en ed by he p esence o NAC (Fig.7a,b), indica -
ing ha i is, a leas pa ially, media ed by eac i e oxygen species. To e alua e whe he F h1−/− mac ophages
had an in insically highe suscep ibili y o oxida i e s ess, we ea ed hem wi h ei he hyd ogen pe oxide o
e -bu yl-hyd ogen pe oxide ( BHP). Cell dea h inc eased wi h bo h ea men s in a dose-dependen manne ,
bu no di e ences we e obse ed be ween F h1+/+ and F h1−/− mac ophages (Fig.7c,d). Al oge he , hese esul s
(Fig.7a–d) indica e ha bo h geno ypes a e able o wi hs and simila le els o oxida i e s ess. Howe e , when
aced wi h equal concen a ions o i on, he oxida i e damage in F h1−/− cells appea s o be highe han in F h1+/+
cells. To con i m ha , we measu ed he le els o ROS, lipid pe oxida ion and p o ein ca bonyla ion. FAC induced
a clea ly highe le el o ROS o ma ion on F h1−/− cells compa ed o F h1+/+ cells, especially 1 and 3 hou s
a e ea men (Fig.7e). F h1−/− mac ophages also had signi ican ly highe le els o lipid pe oxida ion a 6 and
12 hou s a e ea men wi h FAC (Fig.7 ). In he same line, p o ein ca bonyla ion was highe in i on- ea ed
mac ophages o he F h1−/− han F h1+/+ geno ype (Fig.7g,h). In e es ingly, basal le els o p o ein ca bonyla ion
ended o be highe in non- ea ed F h1 de icien cells, al hough he di e ence ela i e o wild- ype was no s a-
is ically signi ican .
In o de o unde s and whe he F h1−/− mac ophages had mo e oxida i e s ess due o some de ec in he
induc ion o cell-p o ec i e mechanisms, we measu ed he exp ession o se e al genes in ol ed in hese mech-
anisms. F h1−/− mac ophages up egula ed hei exp ession o Hmox1, supe oxide dismu ase (Sod2), ca alase
(Ca ), hio edoxin educ ase (T x ) and glu ama e-cys eine ligase ca aly ic subuni (Gclc) simila ly o F h1+/+
cells (Table3).
O e all, he da a ob ained in his wo k, clea ly indica es ha F h1 has an essen ial, non- edundan ole in cell
p o ec ion agains i on-induced oxici y.
Discussion
H- e i in is c ucial o mouse de elopmen and su i al, wi h he o al knockou being emb yonically le hal9. In
his wo k, we show ha H- e i in is no necessa y o in i o di e en ia ion o mu ine BMDM no has a signi -
ican impac in hese mac ophages’ basal s a e, bu i in luences mac ophage esponse o immune ac i a ion o
i on ea men s. In pa icula , H- e i in-de icien BMDM p oduce less ni ic oxide in esponse o IFNG + LPS
ea men and a e mo e p one o oxida i e s ess and cell dea h induced by exogenously added i on.
H- e i in-de icien BMDM we e indis inguishable om wild- ype BMDM ega ding he kine ics o di e en-
ia ion, mo phology, iabili y, and he exp ession o se e al i on- and ac i a ion- ela ed genes. In pa icula , no
signi ican compensa o y inc ease in L- e i in exp ession was ound in H- e i in-de icien mac ophages. This
is in con as wi h he esul s ob ained by Bolise y e al. using he same ype o cells19. The explana ion could be
ela ed o di e ences in cul u e condi ions. O no e, in p e ious s udies, i was obse ed ha ansien silencing
o F h1 exp ession esul ed in up egula ion o F l20,21, bu s able H- e i in-de icien HeLa cell clones had no mal
le els o L- e i in21. Toge he wi h he esul s p esen ed he e, his indica es ha when cells unde go a long- e m
Gene o in e es Basal Le ela
F h1 0.02 ± 0.02*
F l 0.69 ± 0.35
T c 1.39 ± 1.11
Slc40a1 1.16 ± 1.08
Hmox1 0.60 ± 0.33
Nos2a 0.83 ± 1.19
A g1 0.87 ± 1.19
Tn 0.63 ± 0.34
Table 1. Basal gene exp ession in F h1−/− BMDM as compa ed o F h1+/+ cells. aThe alues we e calcula ed as
he old change ela i e o F h1+/+ cells, and ep esen he mean ± SD o ou o ele en independen cul u es pe
gene. S a is ical analysis was pe o med using mul iple - es *p < 0.05.
