A icle
Type I In e e on Signaling Dis up s he Hepa ic U ea
Cycle and Al e s Sys emic Me abolism o Supp ess T
Cell Func ion
G aphical Abs ac
Highligh s
dCh onic i al in ec ion causes long- e m ansc ip ome and
p o eome changes in he li e
dHepa ocy e-in insic ype I in e e on (IFN-I) signaling
egula es hepa ic me abolism
dIFN-I signaling ep og ams he u ea cycle in hepa ocy es and
al e s se um me aboli es
dSe um le els o a ginine and o ni hine modula e T cell
esponses and pa hology
Au ho s
Alexande Le che ,
Anannya Bha acha ya,
Alexand a M. Popa, ...,
Michael T aune , K is aps Kla ins,
And eas Be g hale
Co espondence
abe g hale @cemm.oeaw.ac.a
In B ie
The li e is a cen al egula o o
me abolism, bu how in ec ion-induced
changes in li e me abolism a ec
immune esponses and pa hology
emains enigma ic. Le che e al.
demons a e ha hepa ocy e-in insic
ype I in e e on (IFN-I) ep esses
me abolism du ing ch onic i al in ec ion.
Speci ically, IFN-I dis up s he u ea cycle
in hepa ocy es, al e ing se um a ginine
and o ni hine le els, which dampen
an i i al T cells esponses and li e
pa hology.
Le che e al., 2019, Immuni y 51, 1074–1087
Decembe 17, 2019 ª2019 The Au ho (s). Published by Else ie Inc.
h ps://doi.o g/10.1016/j.immuni.2019.10.014
Immuni y
A icle
Type I In e e on Signaling Dis up s he Hepa ic
U ea Cycle and Al e s Sys emic Me abolism
o Supp ess T Cell Func ion
Alexande Le che ,
1,14
Anannya Bha acha ya,
1,13,14
Alexand a M. Popa,
1
Michael Calde a,
1
Mo i z F. Schlapansky,
1
Ha oon Baazim,
1
Benedik Age e ,
1
Be ina G€
u l,
1
Lindsay Kosack,
1
Pe e Ma
´jek,
1
Julia S. B unne ,
2,3
Dijana Vi ko,
1,4
The esa Pin e ,
1,5
Jakob-Wendelin Genge ,
1
Anna O lo a,
6
Na alia Piko ,
7
Daniela Reil,
1
Ma ia Ozs a
´ -Kozma,
1,8
Ul ich Kalinke,
9
Bu kha d Ludewig,
7
Richa d Mo iggl,
6,10
Kei yn L. Benne ,
1
Jo
¨ g Menche,
1
Paul N. Cheng,
11
Ge no Schabbaue ,
2,3
Michael T aune ,
12
K is aps Kla ins,
1
and And eas Be g hale
1,15,
*
1
CeMM Resea ch Cen e o Molecula Medicine o he Aus ian Academy o Sciences, Laza e gasse 14 AKH BT25.3, 1090 Vienna, Aus ia
2
Depa men o Th ombosis Resea ch and Vascula Biology, Medical Uni e si y o Vienna, 1090 Vienna, Aus ia
3
Ch is ian Dopple Labo a o y o A ginine Me abolism in Rheuma oid A h i is and Mul iple Scle osis, 1090 Vienna, Aus ia
4
Depa men o U ology, Bos on Child en’s Hospi al, Ha a d Medical School, Bos on, MA 02115, USA
5
Resea ch Ins i u e o Molecula Pa hology (IMP), Vienna BioCen e (VBC), 1030 Vienna, Aus ia
6
Ins i u e o Animal B eeding and Gene ics, Uni e si y o Ve e ina y Medicine Vienna, 1210 Vienna, Aus ia
7
Ins i u e o Immunobiology, Kan onsspi al S . Gallen, 9007 S . Gallen, Swi ze land
8
Depa men o Labo a o y Medicine, Medical Uni e si y o Vienna, 1090 Vienna, Aus ia
9
Ins i u e o Expe imen al In ec ion Resea ch, TWINCORE, Cen e o Expe imen al and Clinical In ec ion Resea ch, a join en u e be ween
he Helmhol z Cen e o In ec ion Resea ch, B aunschweig, and he Hanno e Medical School, 30625 Hanno e , Ge many
10
Medical Uni e si y o Vienna, 1090 Vienna, Aus ia
11
Bio-Cance T ea men In e na ional Limi ed, Hong Kong, China
12
Di ision o Gas oen e ology & Hepa ology, Depa men o In e nal Medicine III, Medical Uni e si y o Vienna, 1090 Vienna, Aus ia
13
P esen add ess: Me ck Resea ch Labo a o ies, 33 A enue Louis Pas eu , Bos on, MA 02115, USA
14
These au ho s con ibu ed equally
15
Lead Con ac
*Co espondence: abe g hale @cemm.oeaw.ac.a
h ps://doi.o g/10.1016/j.immuni.2019.10.014
SUMMARY
In ec ions induce complex hos esponses linked
o an i i al de ense, in lamma ion, and issue dam-
age and epai . We hypo hesized ha he li e , as a
cen al me abolic hub, may o ches a e sys emic
me abolic changes du ing in ec ion. We in ec ed
mice wi h ch onic lymphocy ic cho iomeningi is i-
us (LCMV), pe o med RNA sequencing and p o e-
omics o li e issue, and in eg a ed hese da a
wi h se um me abolomics a di e en in ec ion
phases. Widesp ead ep og amming o li e me a-
bolism occu ed ea ly a e in ec ion, co ela ing
wi h ype I in e e on (IFN-I) esponses. Vi al in ec-
ion induced me abolic al e a ions o he li e ha
depended on he in e e on alpha/be a ecep o
(IFNAR1). Hepa ocy e-in insic IFNAR1 ep essed
he ansc ip ion o me abolic genes, including
O c and Ass1, which encode u ea cycle enzymes.
This led o dec eased a ginine and inc eased
o ni hine concen a ions in he ci cula ion, esul ing
in supp essed i us-speci ic CD8
+
T cell esponses
and amelio a ed li e pa hology. These indings
es ablish IFN-I-induced modula ion o hepa ic
me abolism and he u ea cycle as an endogenous
mechanism o immuno egula ion.
INTRODUCTION
Pa hogens and concu en issue damage elici complex
con ex -dependen in lamma o y esponse p og ams (Ho ami-
sligil, 2017). Ch onic in ec ions ep esen a pa icula challenge
o he hos o ganism, which is exposed o p olonged in lamma-
ion ha may p edispose o a ious co-mo bidi ies, such as
suscep ibili y o seconda y in ec ions and cance (Okin and
Medzhi o , 2012; Rehe mann and Nascimbeni, 2005; Vi gin
e al., 2009). The same in lamma o y pa hways can also con ol
cellula issue homeos asis and me abolism (Ko as and Medzhi-
o , 2015; Medzhi o , 2008; O’Neill and Pea ce, 2016). Di e en
cell ypes and o gans communica e wi h each o he h ough sol-
uble cy okines, he eby de e mining he quali y, magni ude, and
du a ion o bo h local and sys emic immune esponses (Medzhi-
o , 2008). The li e is a cen al me abolic o gan bu also ep e-
sen s an immuno egula o y hub be ween blood-bo ne pa ho-
gens and he immune sys em (Jenne and Kubes, 2013; P o ze
e al., 2012; Racanelli and Rehe mann, 2006). Hepa ocy es a e
he unc ional uni o he li e pa enchyma and he mos abun-
dan cell ype o he li e . As such, hei p incipal ask is he
u no e o me aboli es du ing homeos asis. Ye hey a e also
impo an immune signaling pla o ms ha p oduce and eac
o a ange o cy okines upon in lamma ion (C ispe, 2016; Raca-
nelli and Rehe mann, 2006; Zhou e al., 2016). Hepa ocy es
hemsel es a e pe missible o a a ie y o ch onic i uses,
including hepa i is B i us (HBV) and hepa i is C i us (HCV) in
humans and lymphocy ic cho iomeningi is i us (LCMV) in
1074 Immuni y 51, 1074–1087, Decembe 17, 2019 ª2019 The Au ho (s). Published by Else ie Inc.
This is an open access a icle unde he CC BY license (h p://c ea i ecommons.o g/licenses/by/4.0/).
ABC
D E F
G
H
(legend on nex page)
Immuni y 51, 1074–1087, Decembe 17, 2019 1075
mice (Guido i e al., 1999; Rehe mann and Nascimbeni, 2005).
The ch onic in ec ion model o LCMV ep esen s a well-es ab-
lished and pa hophysiologically ele an expe imen al model
o he s udy o hos -pa hogen in e ac ions and immune e-
sponses ha induce a igo ous CD8 T-cell-dependen hepa i is
(Zehn and Whe y, 2015; Zinke nagel e al., 1986). The in ol ed
immunopa hologic mechanisms seen in he LCMV model a e
simila o hose obse ed in pa ien s ch onically in ec ed wi h
HBV o HCV (Guido i e al., 1999; Rehe mann and Nascimbeni,
2005).
Soluble in lamma o y signals ac mainly h ough cy okine e-
cep o s (Ho amisligil, 2017; P o ze e al., 2012; Racanelli and
Rehe mann, 2006). Type I in e e ons (IFN-Is) a e cen al an i i al
cy okines ha signal h ough he ubiqui ously exp essed IFNAR
ecep o , which is composed o he wo subuni s IFNAR1 and
IFNAR2. This induces he exp ession o a b oad a ay o genes
desc ibed as in e e on-s imula ed genes (ISGs). ISGs exe an i-
i al unc ions by di ec in e e ence wi h i al eplica ion and
immuno egula o y p ope ies (McNab e al., 2015; Schoggins
e al., 2011). Mo e ecen ly, IFN-Is a e also ecognized as modu-
la o s o me abolism, such as cellula lipid me abolism and edox
homeos asis (Bha acha ya e al., 2015; Pan el e al., 2014; Wu
e al., 2016; Yo k e al., 2015). Cy okine-induced egula ion o li e
me abolism is expec ed o esul in al e ed me aboli e u no e
and elease ha impac s dis al o gans ( an den Be ghe, 1991;
No a a e al., 2015). Immune cells and in pa icula T cells c i ically
depend on ce ain me aboli es o e icien ly pe o m hei unc-
ions and a e hus suscep ible o al e ed me aboli e a ailabili y
(Chang e al., 2015; Geige e al., 2016; Johnson e al., 2018;
Pea ce e al., 2009). In line wi h his, a equen immune e asion
mechanism o cance is he deple ion o essen ial amino acids
o glucose in he umo mic oen i onmen (Buck e al., 2017;
Chang e al., 2015; Ma e al., 2017; Mu ay, 2016).