5
Scien i ic RepoR S | (2020) 10:3061 | h ps://doi.o g/10.1038/s41598-020-59898-0
www.na u e.com/scien i ic epo s
www.na u e.com/scien i ic epo s/
adap ion o he lack o H- e i in he exp ession o L- e i in is no inc eased. This adap a ion may include a
dec ease in in acellula i on accumula ion, as indica ed by ou esul s.
H- e i in is known o be in ol ed in mac ophage ac i a ion. When we ea ed BMDM wi h IFNG and LPS,
he exp ession o F h1 doubled, in ag eemen wi h p e ious esul s ob ained wi h se e al TLR agonis s14,22. An
inc ease in exp ession o F h1 upon IFNG + LPS ea men is appa en in F h1−/− cells (Supplemen a y Table2),
which is p obably ela ed o a mino leakage o F h1 exp ession due o C e ailu e, bu his emained a esidual
le els compa ed o wild- ype cells (Fig.2a). In con as , F l exp ession was no signi ican ly al e ed by IFNG + LPS
ea men , i espec i e o F h1 exp ession. In ag eemen wi h p e ious epo s22,23, he e opo in-coding gene
Slc40a1 was down- egula ed by IFNG + LPS ea men in wild- ype mac ophages. Howe e , in F h1−/− cells, a
signi ican up- egula ion o his gene was obse ed a 24 hou s pos - ea men , indica ing ha he exp ession o
his impo an i on-expo gene is egula ed by cellula H- e i in le els. In ag eemen o wha has been desc ibed
Figu e 2. Gene exp ession is al e ed in F h1+/+ and F h1−/− BMDM upon ea ed wi h IFNG plus LPS. F h1+/+
(black) and F h1−/− (g ey) BMDM we e ea ed wi h IFNG plus LPS, o 12, 24 o 72 hou s. Gene exp ession
o he genes o in e es (a) F h1, (b) F l, (c) Slc40a1, (d) Hmox1, (e) T c and ( ) Nos2a we e no malized o he
le el o Hp 1. Da a a e shown as old change (mean ± SD) ela i e o he non- ea ed F h1+/+ cells, calcula ed
wi h he 2−ΔΔCT me hod, o a leas h ee independen expe imen s. S a is ical analysis was ob ained by wo-way
ANOVA, *p < 0.05, **p < 0.01 and ***p < 0.001.

6
Scien i ic RepoR S | (2020) 10:3061 | h ps://doi.o g/10.1038/s41598-020-59898-0
www.na u e.com/scien i ic epo s
www.na u e.com/scien i ic epo s/
p e iously by Ludwiczek e al. in THP-1 monocy es24, wild- ype BMDM had a signi ican ly highe exp ession
o he T c gene (coding o he ans e in ecep o 1, in ol ed in cellula i on up ake) om 24 hou s onwa d
a e IFNG + LPS ea men . In con as , in F h1−/− mac ophages his gene was down- egula ed by ea men a
all he ime-poin s analysed. Toge he , he di e en ial egula ion o bo h e opo in and ans e in ecep o
genes in F h1−/− mac ophages emphasize he inabili y o hese cells o s o e i on. Addi ionally, F h1−/− BMDM
had a signi ican ly highe exp ession o Hmox1 24 hou s pos - ea men , in compa ison o F h1+/+ mac ophages.
Hmox1 gene exp ession is known o inc ease in esponse o ROS gene a ing agen s and o be up- egula ed in an
in lamma o y en i onmen , such as mycobac e ial in ec ions25,26. This inc ease in Hmox1 exp ession indica es
ha , besides an impai men on i on- e aining capaci y, F h1−/− cells exhibi a highe deg ee o oxida i e s ess
in esponse o IFNG + LPS ea men . In iguingly, his inc eased exp ession o Hmox1 occu ed concomi an ly
wi h a dec ease in Nos2a exp ession and ni i e elease (Figs.2 and 3), which is in con as wi h a p e ious s udy
ha sugges ed ha Hmox1 up egula ion upon IFNG + LPS ea men was due o inc eased Nos2a exp ession and
ni i e p oduc ion25. Con e sely, ou da a sugges ha o e all ROS le el, a he han NOS2 ac i i y, unde lies
Hmox1 induc ion. O no e, al hough we did no ho oughly s udied he le el o ROS in F h1−/− e sus F h1+/+
BMDM a e IFNG + LPS ea men , he e was a endency o ha e highe le els o ca bonyla ed p o eins in
F h1−/− cells unde basal condi ions (Fig.7g). The clea dampening o Nos2a gene induc ion in he absence o
H- e i in and in a si ua ion o low in acellula i on le el, was also an in iguing obse a ion pe se and co obo-
a es a p e ious obse a ion by Bolise y e al.19. Se e al p e ious s udies showed ha ea ing mac ophages wi h
excess i on dec eased Nos2a exp ession and ni i e p oduc ion upon ea men wi h IFNG and/o LPS27–29. We
now pos ula e ha hese e ec s we e due o inc eased in acellula oxida i e s ess a he han o a di ec e ec
o i on. Di e en signalling pa hways ha e been implica ed in he ela ionship be ween i on and Nos2a exp es-
sion, including signal ansduce and ac i a o o ansc ip ion 1 (STAT1)27 and hypoxia-inducible ac o 1-alpha
(HIF1A)30. The in acellula i on le els, labile i on pool (LIP) and I on Responsi e P o eins (IRP) ha e also been
implica ed31. The exac mechanisms linking he lack o H- e i in o he modula ion o Nos2a exp ession a e
beyond he scope o his wo k bu will be he subjec o u u e s udies.
H- e i in is an essen ial playe in i on me abolism since i has i on s o age and de oxi ie ac i i ies. To unde -
s and he impo ance o H- e i in in his scena io, we ea ed mac ophages wi h exogenous i on in he o ms o
FAC and hemin. In ag eemen wi h p e ious epo s25,28, bo h o ms o i on caused a dose-dependen educ ion
in mac ophages’ iabili y (and an inc ease in cell dea h). Howe e , he e was a clea ly highe suscep ibili y o
exogenous i on in he absence o F h1. We easoned ha his occu ed because wi hou F h1, cells a e unable o
oxidize Fe2+, and his will accumula e in a labile eac i e pool. P e ious s udies ha e indica ed ha in he absence
o H- e i in cells ha e an inc eased LIP20,32,33. In o de o ha e an insigh on he al e a ions in in acellula i on
dis ibu ion caused by he lack o H- e i in in ou model, we used elec on mic oscopy coupled wi h EDS. To
ou su p ise, F h1−/− BMDM had signi ican ly less in acellula i on han no mal BMDM, making i echnically
impossible o quan i y i on in di e en in acellula compa men s. This dec ease in in acellula i on may be
caused by an inabili y o s o e i on in a no mal e i in nanocage in he absence o FTH134, as well as a endency
o ha e inc eased le els o e opo in, he only known p o ein expo e . In acco dance wi h his inding, s udies
on he issue i on dis ibu ion in F hlox/lox; MxC e mice e ealed ha he lack o FTH1 esul s in dec eased i on
s o age in he li e and spleen17.
Despi e he dec ease in he amoun o in acellula i on, F h1-de icien BMDM had inc eased le els o oxida i e
s ess when exposed o exogenous i on, as e alua ed by he le els o ROS, lipid pe oxida ion and p o ein ca bony-
la ion. Ou da a indica e ha i on-induced oxida i e s ess is he main cause o inc eased cell dea h in he F h1−/−
mac ophages upon i on ea men since in he p esence o NAC, F h1−/− cells we e as p o ec ed om cell dea h as
Figu e 3. F h1−/− mac ophages ha e dec eased p oduc ion o ni ic oxide bu no mal elease o TNFa as
compa ed o F h1+/+ mac ophages. F h1+/+ (black) and F h1−/− (g ey) BMDM we e ea ed wi h IFNG plus
LPS, o 3 days. (a) Ni i e p oduc ion was quan i ied in he supe na an by he G iess assay. (b) The elease
o TNFa om BMDM was measu ed by ELISA. The esul s shown ep esen he mean + SD o wo o six
independen cul u es pe g oup ( wo-way ANOVA wi h Tukey mul iple compa isons es ; *p < 0.05).
7
Scien i ic RepoR S | (2020) 10:3061 | h ps://doi.o g/10.1038/s41598-020-59898-0
www.na u e.com/scien i ic epo s
www.na u e.com/scien i ic epo s/
F h1 su icien cells. As men ioned be o e, his e ec is p obably due o he ulne abili y o F h1−/− cells o Fe2+ as
hey lack he e oxidase ac i i y con e ed by H- e i in. Impo an ly, he lack o H- e i in did no inc ease he
suscep ibili y o BMDM o di e en pe oxide sou ces, indica ing ha H- e i in is no di ec ly in ol ed in pe oxide
de oxi ica ion, bu i is in ol ed in dec easing pe oxide o ma ion om i on. Fu he mo e, he da a show ha mac-
ophages do no ha e an al e na i e mechanism o deal wi h i on-media ed oxida i e s ess, e en hough hey can
up egula e he ansc ip ion o se e al genes in ol ed in he esponse o oxida i e s ess. O e all, he al e a ions seen
in F h1−/− mac ophages a e consis en wi h he ecen ly desc ibed ype o p og ammed cell dea h, “ e op osis”35–38,
al hough he exac mechanisms o cell dea h we e no ho oughly cha ac e ized he e.