In his s udy, we in es iga ed he ch onic in ec ion model o
LCMV using an unbiased in eg a i e app oach o un eil in lam-
ma ion-d i en endogenous egula ion o li e me abolism and
i s impac on sys emic immune esponses and issue pa hology.
RESULTS
Iden i ica ion o In lamma o y-Me abolic Changes in he
Li e du ing Ch onic In ec ion
We in ec ed C57BL/6J wild- ype mice wi h 2 310
6
ocus o ming
uni s (FFUs) o he ch onic s ain clone 13 o LCMV and quan i-
ied in ec ious i us pa icles and i al RNA in he li e (Figu e 1A).
This con i med he peak o i al p opaga ion a ound day 8 a e
in ec ion wi h a subsequen decline o i al loads. Li e damage
was assessed by he clinical hallma k pa ame e s alanine amino-
ans e ase (ALT) and aspa a e amino ans e ase (AST) ha
peaked on day 8–12 a e in ec ion (Figu e 1B). In an unbiased
app oach o i us-induced changes in he li e , we collec ed li e
issue om in ec ed mice a di e en phases o in ec ion (day
2zinna e phase; day 8 zpeak o disease; day 30 zch onic
phase; day 60 z esol ing phase) and pe o med ansc ip ome
analyses by RNA sequencing (RNA-seq). The ansc ip omic
da a showed high in a- eplica ep oducibili y and indi idual
samples clus e ed acco ding o he ime cou se o in ec ion (Fig-
u e 1C), highligh ing phase-speci ic changes in gene exp ession
and he g adual eco e y o mice by 60 days a e in ec ion. In o-
al, 3,626 ansc ip s we e di e en ially egula ed (Table S1A)
and he mos di e en ially exp essed genes we e ound on
days 2 and 8 a e in ec ion (Figu e S1A). Hie a chical clus e ing
iden i ied h ee b oad ca ego ies o gene exp ession p og ams,
including ansc ip s ha we e ound al eady egula ed on day
2 (clus e s 1–4) o induced on day 8 (clus e s 5–10) as well as
ansc ip s ha we e ep essed du ing in ec ion (clus e s 11–
15; Figu e 1D; Table S2A). To co obo a e ansc ip ional al e -
a ions, we pe o med issue p o eomics and quan i ied 5,586
p o eins in he li e (Table S1B). P o ein changes co ela ed
wi h di e en ially exp essed ansc ip s (Figu es 1E and S1B;
Table S2C). O he iden i ied clus e s, di e en ially exp essed
genes associa ed wi h an i i al IFN-I signaling we e ound o
be en iched on day 2 (clus e s 1–4; Figu e 1F; Table S2B)
al hough leukocy e-associa ed genes we e mainly ound on
day 8 a e in ec ion (clus e s 5–10; Figu e 1F; Table S2B).
Some o hese di e en ial exp ession changes a e likely o e lec
hepa ic immune cell in il a ion (Figu e S1C).
Ou en ichmen analyses indica ed widesp ead me abolic e-
p og amming o li e issue du ing LCMV in ec ion (Figu e S1D),
wi h pa icula ly p onounced e ec s seen o down egula ed
me abolic pa hways ela ing o lipid and amino acid me abolism
on days 2 and 8 a e in ec ion (clus e s 11–15; Figu e 1F; Table
S2B). This is in line wi h ou ecen s udy abou he modula ion o
ci cula ing lipids du ing acu e LCMV in ec ion (Kosack e al.,
2017). Fo an in-dep h in eg a ion o hese me abolic changes,
we pe o med en ichmen analyses o he union o di e en ially
egula ed ansc ip s and p o eins a each s age o i al in ec ion
on Kyo o Encyclopedia o Genes and Genomes (KEGG) me a-
bolic pa hways. We iden i ied global ep og amming o p o-
cesses ela ed o all he majo classes o me aboli es. Modula-
ion o hepa ic pa hways in ol ing lipids and essen ial amino
acids was ini ia ed on day 2 and peaking on day 8 a e in ec ion
Figu e 1. LCMV Cl13 Induces Hepa ic Me abolic Rep og amming, T ansla ing o Changes in Sys emic Me abolism du ing he Cou se o
In ec ion
(A and B) Vi emia and RNemia o LCMV-clone-13-in ec ed li e issue (A; n = 3) and se um alanine ans e ase (ALT) and aspa a e amino ans e ase (AST) le els
upon LCMV clone 13 in ec ion up o 60 days a e in ec ion (B; n = 4–12).
(C) P incipal-componen analyses (PCA) o li e issue ansc ip omes a 0, 2, 8, 30, and 60 days a e in ec ion (n = 3).
(D) Hie a chical clus e ing ( agmen s pe kilobase o ansc ip pe million [FPKM]; k-means; Pea son’s co ela ion) o signi ican ly changed genes a any indi-
ca ed ime poin (n = 3).
(E) Regula ion o de ec ed p o eins in li e issue a he co esponding ime poin s (n = 3).
(F) En iched GO e ms and pa hways (ClueGO) o ansc ip s iden i ied in he g oups o clus e s shown in (D).
(G) En ichmen o he union o signi ican ly egula ed me abolic ansc ip s and p o eins on KEGG me abolic da abase a any ime poin .
(H) Hie a chical clus e ing o signi ican ly egula ed se um me aboli es (n = 4; k-means; Pea son’s co ela ion). dpi, days pos -in ec ion.
Fo (A), (B), and (H), one ou o a leas wo ep esen a i e expe imen s is shown. Fo (C)–(G), ansc ip omic and p o eomic da a a e de i ed om one expe imen .
Symbols ep esen he a i hme ic mean ± SEM; ns, no signi ican ; *p < 0.05; **p < 0.01; ***p < 0.001 (S uden ’s es ). See also Figu e S1.
1076 Immuni y 51, 1074–1087, Decembe 17, 2019
(Figu e 1G; Table S1C). Di e en ially exp essed me abolic p o-
cesses we e also ound in he ch onic and esol ing phases o
in ec ion, al hough he numbe o modula ed ansc ip s and p o-
eins we e lowe (Figu es 1G, S1A, and S1B).
To add ess he e ec s o i us-induced al e a ions on
he me abolic ou pu o he li e , we pe o med a ge ed me abo-
lomicso se um, ocusingonaminoacids,biogenic amines,sphin-
golipids, acylca ni ines, and glyce ophospholipids. We ound 99
o 180 me aboli es o be signi ican ly egula ed a one o mo e
ime poin s (Figu e 1H; Table S3A). Down egula ion o me aboli es
on day 2 (Figu e S1E) was obse ed o glyce ophospholipids and
sphingolipids (Figu e 1H; clus e s 1 and 3) al hough acylca ni ines
(Figu e 1H; clus e 4) we e a he up egula ed on day 8 a e in ec-
ion (Figu e 1H). Sys emic amino acids and biogenic amines ex-
hibi ed a sus ained decline om day 2 o day 8 and g adually
eco e ed in he la e phases o in ec ion (Figu e 1H; clus e 2).
This a ec ed almos exclusi ely essen ial and semi-essen ial p o-
einogenic amino acids (His, Ile, Leu, Me , T p, Val, and A g;
Figu e 1H) and coincided wi h ansc ip ional egula ion o
amino-acid- ela ed me abolic pa hways on day 8 (Figu es 1F–
1H and S1F; Table S1C). Toge he , ou in eg a ed ansc ip omic
and p o eomic analysis linked changes in he li e wi h al e ed
sys emic me aboli e le els du ing ch onic i al in ec ion.
Reduced Food In ake du ing Vi al In ec ion Mildly
A ec s Gene Exp ession in he Li e
Mice in ec ed wi h LCMV s ain clone 13 de elop in ec ion-asso-
cia ed cachexia (Baazim e al., 2019) and display educed ood
in ake (Figu e S2A). We he e o e pe o med a pai - eeding
expe imen o in es iga e he impac o in ec ion-associa ed
changes in ood up ake o he me abolic ep og amming o he
li e 8 days a e in ec ion. Unin ec ed mice ecei ed es ic ed
amoun s o ood equi alen o he amoun s consumed by in-
ec ed mice. Subsequen ly, we collec ed li e issue o in ec ed,
pai - ed, and nai e animals and pe o med ansc ip ome
analyses. These esul s indica ed ha mos o he obse ed
ansc ip ional changes we e linked o i al in ec ion and
independen o ano exic beha io (Figu es 2A, S2B, and S2C).
Unin ec ed pai - ed mice showed only mino di e en ial gene
exp ession changes ha pe ained mainly o lipid me abolism
(Figu es 2B and 2C; Table S4). Toge he , hese expe imen s
demons a ed ha he majo i y o he obse ed i us-induced
me abolic changes in he li e , including amino-acid- ela ed
pa hways, co ela ed wi h he in ec ion s a us o he mice and
we e independen o al e ed ood in ake (Figu es 1H and 2D).
IFN-I Signaling Rep og ams Hepa ic Me abolism
The obse ed changes o hepa ic amino acid me abolism
became no iceable as ea ly as 2 days, which coincides wi h
peak se um le els o IFN-aand IFN-bin mice in ec ed wi h
LCMV (Bha acha ya e al., 2015). Thus, we aimed o dissec a
po en ial ole o ea ly IFN-I signaling in he obse ed changes
in he li e . To measu e he impac o IFN-I signaling on li e pa-
enchyma, we pe o med an ex acellula me abolic lux analysis
o p ima y mu ine hepa ocy es upon s imula ion wi h IFN-b
in i o. IFN-ba ec ed p oxies o mi ochond ial espi a ion
and glycolysis in an IFNAR1-dependen manne (Figu es 3A
and 3B), indica ing ha IFN-I signaling a ec s me abolism o he-
pa ocy es. To decon olu e issue he e ogenei y in i o and o
assess he con ibu ion o IFN-I signaling o me abolic ep og-
amming in he li e , we ook ad an age o a gene ic model o
hepa ocy e-speci ic abla ion o I na 1 (Alb-C e ERT2 I na 1
l/ l
[I na 1
D/D
]) and espec i e li e ma e con ols (Alb-C e ERT2
I na 1
+/+
[I na 1
+/+
]). These mice did no e eal any geno ype-
speci ic di e ences in i emia o se um concen a ion o IFN-a
1.5 days a e in ec ion (Figu es S3A–S3C), sugges ing ha any
po en ial di e ences seen in exp ession p o iling o li e issue
om I na 1
D/D
e sus I na 1
+/+
mice is unlikely o be due o
al e ed i al loads and/o sys emic IFN-I esponses. Mo eo e ,
nai e and in ec ed I na 1
D/D
e sus I na 1
+/+
animals displayed
compa able abundances o immune-cell- ela ed ansc ip s,
sugges ing only mino di e ences o immune cell in il a ion
be ween he geno ypes a his ea ly ime poin a e in ec ion
(Figu e S3D). Nex , we analyzed ansc ip omic changes o li e
issue aken om unin ec ed and in ec ed mice o ei he geno-
ype a he peak o se um IFN-ale els and employed a limma
(2 32 ac o ial) in e ac ion model. This esul ed in a se o 526 he-
pa ocy e-in insic IFNAR1- egula ed genes, which we e ound o
be associa ed wi h bo h classical ISG esponses as well as
me abolic p ocesses (Figu es 3C, 3D, and S3E; Table S5A).
The egula ed genes could be di ided in o wo majo classes—
IFNAR1-s imula ed (clus e 1) and IFNAR1- ep essed (clus e 2)
genes (Figu e 3C; Table S5B). The majo i y o induced genes
we e well-known classical ISGs encoding o an i i al e ec o s
(clus e 1; Figu e 3D; Table S5C; Schoggins e al., 2011). In e -
e on- ep essed genes (IRGs), which a e no ha well cha ac e -
ized (Mos a a i e al., 2016; Schoggins e al., 2011), we e ound
o be s ongly en iched o me abolism-associa ed p ocesses
(clus e 2; Figu e 3D; Table S5C). In addi ion, clus e s 3 and 4
con ained genes whose main ained exp ession depended on
in ac IFNAR1 signaling and we e also associa ed wi h me abolic
p ocesses (Figu e 3D; Table S5C). Hepa ocy e-in insic IFNAR1
signaling mainly egula ed me abolic genes on day 2 a e in ec-
ion. No ably, a bioin o ma ic in e sec ion wi h ou longi udinal
da a ob ained om ch onically in ec ed wild- ype mice sug-
ges ed ha he IFN-I-dependen egula ion o many o hese
genes is main ained beyond hese ea ly ime poin s (Figu e 3E).
Speci ically, i us-induced gene egula ion o amino-acid-
ela ed pa hways (Figu e 1) was d i en by hepa ocy e-in insic
IFNAR1 signaling (Figu e 3E) and co esponded wi h he di e -
en ially egula ed me aboli es obse ed in in ec ed wild- ype
mice (Figu e 1H). To in es iga e whe he hese me abolic
changes depended on hepa ocy e-in insic IFNAR1 signaling,
we pe o med me abolomics and ound ha sys emic se um
le els we e indeed egula ed by local IFNAR1 signaling o hepa-
ocy es (Figu e S3F). In a mo e s ingen analysis using he limma
(2 32 ac o ial) in e ac ion model, we ound 15 se um me abo-
li es o be egula ed by hepa ocy e-in insic IFNAR1 signaling,
including he semi-essen ial amino acid a ginine and i s down-
s eam me aboli e o ni hine (Figu e 3F). In summa y, ou da a
indica e ha hepa ocy e-in insic IFNAR1 signaling ac s as a
ansc ip ional egula o o li e me abolism and esul s in
changes o ci cula ing me aboli es du ing in ec ion.
Hepa ocy e-In insic I na 1 Signaling B eaks he
U ea Cycle
A ginine and o ni hine a e bo h key me aboli es o he u ea
cycle. The in ol ed enzymes a e encoded by A g1,O c,and
Immuni y 51, 1074–1087, Decembe 17, 2019 1077

Ass1, which a e p ima ily exp essed in he li e , as well as Asl,
which is ubiqui ously exp essed (Uhle
´n e al., 2015). Exp ession
analysis o li e issue o unin ec ed and in ec ed I na 1
D/D
and
I na 1
+/+
mice highligh ed ha exp ession o O c and Ass1
was ep essed in a hepa ocy e-in insic IFNAR1-dependen
manne (Figu es 4A, S4A, and S4B). A g1 was induced, whe eas
Asl was ep essed upon in ec ion, bo h independen o hepa o-
cy e-in insic IFNAR1 signaling (Figu e 4A). These in ec ion-
induced ansc ip ional changes o he u ea cycle genes O c,
Ass1,Asl,andA g1 we e main ained on ansc ip omic and
p o eomic le el up o 30 days a e in ec ion, wi h he mos p o-
nounced e ec s seen on days 2 and 8 (Figu es S4C–S4F). O
no e, he egula ion o O c and Ass1 was independen o ood
in ake (Figu e S4G).
To co obo a e he hepa ocy e-speci ic o igin o he obse ed
ep og amming o he hepa ic u ea cycle, we pe o med single-
cell RNA-seq (scRNA-seq) o p ima y mu ine hepa ocy es iso-
la ed om nai e o LCMV-in ec ed wild- ype mice 2 days a e
in ec ion. Bo h samples showed compa able numbe s o eads
(unin ec ed: 1,493.5; in ec ed: 1,429.0) and genes (unin ec ed:
705.5; in ec ed: 761.0) pe cell. We iden i ied 348 di e en ially ex-
p essed genes in hepa ocy es ha co ela ed well wi h changes
obse ed in bulk li e issue (Figu es 1,S4H, and S4I). Acco d-
ingly, genes induced in hepa ocy es we e associa ed wi h an i-
i al inna e immune esponses, whe eas ep essed genes we e
as ly associa ed wi h me abolic p ocesses (Figu es S4J and
S4K). Hepa ocy es clus e ed by he in ec ion s a us o he mice
(Figu e 4B), and we con i med hepa ocy e-in insic ep ession
o Ass1 and O c upon in ec ion, which in e sely co ela ed wi h
exp ession le els o he ISG I i 1 on he single-cell le el (Figu es
4C and 4D). T ansc ip ional ep ession o O c and Ass1 mani-
es ed in educed abundance o OTC and ASS1 p o eins in hepa-
ocy es as obse ed by immunohis ochemis y on days 2 and 8
a e in ec ion (Figu es 4E and 4F). As expec ed, STAT1, a mas e
egula o o IFNAR1 downs eam signaling, was signi ican ly up-
egula ed in hepa ocy es (Figu es 4E and 4F). The modula ion o
AB
CD
Figu e 2. Reduced Food In ake upon LCMV In ec ion Mainly A ec s Lipid-Me abolism-Associa ed Genes in he Li e
(A) PCA o li e ansc ip omes o nai e, 8 days a e LCMV in ec ion, and pai - ed (8 days) mice (n = 3).
(B) Signi ican ly de egula ed ansc ip s in li e issue upon LCMV in ec ion a 8 days a e in ec ion compa ed o nai e and pai - ed and be ween pai - ed and
nai e animals (n = 3).
(C) En iched GO e ms and pa hways (ClueGO) o signi ican ly up- and down egula ed genes o pai - ed e sus nai e mice (n = 3).
(D) Signi ican ly egula ed me abolic genes in he li e s o pai - ed e sus nai e animals supe -imposed on he p e iously iden i ied signi ican ly egula ed
me abolic ansc ip s and p o eins a he indica ed ime poin s a e LCMV in ec ion (based on Figu e 1G).
Fo (A)–(C), ansc ip omic da a a e de i ed om one expe imen . See also Figu e S2.
1078 Immuni y 51, 1074–1087, Decembe 17, 2019
O c and Ass1 on day 2 and day 8 a e in ec ion was con i med by
eal- ime PCR in so ed hepa ocy es (Figu e 4G).
In line wi h his, li e issue o in ec ed mice exhibi ed educed
enzyma ic ac i i y o OTC and ASS1, which we measu ed by he
abundance o he
13
C-labeled eac ion p oduc s a gininosucci-
na e (ASS1) o ci ulline (OTC) a e pulsing wi h
13
C-labeled sub-
s a es. Addi ion o he ASS1-speci ic inhibi o a-me hyl-DL-as-
pa ic acid (MDLA) abolished p oduc ion o a gininosuccina e
and indica ed speci ici y o he assay (Figu e S4L). In line wi h
he IFNAR1-dependen inc ease o sys emic o ni hine le els
upon in ec ion (Figu e 4A), hese da a indica e ha IFNAR1 e-
p esses he deg ada ion o o ni hine ia OTC in hepa ocy es.
AC
D
B
EF
Figu e 3. Hepa ocy e-In insic IFNAR1 Signaling Is a T ansc ip ional Regula o o Li e Me abolism and Shapes Sys emic Me abolism
(A and B) Oxygen consump ion a e (OCR) (A) and ex acellula acidi ica ion a e (ECAR) (B) o wild- ype and I na 1
/
p ima y hepa ocy es ea ed o 4 h wi h
IFN-b(n = 11).
(C) Clus e ing by exp ession p o ile (FPKM; k-means; Pea son’s co ela ion) o ansc ip s signi ican ly egula ed (limma in e ac ion model) by hepa ocy e-
in insic IFNAR1 signaling (n = 3).
(D) En iched GO e ms and pa hways (ClueGO) o ansc ip s iden i ied in he g oups clus e s in (C).
(E) Me abolism-associa ed ansc ip s signi ican ly egula ed by hepa ocy e-in insic IFNAR1 signaling supe -imposed on he p e iously iden i ied signi ican ly
egula ed me abolic ansc ip s and p o eins a he indica ed ime poin s a e LCMV in ec ion (based on Figu e 1G).