Figu e 4. H- e i in suppo s mac ophage su i al in he p esence o exogenous i on. F h1+/+ (black) and
F h1−/− (g ey) BMDM we e ea ed wi h inc easing concen a ions o (a,b) FAC o (c,d) hemin, o 3 days.
(a,c) Cell iabili y was measu ed by esazu in educ ion. The esul s ep esen he mean ± SD o h ee o se en
independen cul u es pe condi ion and a e exp essed as he pe cen age o iable cells ela i e o non- ea ed cells
o each geno ype. (b,d) Cell dea h was measu ed h ough SYTOX G een inco po a ion in o dead cells. The esul s
ep esen he mean ± SD o h ee o se en independen cul u es pe condi ion ( wo-way ANOVA wi h Tukey
mul iple compa isons es ; *p < 0.05). (e) T ansmission elec on mic oscopy images o F h1+/+ (uppe panels)
and F h1−/− (lowe panels) BMDM ea ed wi h 10 μM o FAC o 100 μM o hemin o 12 h. Scale ba : 1 μm. The
images a e ep esen a i e o h ee independen cul u es pe condi ion. A owheads: memb ane damage.
8
Scien i ic RepoR S | (2020) 10:3061 | h ps://doi.o g/10.1038/s41598-020-59898-0
www.na u e.com/scien i ic epo s
www.na u e.com/scien i ic epo s/
In conclusion, his wo k shows ha al hough H- e i in is no essen ial o he in i o di e en ia ion o bone
ma ow-de i ed mac ophages, i modula es mac ophage ac i a ion by immune and mic obial s imuli and, mo e
impo an ly, i has a non- edundan ole in cell p o ec ion agains i on oxici y.
Compound
IC50 ± SD (µM)a
F h1+/+F h1−/−
FAC 26.6 ± 1.2 9.7 ± 1.1
Hemin 150.9 ± 1.1 49.6 ± 1.0
Table 2. Rela i e oxici y o FAC and hemin o F h1+/+ and F h1−/− BMDM. aEach alue ep esen s he
a e age ± SD o h ee o se en independen cul u es pe condi ion. IC50 is he concen a ion ha inhibi s 50% o
mac ophage iabili y a e 3 days o incuba ion wi h each compound.
Figu e 5. H- e i in-de icien mac ophages e ain less exogenously added i on. I on was quan i ied in F h1+/+
(black) and F h1−/− (g ey) BMDM ea ed wi h 10 μM o FAC o 100 μM o hemin. (a) I on quan i ica ion by
TEM-EDS a 30 minu es, 1 o 12 hou s, pos - ea men s. The esul s ep esen he mean + SD o 6 o 24 cells
and a e exp essed as he pe cen age o i on mass inside a single cell ( wo-way ANOVA wi h Tukey mul iple
compa isons es ; *p < 0.05). (b) I on quan i ica ion by a omic abso p ion a 24 h pos - ea men s. The esul s
ep esen he mean + SD o 4 independen cul u es pe condi ion and a e exp essed as he a io be ween he
concen a ion o i on and he concen a ion o phospho us in each sample ( wo-way ANOVA wi h Tukey
mul iple compa isons es ; *p < 0.05).