(F) Se um me aboli es signi ican ly egula ed (limma in e ac ion model) by hepa ocy e-in insic IFNAR1 signaling in nai e and LCMV-in ec ed animals (n = 3).
Fo (A) and (B), da a o one ou o wo ep esen a i e expe imen s a e shown. Fo (C), (D), and (F), ansc ip omic and me abolomic da a a e de i ed om one
expe imen . Symbols ep esen he a i hme ic mean ± SEM; *p < 0.05; **p < 0.01 (S uden ’s es ). See also Figu e S3.
Immuni y 51, 1074–1087, Decembe 17, 2019 1079
(legend on nex page)
1080 Immuni y 51, 1074–1087, Decembe 17, 2019
The down egula ion o Asl exp ession, oge he wi h dec eased
a gininosuccina e, is expec ed o agg a a e he in ec ion-
induced b eak in he u ea cycle by limi ing a ginine e-syn hesis
(Figu e 4A).
We con i med he a o emen ioned ansc ip ional changes o
heu eacycleandup egula iono I i 1 in he i al in ec ion
model o co ona i us mouse hepa i is i us (MHV) by eal-
ime PCR (Figu e S4M). To es whe he simila esponses in
he li e a e seen in an in lamma o y con ex , we adminis e ed
he syn he ic i al RNA analog and TLR3 agonis polyinosinic-
polycy idylic acid (poly(I:C)) in ape i oneally o wild- ype
mice. Poly(I:C) ea men esul ed in s ong induc ion o I i 1
and co obo a ed he up egula ion o A g1 and down egula ion
o O c and Ass1 in li e issue by eal- ime PCR (Figu e S4N).
T ea men wi h 100 ng ecombinan IFN-ainduced an ISG
signa u e in he li e bu did no yield ep oducible changes o
exp ession o hese genes (da a no shown), sugges ing ha
in e e on signaling is equi ed, bu no su icien o egula e
he u ea cycle.
Nex , we aimed o add ess whe he i al in ec ion leads o
hype ammonemia, a hallma k pa ame e o u ea cycle de i-
ciency (UCD). Indeed, blood ammonia le els we e signi ican ly
inc eased om 2 up o 8 days a e LCMV in ec ion, co ela ing
wi h down egula ion o he ammonia- ixa ing enzyme Cps1
(Figu es 5AandS5A). This p omp ed us o u he cha ac e ize
he obse ed u ea-cycle- ela ed me abolic changes on a sys-
emic le el by in i o acing o
13
C
6
hea y-iso ope-labeled
a ginine (Figu e 5B). Li e issue o in ec ed mice ha ecei ed
a bolus o
13
C
6
a ginine showed inc eased concen a ion o
labeled o ni hine and ci ulline (Figu e 5C). A ginine was no de-
ec ed in he li e as epo ed p e iously (Wu e al., 2009). These
local me abolic changes in he li e issue, oge he wi h an-
sc ip omic and p o eomic changes in hepa ocy es highly co e-
la ed wi h changes in sys emic me abolism. We de ec ed
inc eased
13
C
6
a ginine deg ada ion in he se um (Figu e 5D),
which is consis en wi h he induc ion o A g1 (Figu es 4Aand
S4C). Toge he wi h IFNAR1-dependen ep ession o O c
and Ass1 (Figu es 4A–4D), his esul ed in he sys emic accu-
mula ion o
13
C
5
o ni hine and
13
C
5
ci ulline, espec i ely (Fig-
u e 5D). In acco dance wi h inc eased a ginine deg ada ion and
dec eased o ni hine deg ada ion, we de ec ed ele a ed le els
o
13
C
1
-labeled u ea in li e issue (Figu e S5B), whe eas
13
C
1
-labeled u ea was only mildly ele a ed in he ci cula ion
(Figu e S5C). Toge he wi h he educ ion o
13
C
5
a ginine, he
end p oduc o his pa hway, hese esul s suppo he no ion
ha i al in ec ion ini ia es a b eak o he hepa ic u ea cycle
as ea ly as 2 days a e in ec ion ha pe sis s a leas un il
8 days a e in ec ion and al e s le els o me aboli es in he ci -
cula ion (Figu e S5D).
Consis en ly, we obse ed a signi ican dec ease o he a gi-
nine- o-o ni hine a io in he se um o mice in ec ed wi h LCMV
as a esul o bo h educed a ginine and inc eased o ni hine
le els (Figu es 5E and S5E). Simila changes in se um concen a-
ion o a ginine and o ni hine we e seen upon in ec ion wi h MHV
(Figu es S5F and S5G). Taken al oge he , ou esul s de i ed
om di e en expe imen al models demons a e conse ed
ansc ip ional and me abolic modula ion o he u ea cycle
upon i al in ec ion.
Sys emic A ginine-O ni hine Homeos asis Is a Regula o
o An i i al Adap i e Immuni y
Because ac i a ed T cells a e auxo ophic o a ginine (Geige
e al., 2016; Mu ay, 2016), we hypo hesized ha an al e ed a gi-
nine- o-o ni hine se um a io migh exe immunomodula o y
unc ion. To es his in i o, p ima y nai e mu ine splenic CD8
T cells we e ac i a ed wi h an i-CD3/CD28 an ibodies o
3 days and IFN-gand umo nec osis ac o alpha (TNF-a) p o-
duc ion was assessed by low cy ome y. CD8 T cells cul u ed
in medium wi h 11.5 mM a ginine, a concen a ion compa able
o he obse ed changes in i o, displayed educed cy okine
p oduc ion compa ed o s anda d cell cul u e medium con ain-
ing 1,150 mM a ginine (Figu e 6A). Addi ion o o ni hine, which
is absen in s anda d cell cul u e medium, esul ed in an addi i e
supp essi e e ec on cy okine p oduc ion (Figu e 6A). This indi-
ca ed ha dec eased concen a ions o a ginine impac CD8
T cell esponses and ha hese e ec s a e agg a a ed by a
simul aneous inc ease o o ni hine.
To expe imen ally uncouple he obse ed changes in he u ea
cycle om o he co-occu ing e ec s in i o, we ea ed
wild- ype mice wi h ecombinan pegyla ed human a ginase 1
( ecA g1) (Cheng e al., 2007) o agg a a e he obse ed endog-
enous egula ion o a ginine and o ni hine in he ci cula ion. As
expec ed, ecA g1 con e ed a ginine o o ni hine in he se um
and closely ecapi ula ed he dec eased a ginine- o-o ni hine a-
io seen upon i al in ec ion (Figu es 5E and S6A). Adminis a ion
o ecA g1 did no a ec he abundance o splenic T cells o in-
ec ed and unin ec ed animals on day 8 a e in ec ion (Fig-
u e S6B). Ye , we ecognized an impai ed shi om nai e
(CD62L
+
CD44

) o e ec o (CD62L

CD44
+
) T cells (Figu es
S6C and S6D). Fu he , he numbe s o i us-speci ic CD8
Figu e 4. IFNAR1 Con ibu es o In ec ion-Induced Me abolic Rep og amming o Hepa ocy es
(A) Depic ion o he u ea cycle, exp ession, and concen a ions o he associa ed genes and se um me aboli es in nai e and LCMV-clone-13-in ec ed Alb-C e
ERT2 I na 1
l/ l
(I na 1
D/D
) and I na 1
+/+
mice (n = 3).
(B) Uni o m Mani old App oxima ion and P ojec ion (UMAP) clus e ing o single-cell RNA-seq da a o p ima y hepa ocy es so ed om nai e o in ec ed wild- ype
animals 2 days a e LCMV in ec ion.
(C and D) Exp ession le els and co ela ion wi h I i 1 le els (IFN-I-s imula ed gene) o (C) O c and (D) Ass1 in hepa ocy es isola ed om nai e e sus in ec ed
animals (n = 2; pooled o each condi ion). Each do ep esen s a single hepa ocy e.
(E) Rep esen a i e immunohis ochemical s aining o OTC, ASS1, and STAT1 in li e sec ions o nai e o LCMV-in ec ed (2 and 8 days a e in ec ion) wild-
ype mice.
(F) Quan i ica ion o immunohis ochemical s aining o OTC, ASS1, and STAT1 using His oQues so wa e (n = 4 and 8 pic u es pe mouse we e quan i ied).
(G) Exp ession o O c and Ass1 in so ed p ima y hepa ocy es isola ed om nai e and LCMV-in ec ed (2 and 8 days a e in ec ion) mice (n = 3).
Fo me aboli e da a in (A), one o wo ep esen a i e expe imen s is shown. Fo (B)–(G), single-cell ansc ip omic and his ological da a a e de i ed om one
expe imen . Symbols ep esen he a i hme ic mean ± SEM; *p < 0.05; **p < 0.01; ***p < 0.001 (S uden ’s es ). Log 2 old changes and adjus ed p alues o
single-cell RNA-seq da a we e compu ed as desc ibed in he STAR Me hods. See also Figu e S4.
Immuni y 51, 1074–1087, Decembe 17, 2019 1081
STAR+METHODS
KEY RESOURCES TABLE
REAGENT o RESOURCE SOURCE IDENTIFIER
An ibodies
an i-CD3e BD Biosciences Ca #553238
an i-CD28 BD Biosciences Ca #553295
an i-CD8.2b:Paci ic Blue, clone 53-5.8 Biolegend Ca #140414
an i-CD4:FITC, clone H129.19 Biolegend Ca #100408Ca #130308
an i-CD4:PE, clone GK1.5 Biolegend Ca #100408
an i-CD3e:APC, clone 145-2C11 Biolegend Ca #100312
an i-CD44:BV605, clone IM7 Biolegend Ca #103047
an i-CD62L:AF700, clone MEL-14 Biolegend Ca #104426
an i-CD19:APC-Cy7, clone 6D5 Biolegend Ca #115530
an i-IFNg: PE-Cy7, clone XMG1.2 Biolegend Ca #505826
an i-IL-2: PE, clone JES6-5H4 Biolegend Ca #503808
an i-TNFa: APC, clone MP6-XT22 Biolegend Ca #506308
an i-STAT1 Cell Signaling Technology Ca #39172
an i-ASS1 Abcam Ca #ab170952
an i-OTC San a C uz Ca #sc-5157791
an i-CD16/32, clone: 93 Biolegend Ca #101330
an i-CD16/32, clone: 93 Biolegend Ca #101330
an i-CD16/32, clone: 93 Biolegend Ca #101330
a an i-IFN-aPBL In e e on Sou ce Ca #22100-1
abbi an i-IFN-aPBL In e e on Sou ce Ca #32100-1
an i- abbi HRP an ibody Jackson ImmunoRese ach Ca #711-036-152
Bac e ial and Vi us S ains
MHV s ain A59 Ce an es-Ba agan e al., 2007 NA
LCMV s ain Cl13 Bha acha ya e al., 2015 NA
Chemicals, Pep ides, and Recombinan P o eins
Polyinosine-olycy idylic acid (poly(I:C)) in i ogen Ca # l l-pic
Recombinan mu ine IFNbPBL Assay Science Ca #12405-1
Recombinan mu in IL-2 The mo Fishe Scien i ic Ca #34-8021-82
Libe ase TM Sigma Ca #5401127001
Tamoxi en Sigma Ca #T5658
Sun lowe oil Sigma Ca #S5007
a-Me hyl-DL-aspa ic acid Sigma Ca #M6001
L-Ci ulline Sigma Ca #C7629
L-A ginine Sigma Ca #A5006
L-O ni hine Sigma Ca #O2375
b-me cap oe hanol Sigma Ca #M6250
13
C
6
a ginine Camb idge Iso ope Labo a o ies Ca #CLM-2265-H-PK
13
C
5
o ni hine Camb idge Iso ope Labo a o ies Ca #CLM-4724-PK
13
C
4
aspa ic acid Camb idge Iso ope Labo a o ies Ca #CLM-1801-H-PK
Ca bamoyl phospha e Sigma Ca #C4135
Phospha ase inhibi o cock ail The mo Fishe Scien i ic Ca #78447
Adenosine iphospha e The mo Fishe Scien i ic Ca #R1441
QIAzol lysis eagen QIAGEN Ca #79306
Ro i his o ix Ca l Ro h Ca #P087.4
(Con inued on nex page)
e1 Immuni y 51, 1074–1087.e1–e9, Decembe 17, 2019

Con inued
REAGENT o RESOURCE SOURCE IDENTIFIER
Ta ge e ie al solu ion Dako Ca #S1699
Taqman as uni e sal PCR mas e mix The mo Fishe Scien i ic Ca #4352042
Agilen Seaho se XF base medium Agilen Ca #102353-100
Red blood cell lysis bu e eBioscience Ca #00-430054
RPMI 1640 medium o SILAC The mo Fishe Scien i ic Ca #88365
Dialyzed FCS The mo Fishe Scien i ic Ca #A3382001
Hepa in-Na ium-5000- a iopha m Ra iopha m NA
Xylapan Ve oquinol NA
Na ke an Ve oquinol NA
GP33-PE NIH e ame co e acili y NA
NP396-APC NIH e ame co e acili y NA
P o ein anspo inhibi o cock ail eBioscience Ca #00-4980-03
Cell s imula ion cock ail eBioscience Ca #00-4970-93
Saponin Sigma Ca #47036
LCMV gp33-41 pep ide (KAVYNFATC) Pep ide 2.0 Inc. cus om syn hesis
LCMV np396-404 pep ide (FQPQNGAFI) Pep ide 2.0 Inc. cus om syn hesis
LCMV gp276-286 pep ide (SGVENPGGYCL) Pep ide 2.0 Inc. cus om syn hesis
LCMV gp64-80 pep ide (GPDIYKGVYQFKSVEFD) Pep ide 2.0 Inc. cus om syn hesis
Pegyla ed ecombinan human ARG1 Bio-Cance T ea men In e na ional L d. Ca #PEG-BCT-100
Recombinan mu ine IFNaPBL In e e on Sou ce Ca #12100-1
TMB Solu ion The mo Fishe Scien i ic Ca #002023
C i ical Comme cial Assays
Agilen Seaho se XF Cell Mi o S ess Tes Ki Agilen Ca #10315-100
CD8a+ T Cell Isola ion Ki Mil enyi Bio ec Ca #130-104-075
Absolu e IDQ P180 Ki Bioc a es Li e Sciences AG NA
P ecellys CK14 Lysing Ki VWR Ca #432-3751
Re e Aid Fi s S and cDNA Syn hesis Ki The mo Fishe Scien i ic Ca #K1612
Deposi ed Da a
P o eomic da a This s udy PRIDE: PXD011122
T ansc ip omic da a (supe amily) This s udy GEO: GSE123688
Expe imen al Models: O ganisms/S ains
Mouse:C e-Alb ERT2 Schule e al., 2004 NA
Mouse:I na 1
–/–
Mulle e al., 1994 NA
Mouse:I na 1
l/ l
Kamphuis e al., 2006 NA
Oligonucleo ides
E 1aTaqman p ime s and p obes Bha acha ya e al., 2015 NA
O c Taqman Gene Exp ession Assay The mo Fishe Scien i ic Ca #Mm00493267_m1
Ass1 Taqman Gene Exp ession Assay The mo Fishe Scien i ic Ca #Mm00711256_m1
LCMV NP Taqman p ime s and p obes Bha acha ya e al., 2015 NA
So wa e and Algo i hms
FlowJo 8.7 FlowJo FlowJo
T ace Finde 4.1 The mo Fishe Scien i ic T ace Finde
Cell Range 3.0.2 10X Genomics CellRange
Me IDQ, Ve sion 5-4-8-DB100-Bo on-2607 Bioc a es Li e Sciences AG Me IDQ
His oQues TM so wa e TissueGnos ics GmbH His oQues
Xcalibu 2.1 The mo Fishe Scien i ic Xcalibu
Exp essClus e 1.3 Ha a d Medical School Exp essClus e
ClueGO 2.3.3 Bindea e al., 2009 Cy oscape ClueGO
(Con inued on nex page)
Immuni y 51, 1074–1087.e1–e9, Decembe 17, 2019 e2
LEAD CONTACT AND MATERIALS AVAILABILITY
Inqui es o u he in o ma ion o eques s o esou ces and eagen s should be di ec ed and will be ul illed by he lead con ac
And eas Be g hale ([email p o ec ed]). This s udy did no gene a e new unique eagen s.
EXPERIMENTAL MODEL AND SUBJECT DETAILS
Mice
C57BL/6J mice we e o iginally ob ained om The Jackson Labo a o y, I na 1
l/ l
mice om Ul ich Kalinke (TWINCORE, Cen e o
Expe imen al and Clinical In ec i on Resea ch, Hanno e , Ge many) and C e-Alb ERT2 mice om Pie e Chambon (Ins i u de
Ge
´ne
´ ique e de Biologie Mole
´culai e e Cellulai e, Illki ch, F ance). Mice we e b ed and main ained unde speci ic pa hogen- ee
(SPF) condi ions a he Ins i u e o Molecula Bio echnology o he Aus ian Academy o Sciences, Vienna, Aus ia. Animal expe i-
men s we e conduc ed in indi idual en ila ed cages acco ding o he espec i e animal expe imen licenses (BMWFW-66.009/
0199-WF/V/3 /2015 and BMWFW-66.009/0361-WF/V/3b/2017) app o ed by he ins i u ional e hical commi ees and he ins i u-
ional guidelines a he Depa men o Biomedical Resea ch o he Medical Uni e si y o Vienna. Mice we e 8 o 12 weeks old
and age- and sex-ma ched wi hin expe imen s. Male and emale mice we e used in e changeably be ween expe imen s and no s ik-
ing sex-di e ences we e obse ed. Fo expe imen s using mu ine hepa i is i us (MHV), mice we e main ained unde SPF condi ions
a he Kan onsspi al S . Gallen Medical Resea ch Cen e , S . Gallen Swi ze land. MHV expe imen s we e conduc ed in indi idually
en ila ed cages acco ding o he animal expe imen license (SG/14/18.30861) app o ed by ede al and can onal e hical commi ees.
Vi uses and cell lines
Lymphocy ic cho iomeningi is i us (LCMV) was g own on BHK-21 cells (baby hams e kidney cells de i ed om 5 unsexed new-
bo ns, ATCC CCL-10) and i e was de e mined in a modi ied ocus o ming assay using Ve o cells ( emale g een monkey kidney cells,
ATCC CCL-81) (Bha acha ya e al., 2015). Mice we e in a enously in ec ed wi h 2x10
6
ocus o ming uni s (FFU) o LCMV.
Mu ine hepa i is i us s ain A59 (MHV) was gene a ed using 17CL1 cells (spon aneously ans o med 3T3 cell line es ablished om
unsexed BALB/c mouse emb yos) (S u man and Takemo o, 1972) and i e was de e mined by s anda ds plaque assay on L929 cells
(male mu ine ib oblas s, ATCC CCL-1) (Ce an es-Ba agan e al., 2007). Mice we e in ape i oneally in ec ed wi h 10
3
plaque o m-
ing uni s (PFU) o MHV.
METHOD DETAILS
In ec ions
Mice we e sac i iced a he ime poin s indica ed in he legends. Tissue samples we e snap ozen in liquid ni ogen and s o ed
a 80C un il u he analyses. Fo se um collec ion, blood was collec ed a he espec i e ime poin s, samples we e cen i uged
a 10.000 pm o 5 min a 4C. The se um was ans e ed in o a new ube and s o ed a 80C un il u he analyses.
Pha macological pe u ba ions
Pegyla ed ecombinan human A ginase 1 (PEG-BCT-100, Bio-Cance T ea men In e na ional) was adminis e ed o mice ia
in ape i oneal injec ion a a dosage o 50 mg/kg. Mice we e ea ed wice pe week. T ea men was s a ed one day p io o
LCMV in ec ion. Con ol mice ecei ed he same olume o PBS.