Genes o in e es BMDM geno ype FAC Hemin
F h1 F h1+/+1.65 ± 0.83 2.94 ± 1.26
F h1−/−0.11 ± 0.05*0.10 ± 0.03*
F l F h1+/+3.38 ± 2.60 3.57 ± 0.16
F h1−/−4.61 ± 2.39 5.10 ± 0.27
Slc40a1 F h1+/+3.92 ± 2.30 18.19 ± 0.27
F h1−/−11.53 ± 0.25 18.65 ± 14.23
T c F h1+/+1.39 ± 0.73 0.31 ± 0.14
F h1−/−1.0. ± 0.60 0.30 ± 0.32
Hmox1 F h1+/+6.09 ± 6.70 20.19 ± 7.54
F h1−/−10.24 ± 9.00 17.07 ± 9.34
Sod2 F h1+/+4.05 ± 2.45 4.65 ± 1.47
F h1−/−4.40 ± 2.08 5.06 ± 2.75
Ca F h1+/+7.36 ± 5.31 11.57 ± 1.87
F h1−/−14.22 ± 0.42 17.85 ± 3.69
T x F h1+/+6.34 ± 5.58 16.33 ± 8.17
F h1−/−6.16 ± 3.24 11.50 ± 1.98
Gclc F h1+/+1.68 ± 0.71 12.11 ± 7.23
F h1−/−2.48 ± 0.18 3.64 ± 1.66
Table 3. Al e a ions in gene exp ession ain F h1+/+ and F h1−/− BMDM upon FAC o hemin ea men s, a
12 h. aThe alues a e calcula ed as he old change ela i e o he F h1+/+ non- ea ed cells, and ep esen he
mean ± SD o h ee independen cul u es pe condi ion; wo-way ANOVA *p < 0.05.
9
Scien i ic RepoR S | (2020) 10:3061 | h ps://doi.o g/10.1038/s41598-020-59898-0
www.na u e.com/scien i ic epo s
www.na u e.com/scien i ic epo s/
Ma e ials and Me hods
Chemicals. All chemicals we e ob ained om Sigma-Ald ich (S . Louis, MO, USA), unless o he wise
speci ied.
Mac ophage cul u es. Bone ma ow-de i ed mac ophages (BMDM) we e ob ained as desc ibed p e i-
ously39 om he bone ma ow o F h1+/+ o F h1−/− mice.
Mice e e ed o as F h1−/− mice a e condi ional F h1 de icien (F h1Fl/Fl; Lyz2c e/+) mice, ob ained by c oss-
ing F h1Fl/Fl mice17 wi h Lyz2c e/+ mice. An ini ial b eeding pai was kindly p o ided by P o . Lukas Kuhn (Swiss
Ins i u e o Expe imen al Cance Resea ch, Lausanne, Swi ze land). In hese mice, C e ecombinase dele es he
F h1 gene in cells exp essing Lyz2 (cells o he myeloid lineage). F h1Fl/Fl; Lyz2+/+ C e-nega i e li e ma e mice
we e used as con ols and will be e e ed o as F h1+/+ mice.
All expe imen al animal p ocedu es desc ibed in his wo k we e app o ed by he Local Animal E hics
Commi ee o IBMC/i3S and licensed by he Po uguese Au ho i y “Gene al Di ec o y o Ag icul u e and
Ve e ina y” (DGAV), on July 6 h, 2016 wi h e e ence 0421/000/000/2016. All animals we e handled in s ic
acco dance wi h good animal p ac ice as de ined by na ional au ho i ies (DGAV, Dec e o-Lei 113/2013, augus
7 h) and Eu opean Di ec i e 2010/63/EU.
Cell cul u es we e isualized and imaged e e y day du ing he expe imen in an in e ed op ical mic oscope
(Leica DMI6000 Timelapse).
Fo mac ophage ac i a ion expe imen s, ecombinan mouse IFNG (Gibco, MD, USA) was added a 16 ng/ml,
a he 7 h day o cul u e. Fo kine ic s udies, IFNG was added wo addi ional imes he ea e wi h 24 h in e als.
Lipopolysaccha ide (LPS) was gi en once a 10 ng/ml.
Figu e 6. H- e i in-de icien mac ophages exp ess highe e opo in p o ein le els upon hemin ea men .
Quan i ica ion o SLC40A1, HMOX1 and FTL by Wes e n blo in p o ein ex ac s om F h1+/+ (black) and
F h1−/− (g ey) BMDM, 24 hou s a e ea men wi h 10 µM o FAC o 100 µM o hemin. (a) Rep esen a i e
image o he quan i ica ion o each p o ein by wes e n blo . (b–d) Quan i ica ion o SLC40A1 (b), HMOX1
(c) and FTL (d), made by densi ome y analysis wi h ImageLabTM so wa e. The bands we e no malized o
ACTB, and he esul s a e p esen ed as pe cen age ela i e o he non- ea ed F h1+/+ cells, a e age ± s anda d
de ia ion om 3 independen expe imen s. S a is ical analysis was pe o med using wo-way ANOVA wi h
Tukey’s mul iple compa isons es . ***p < 0.001 when compa ed wi h F h1+/+, #< 0.05 when compa ed wi h
non- ea ed F h1−/−.