Con inued
REAGENT o RESOURCE SOURCE IDENTIFIER
TopHa 2 2.0.10 John’s Hopkins Uni e si y TopHa
Cu links/Cu di /Cu me ge 2.2.1 Cole T apnell Lab Cu linds/Cu di
ea u eCoun s Wal e and Eliza Hall Insi u e o Medical Resea ch Sub ead Package
Seu a 3.1.0 Sa ija Lab, New Yo k Genome Cen e Sa ija Lab
R C an R-P ojec
limma R package Bioconduc o limma R package
oom limma unc ion oom oom
P o eoWiza d Lib a y 2.1.2708 P o eoWiza d P o eoWiza d
MASCOT 2.3.02 Ma ixScience Ma ixScience
Phenyx 2.5.14 GeneBio GeneBio
Isoba So wa e B ei wiese and Colinge, 2013 NA
P ism 7-8 G aphpad So wa e G aphpad So wa e
Wa e Desk op 2.0 Agilen Wa e Desk op
e3 Immuni y 51, 1074–1087.e1–e9, Decembe 17, 2019
Condi ional I na 1 abla ion on hepa ocy es
Tamoxi en (T5658, Sigma) was dissol ed in s e ile sun lowe oil (S5007, Sigma) con aining 10% e hanol ( / ) and s o ed a 20C o
a maximum o 2 weeks. To induce hepa ocy e speci ic I na 1 abla ion, C e-Alb ERT2 I na 1
l/ l
and C e-Alb ERT2 I na 1
+/+
con ol mice
we e adminis e ed 50 mg/kg amoxi en in ape i oneally o 5 consecu i e days (Me zge e al., 2005). Expe imen s we e s a ed he
day a e he las dose.
poly(I:C) ea men
Mice we e challenged wi h poly(I:C) (in i ogen, # l l-pic) ia in ape i oneal injec ion a a dosage o 4 mg/kg. Con ol mice ecei ed
PBS. Li e issue was ha es ed and analyzed ia eal- ime PCR a he indica ed ime poin s.
Isola ion o p ima y mu ine hepa ocy es
Mice we e anes he ized (Ke amine/Xylazine: 1:3, 0.1 ml/10 g mouse, Ve oquinol) and he li e was cannula ed ia he po al ein and
pe used wi h 20 mL HBSS (GIBCO) con aining 0.5 mM EGTA (Sigma) ollowed by diges ion o he li e wi h 20 mL L15 medium
(GIBCO) con aining 40 mg/L Libe ase (Roche) a a a e o 5 mL/min. The li e was isola ed, placed in a Pe i dish wi h diges ion me-
dium (L15 wi h 40 mg/L Libe ase) and he li e capsule was diligen ly emo ed. Cells we e cen i uged a 50 g o 5 min a 4C, e-
suspended in William’s E medium con aining 10% FCS (PAA) and 1% Penicillin-S ep omycin-Glu amine (The mo Fishe Scien i ic)
and pla ed. Cells we e s imula ed a he ime o pla ing.
Me aboli e acing
Nai e o LCMV-in ec ed C57BL/6J mice we e gi en an in a enous bolus o 500 mg
13
C
6
labeled a ginine (Camb idge Iso ope Lab-
o a o ies, CLM-2265-H-PK) dissol ed in PBS. Mice we e sac i iced 20 min a e wa d and se um and issues o in e es we e ha -
es ed and s o ed a 80C.
Pe mg issue (app oxima ely 100 mg pe sample), 3 mL 80% ( / ) me hanol we e added and issue samples we e homogenized
using a P ecellys 24 issue homogenize (P ecellys CK14 lysing ki , Be in). 50 mL o homogenized issue o se um we e mixed wi h
450 mL o me hanol and 250 mL o wa e and o exed o 10 s. A e wa d, 450 mL chlo o o m we e added pe sample and mixed by
o exing o 10 s, incuba ed on ice o 5 min and o exed again o 10 s be o e cen i uging hem o 10 min a 1000 g. The uppe
aqueous phase was collec ed, d ied (ni ogen e apo a o ) and econs i u ed in 50 mL o me hanol. Samples we e cen i uged o
10 min a 1000 g and supe na an s we e used o LC-MS analysis.
A Vanquish UHPLC sys em (The mo Scien i ic) coupled o an O bi ap Fusion Lumos (The mo Scien i ic) mass spec ome e was
used o he LC-MS analysis. The ch oma og aphic sepa a ion o samples was ca ied ou on an ACQUITY UPLC BEH Amide,
1.7 mm, 2.1x100 mm analy ical column (Wa e s) equipped wi h a VanGua d: BEH C18, 2.1x5mm p e-column (Wa e s). The column
was main ained a a empe a u e o 40C and 2 mL sample we e injec ed pe un. The mobile phase A was 0.15% / o mic acid in
wa e and mobile phase B was 0.15% / o mic acid in 85% / ace oni ile wi h 10 mM ammonium o ma e. The g adien elu ion
wi h a low a e 0.4 mL/min was pe o med wi h a o al analysis ime o 17 min. The mass spec ome e was ope a ed in a posi i e
elec osp ay ioniza ion mode: sp ay ol age 3.5 kV; shea h gas low a e 60 a b; auxilia y gas low a e 20 a b; capilla y empe a u e
285C. Fo analysis, a ull MS scan mode wi h a scan ange m/z 50 o 400, esolu ion 120.000, AGC a ge 2e5 and a maximum in-
jec ion ime 50 ms was applied. Fo de ec ion o u ea he scan ange was adjus ed o m/z 50 o 250 a a esolu ion o 500.000.
Enzyma ic assays
Li e issue lysis and bu e condi ions o enzyma ic assays o OTC and ASS1 we e p e iously desc ibed (Diez-Fe nandez e al.,
2016; Gue ei o e al., 2009). B ie ly, li e issue was lysed in 10 mM HEPES, 0.05% T i on X-100, 0.5 mM DTT, 2 mM EDTA, 1X
Hal P o ease and Phospha ase Inhibi o Cock ail (The mo Fishe Scien i ic, 78447), pH 7.4 using a TissueLyse (QIAGEN) a
30 Hz o 3x 30 s. P o ein concen a ions we e measu ed in he clea ed lysa e using a B ad o d assay. 100 mg o p o ein we e
used o each assay. Fo OTC ac i i y, 200 mL o 50 mM T is-Ace a e, 2 mM EDTA, 1 mM L-O ni hine
13
C
5
(Camb idge Iso ype Lab-
o a o ies, CLM-4724-PK, 1 mM Ca bamoyl Phospha e (Sigma, C4135), 1X Hal P o ease and Phospha ase Inhibi o Cock ail
(The mo Fishe Scien i ic, 78447), pH 8.3 and samples we e incuba ed o 30 min a 37C. Fo ASS1 ac i i y, 200 mL o 20 mM HEPES,
2 mM ATP (The mo Fishe Scien i ic, R1441), 5 mM MgCl
2
, 1 mM L-Aspa a e
13
C
4
(Camb idge Iso ype Labo a o ies, CLM-1801-H-
PK), 1 mM L-Ci ulline (Sigma, C7629), 1X Hal P o ease and Phospha ase Inhibi o Cock ail (The mo Fishe Scien i ic, 78447), pH 7
and samples we e incuba ed o 30 min a 37C. Fo expe imen s using he ASS1-speci ic inhibi o a-Me hyl-DL-aspa ic acid
(MDLA, M6001, Sigma), 100 mg o o al li e p o ein ex ac we e ea ed wi h 10 mM MDLA and he enzyma ic assay was pe o med
as desc ibed abo e (Gue ei o e al., 2009).
Following he incuba ion, 10 mL o sample was immedia ely mixed wi h 90 mL me hanol and subsequen ly p epa ed o mass spec-
ome ic analysis o labeled downs eam me aboli es.
13
C labeled a gininosuccina e o
13
C labeled ci ulline was de ec ed using a Vanquish UHPLC sys em (The mo Scien i ic) coupled
o an O bi ap Fusion Lumos (The mo Scien i ic) mass spec ome e was used o he LC-MS analysis. The ch oma og aphic sepa-
a ion o samples was ca ied ou on an ACQUITY HSS T3, 1.8 mm, 2.1x100 mm analy ical column (Wa e s) equipped wi h a
VanGua d HSS T3, 2.1x5 mm p e-column (Wa e s). The column was main ained a a empe a u e o 40C and 2 mL sample we e in-
jec ed pe un. The mobile phase A was 0.1% o mic acid ( / ) in wa e and mobile phase B was 0.1% o mic acid ( / ) in me hanol.
The g adien elu ion wi h a low a e 0.5 mL/min was pe o med wi h a o al analysis ime o 10 min. The mass spec ome e was
Immuni y 51, 1074–1087.e1–e9, Decembe 17, 2019 e4
ope a ed in a posi i e elec osp ay ioniza ion mode: sp ay ol age 3.5 kV; shea h gas low a e 60 a b; auxilla y gas low a e 20 a b;
capilla y empe a u e 285C. Fo he analysis a ull MS scan mode wi h a scan ange m/z 80 o 400, esolu ion 500.000, AGC a ge
2e5 and a maximum injec ion ime 50 ms was applied.
Gene exp ession analyses o so ed p ima y mu ine hepa ocy es
P ima y mu ine hepa ocy es om nai e o LCMV Cl13 in ec ed emale animals we e isola ed as desc ibed abo e. A e isola ion, cells
de i ed om one li e we e esuspended in 500 mL PBS con aining an i-CD16/32 (1:200, clone: 93, Biolegend) and incuba ed o
10 min a oom empe a u e. Nex , immune cells (CD45.2: PE clone: 104, Biolegend, 1:200) and dead cells (Fixable Viabili y Dye:
eFluo 780, eBioscience, 1:2000) we e s ained o 20 min a 4C. Cells we e washed wi h PBS and 50.000 o 200.000 hepa ocy es
( iable CD45.2
–
, pu i y > 95%) we e so ed in o PBS con aining 0.04% BSA on a Sony SH800 Cell So e using a 130 mm chip and
used o gene exp ession analyses. Fo eal- ime PCR analyses, cells om each mouse (n = 3) we e cen i uged a 1500 pm o 5 min
and esuspended in QIAzol o u he analyses. Fo single cell RNA-seq, cells om 2-3 mice we e pooled in equal amoun s, cen i-
uged, and esuspended in PBS con aining 0.04% BSA o ob ain a concen a ion o 800.000 cells/mL. This cells suspension was
subsequen ly used o d ople based single cell RNA-seq using a 10X Genomics Ch omium Single Cell Con olle and s anda d p o-
ocols. Samples we e sequenced wi h an Illumina HiSeq 3000/4000 ins umen wi h 3 samples mul iplexed pe lane and un on a
75bp pai ed-end low cell.
Immunohis ochemis y
Li e pieces we e ixed in Ro i His o ix (P087.4, Ca l Ro h) o 48h. A e wa d he ixa ion solu ion was exchanged o 70% ( / ) e hanol
and s o ed un il pa a in-embedmen . Fo immunohis ochemical s ainings 2 mm FFPE consecu i e sec ions cu , hea -media ed an-
igen e ie al was pe o med in ci a e bu e a pH 6.0 (S1699; Dako, Agilen , San a Cla a, CA, USA). Sec ions we e s ained wi h
an ibodies speci ic o STAT1 monoclonal abbi (1:200; 9172; Cell Signaling Technology), ASS1 monoclonal abbi (1:400;
ab170952; Abcam), OTC monoclonal abbi (1:500, sc-5157791; San a C uz), using s anda d p o ocols.
Cy okine de e mina ion
IFN-ase um le els we e de e mined ia enzyme-linked immunoso ben assay (ELISA) as desc ibed p e iously (Bha acha ya e al.,
2015). In b ie , mic opla es (675074, G eine Bio-One) we e coa ed wi h a an i-mIFN-acap u e an ibody (22100-1, PBL In e e on
Sou ce, 1:54 dilu ion in PBS). P e-dilu ed se um samples (1:10 o 1:20 in PBS) and mIFN-as anda ds (12100-1, PBL In e e on
Sou ce) we e added and incuba ed in a humidi ied chambe o e nigh a 4C. Fo de ec ion, a p ima y abbi an i-mIFN-a
an ibody (32100-1, PBL In e e on Sou ce, 1:738 dilu ion in PBS), a seconda y an i- abbi HRP an ibody (711-036-152, Jackson
ImmunoResea ch) and TMB solu ion (002023, The mo Fishe Scien i ic) we e used. The colo eac ion was ead ou on a pla e eade
a 450 nm.
Blood chemis y
Mouse se um was p e-dilu ed 1:8 in PBS and le els o alanine amino ans e ase (ALT) and aspa a e amino ans e ase (AST) we e
spec opho ome ically analyzed using a Cobas C311 Analyze (Roche). Fo blood ammonia, mice we e e minally anes he ized
(Ke amine/Xylazine: 1:3, 0.1 ml/10 g mouse, Ve oquinol) and blood was sampled ia hea punc u e. Blood was collec ed in
MiniCollec blood EDTA ubes (G eine Bio-One). Plasma was ob ained by cen i uga ion o whole blood a 4.000 pm o 10 min
a 4C and p e-dilu ed 1:1 in 0.9% (w/ ) NaCl solu ion. Samples we e kep on ice, in he da k and in closed ubes un il analysis o
blood ammonia on a Cobas 8000 analyze (Roche).
RNA isola ion and eal- ime PCR
Tissues we e homogenized using a TissueLyse II (QIAGEN). To al RNA was ex ac ed om homogenized li e issue using QIAzol
lysis eagen as pe he manu ac u e ’s ins uc ions (79306, QIAGEN). Re e se ansc ip ion om RNA o cDNA was ca ied ou using
andom p ime s and he Fi s S and cDNA Syn hesis Ki (K1612, The mo Fishe Scien i ic). Real- ime PCR was pe o med wi h Taq-
man Fas Uni e sal PCR Mas e mix (4352042, The mo Fishe Scien i ic) and Taqman Gene Exp ession Assays (The mo Fishe
Scien i ic) o O c (Mm00493267_m1) and Ass1 (Mm00711256_m1). Exp ession le els o LCMV NP (50-CAAGTATTCACACGG
CATGGA-30,5
0-TGGGAGAGCACCTATAACTGATA-30and 50-[6FAM]TGATCTCTTCAATGCACAGCCTGGGC[BHQ1]-30) and E 1a
(50-GCAAAAACGACCCACCAATG-30,5
0-GGCCTTGGTTCAGGATA-30, and 50-[6FAM]CACCTGAGCAGTGAAGCCAG[TAM]-30)
we e measu ed by co esponding p obe and p ime se s as desc ibed p e iously (Bha acha ya e al., 2015).
Me abolic lux measu emen s
Oxygen consump ion a e (OCR) and ex acellula acidi ica ion a e (ECAR) we e de e mined on a Seaho se XFe96 Analyze (Agilen )
using he Seaho se XF Cell Mi o S ess es ki (Agilen , 103015-100). 25.000 p ima y hepa ocy es we e pla ed pe well espec i ely.
In b ie , cells we e ea ed wi h selec ed s imuli o he indica ed ime poin s. P io o measu emen , media was changed o XF Base
Medium (Agilen 102353-100) con aining glucose (10 mM), sodium py u a e (1 mM) and L-glu amine (2 mM) and cells we e incuba ed
e5 Immuni y 51, 1074–1087.e1–e9, Decembe 17, 2019
o 1h. The assay was un acco ding o he manu ac u e ’s ins uc ions. Oligomycin (2 mM), Ca bonyl cyanide-p- i luo ome hoxyphe-
nylhyd azone (FCCP, 0.25 mM) and Ro enone/An imycin A (500 nM) we e subsequen ly injec ed in o wells a he desi ed ime
poin s. Raw da a we e analyzed using Wa e Desk op So wa e (Agilen , e sion 2.0) and expo ed and g aphed in G aphPad P ism
(G aphPad So wa e, e sion 7.0a).
CD8 T cell isola ion and in i o ac i a ion
Spleens o mice we e ha es ed and dissocia ed h ough a 40 mm cell s aine (Falcon). The pelle was esuspended in 1 mL ed blood
cell lysis bu e (eBioscience, #00-4300-54) and incuba ed a oom empe a u e o 1 min. Subsequen ly he eac ion was s opped by
addi ion o 9 mL PBS. Cells we e coun ed and CD8 T cells isola ed using a magne ic ac i a ed cell so ing nega i e selec ion (MACS)
ki acco ding o he manu ac u e ’s ins uc ions (Mil enyi Bio ec, #130-104-075). 96 well pla es we e coa ed o e nigh a 4C wi h
1mg/mL an i CD3e (BD, #553238) and 2 mg/mL an i CD28 (BD, #553295) in a o al olume o 60 mL PBS pe well. Wells we e washed
wi h PBS and 5x10
4
cells we e pla ed pe well in RPMI 1640 medium o SILAC (The mo Fishe Scien i ic, #88365), supplemen ed wi h
10% dialyzed FCS (The mo Fishe Scien i ic, # A3382001), 1% Penicillin-S ep omycin-Glu amine (The mo Fishe Scien i ic), 50 mM
b-me cap oe hanol (Sigma, #M6250) and 20 U/mL IL-2 (The mo Fishe Scien i ic, #34-8021-82). Medium was supplemen ed wi h
indica ed concen a ions o L-a ginine (Sigma, #A5006) and/o L-o ni hine (Sigma, #O2375) and cul u ed up o 72h be o e p oceeding
wi h in acellula cy okine s aining.
Flow cy ome y
To sample blood, 3-4 d ops o blood we e collec ed in 1 mL MEM medium (GIBCO) supplemen ed wi h 2000 U/L hepa in (Hepa in-
Na ium-5000- a iopha m). Red blood cells we e lysed by adding 500 mL ed blood cell lysis bu e (eBioscience, #00-4300-54) and
incuba ed a oom empe a u e o 1 min. Samples we e spun down (1500 pm), esuspended in 100 mL PBS and pla ed in a 96 well
pla e. Spleens we e dissocia ed in o a single cell suspension using a 40 mm cell s aine (Falcon) and esuspended in 10 mL o PBS
(GIBCO). An aliquo o cell suspension was used o coun ing o calcula e o al numbe o cells pe spleen. 200 mL pe sample (app ox.
2x10
6
cells) we e pla ed in a 96 well pla e. Pla es we e spun and supe na an s we e emo ed.
Fo e ame s aining, cells we e esuspended in 25 mL PBS con aining GP33 (1:500) and NP396 (1:250) e ame s (NIH Te ame
Co e Facili y) and incuba ed a 37C o 15 min. Nex , 25 mL PBS con aining an i-CD16/32 (Biolegend; 1:200, clone: 93) we e added
and incuba ed o 10 min a oom empe a u e. 25 mL o a mas e mix o he desi ed su ace ma ke an ibodies (CD8.2b: Paci ic Blue
clone 53-5.8; CD4: FITC, clone H129.19; CD4: PE, clone GK1.5; CD3: APC, clone 145-2C11; CD44: BV605, clone IM7; CD62L:
AF700, clone MEL-14; CD19: APC-Cy7, clone 6D5; all Biolegend; 1:200 in PBS) and Fixable Viabili y Dye eFluo 780 (eBioscience;
1:2000 in PBS) we e added and cells we e incuba ed o 20 min a 4C. Cells we e washed wi h FACS bu e (PBS, 2% FCS) and ixed
in 4% Pa a o maldehyde (Sigma) in PBS o 10 min. Subsequen ly samples we e washed wice wi h FACS bu e , esuspended in
100 mL and analyzed by low cy ome y.
Fo in acellula cy okine s aining (ICS), cell pelle s we e esuspended in 50 mL o RPMI 1640 medium (GIBCO) supplemen ed wi h
10% FCS (PAA) and 1% Penicillin-S ep omycin-Glu amine (The mo Fishe Scien i ic), 50 mMb-me cap oe hanol (Sigma), con aining
LCMV pep ides (1:1000, Pep ide 2.0 Inc.) and P o ein T anspo Inhibi o Cock ail (eBioscience, #00-4980-03; 1:500, The mo Fishe
Scien i ic). As posi i e con ol, cells we e ea ed wi h Cell S imula ion Cock ail (eBioscience, #00-4970-93). Cells we e incuba ed o
4ha 37
C and hen su ace an igens we e s ained as desc ibed abo e. A e wa d, a mas e mix o desi ed an ibodies in 25 mL FACS
bu e con aining 0.05% saponin (Sigma, 47036) agains in acellula an igens o in e es we e added and incuba ed o 90 min a 4C
(IFNg: PE-Cy7, clone XMG1.2; IL-2: PE, clone JES6-5H4; TNFa: APC, clone MP6-XT22; all Biolegend. all 1:200). Nex , cells we e
washed wice wi h FACS bu e , esuspended in 100 mL and analyzed by low cy ome y.
Quan i a i e p o eomics
Th ee TMT 6-plex uns we e ca ied ou o moni o he changes in li e p o ein abundance du ing he en i e cou se o in ec ion. The
i s un included biological eplica es om day 2 and day 8 a e in ec ion along wi h he unin ec ed con ols. The second un included
eplica e samples o unin ec ed con ols, day 30 and day 60 a e in ec ion. To accoun o he e ec o aging, he hi d un included
unin ec ed con ols a day 0 and day 123, along wi h in ec ed samples om day 123 (all in biological eplica es). Iden ical unin ec ed
con ols (day 0) we e included in all h ee TMT 6-plex uns as an in e nal con ol o moni o ep oducibili y be ween he uns. Li e
issues we e homogenized (TissueLyse II, QIAGEN) in 1.5 mL 50 mM HEPES bu e , pH 8.5 supplemen ed wi h 2% sodium dodecyl
sul a e (SDS). The p o ein concen a ions we e de e mined by he bicinchoninic acid assay (BCA, Pie ce Bio echnology, The mo Sci-
en i ic, IL). Fu he p ocessing was adap ed om a il e -aided sample p epa a ion (FASP) me hod p e iously desc ibed (Manza e al.,
2005; Wisniewski e al., 2009). Fo each sample, 100 mg o he li e issue lysa e was educed wi h 100 mM di hio h ei ol (DTT) and
ans e ed in o VIVACON 500 il e uni (Vi ap ouc s, Li le on, MA). SDS-con aining bu e was emo ed om he sample by cen i-
uga ion and exchanged wi h 8 M u ea in 100 mM T is-HCl bu e . P o eins we e alkyla ed wi h 50 mM iodoace amide and washed
wi h 50 mM ie hyl ammonium bica bona e (TEAB). Finally, po cine ypsin (P omega, Madison, WI) was used o p o ein diges ion in
an enzyme o p o ein a io o 1:100 w/w.
Immuni y 51, 1074–1087.e1–e9, Decembe 17, 2019 e6

Fo ela i e p o ein quan i a ion, six samples om each un we e sepa a ely de i a ized wi h TMT 6-plex eagen s (ABI,
F amingham, MA) acco ding o he ins uc ions p o ided by he manu ac u e . The combina ion o he TMT 6-plex labels was as
ollows: (i) Run 1: day 0 unin ec ed (TMT 126 and 127), day 2 in ec ed (TMT 128 and 129) and day 8 in ec ed (TMT 130 and 131);
(ii) Run 2: day 0 unin ec ed (126 and 127), day 30 in ec ed (128 and 129) and day 60 in ec ed (130 and 131); and (iii) Run 3: day 0
(126 and 127), day 123 unin ec ed (128 and 129) and day 123 in ec ed (130 and 131).
The TMT-labeled yp ic diges s we e pooled and concen a ed by solid phase ex ac ion (SPE) (Mac oSpin columns 30-300 mg
capaci y, The Nes G oup, Sou hbo ough, MA, USA). Samples we e ea ed wi h 20 mM ammonium o ma e p io o injec ion
on o a Phenomenex column (150 32.0 mm GeminiNX-C18 3 mm 110A
˚; Phenomenex, To ance, CA, USA) in an Agilen 1200 se ies
HPLC (Agilen Bio echnologies, Palo Al o, CA) wi h UV de ec ion a 214 nm. HPLC sol en A consis ed o 20 mM ammonium o ma e,
pH 10 in 5% ace oni ile. Pep ides we e sepa a ed a low a e o 100 mL/min and elu ed om he column wi h a non-linea g adien
anging om 0 o 100% sol en B (20 mM ammonium o ma e, pH 10 in 90% ace oni ile). Se en y- wo ime-based ac ions we e
collec ed, acidi ied, and pooled in o 50 HPLC ials based on he UV ace. A e emo al o o ganic sol en in a acuum cen i uge,
samples we e econs i u ed o 10 mL wi h 5% o mic acid (Benne e al., 2011). Indi idual ac ions we e u he analyzed a pH 2.4 by
Agilen 1200 nano-HPLC sys em (Agilen Bio echnologies, Palo Al o, CA) coupled o a hyb id LTQ O bi ap Velos mass spec ome e
(The moFishe Scien i ic, Wal ham, MA). Da a we e acqui ed u ilizing he Xcalibu so wa e e sion 2.1. B ie ly, single ac ions we e
loaded on o a ap column (Zo bax 300SB-C18 5 mm, 5 30.3 mm, Agilen Bio echnologies, Palo Al o, CA) wi h a bina y pump a a low
a e o 45 mL/min. Sol en s o LCMS sepa a ion we e composed o 0.1% i luo ace ic acid (TFA) in wa e (sol en A) and 0.1% TFA in
70% me hanol and 20% isop opanol (sol en B). The pep ides we e elu ed by back- lushing om he ap column on o a 16 cm used
silica analy ical column wi h an inne diame e o 50 mm packed wi h C18 e e sed-phase ma e ial (Rep oSil-Pu 120 C18-AQ, 3 mm,
D . Maisch GmbH, Amme buch-En ingen, Ge many). Elu ion was achie ed wi h a 27 min g adien anging om 3 o 30% sol en B,
ollowed by a 25 min g adien om 30 o 70% sol en B and, inally, a 7 min g adien om 70 o 100% sol en B a a cons an low a e
o 100 nL/min. The analysis was pe o med in a da a-dependen acquisi ion mode. The 10 mos in ense ions we e isola ed and ag-
men ed by high-ene gy collision-induced dissocia ion (HCD) o pep ide iden i ica ion and ela i e quan i a ion o TMT epo e ions.
Dynamic exclusion o selec ed ions was 60 s and a single lock mass a m/z 445.120024 (Si(CH3)
2
O)6)
20
(Olsen e al., 2005) was used
o in e nal mass calib a ion wi h he a ge loss mass abundance o 0%. Maximal ion accumula ion ime allowed was 500 ms and
o e illing o he C- ap was p e en ed by au oma ic gain con ol se o 10
6
ions o a ull FTMS scan and 5 310
5
ions o MS
n
HCD. In ac pep ides we e de ec ed in he O bi ap mass analyze a esolu ion o 30,000 wi h he signal h eshold o 2,000 coun s
o igge ing an MSMS e en . The maximum ion scan ime was se o 200 ms o acqui ing 1 mic oscan a a esolu ion o 7500.
Ta ge ed LC-MS based me aboli e measu emen s
Tissue samples we e homogenized using a P ecellys 24 issue homogenize (P ecellys CK14 lysing ki , Be in). Pe mg issue, 3 mLo
80% ( / ) me hanol we e added. 10 mL o homogenized issue sample o se um we e mixed wi h 10 mL o an iso opically labeled in-
e nal s anda d mix u e in a hyd ophobic 96 well il e pla e. Aliquo s o 300 mL o me hanol we e added and mixed o 20 min a
450 pm. A e wa d, he sample ex ac s we e collec ed by cen i uging he il e pla e o 5 min a 500 g. A Vanquish UHPLC sys em
(The mo Scien i ic) coupled wi h an O bi ap Q Exac i e (The mo Scien i ic) mass spec ome e was used o he LC-MS analysis. The
ch oma og aphic sepa a ion o samples was ca ied ou on an ACQUITY UPLC BEH Amide, 1.7 mm, 2.1x100 mm analy ical column
(Wa e s) equipped wi h a VanGua d: BEH C18, 2.1x5mm p e-column (Wa e s). The column was main ained a a empe a u e o 40C
and 2 mL sample we e injec ed pe un. The mobile phase A was 0.15% o mic acid ( / ) in wa e and mobile phase B was 0.15%
o mic acid ( / ) in 85% ace oni ile ( / ) wi h 10 mM ammonium o ma e. The g adien elu ion wi h a low a e 0.4 mL/min was pe -
o med wi h a o al analysis ime o 17 min. The O bi ap Q Exac i e (The mo Scien i ic) mass spec ome e was ope a ed in an elec-
osp ay ioniza ion posi i e mode, sp ay ol age 3.5 kV, aux gas hea e empe a u e 400C, capilla y empe a u e 350C, aux gas
low a e 12. The me aboli es o in e es we e analyzed using a ull MS scan mode, scan ange m/z 50 o 400, esolu ion 35000,
AGC a ge 1e6, maximum IT 50ms. The T ace Finde 4.1 so wa e (The mo Scien i ic) was used o he da a p ocessing. Se en-poin
linea calib a ion cu es wi h in e nal s anda diza ion and 1/x weighing was cons uc ed o he quan i ica ion o me aboli es.
Me aboli e measu emen s using he Absolu eIDQ p180 Ki (Bioc a es Li e Science AG) we e pe o med as desc ibed p e iously
(S . John-Williams e al., 2017). In b ie , pe mg o issue, 6 mL e hanol/phospha e bu e (85:15 / e hanol/10 mM phospha e bu e )
we e added and he issue was homogenized (TissueLyse II, QIAGEN). Samples we e cen i uged a 5,000 g o 5 min a 4C and he
supe na an s we e ans e ed o a esh ube and used o analyses. Fo se um samples, blood was collec ed om mice in blood
collec ion ubes and cen i uged a 12,000 g o 5 min o ob ain se a. The se um was ans e ed in o a esh ube and s o ed
a 80C un il analyses. The samples we e analyzed on a Xe o TQ-MS (Wa e s) mass spec ome e using an Acqui y UHPLC (Wa-
e s) sys em, ope a ed wi h MassLynx V4.1 (Wa e s). Samples and addi ional blanks, calib a ion s anda ds measu emen s, quali y
con ols, and analyses we e p epa ed acco ding o he use manual.
QUANTIFICATION AND STATISTICAL ANALYSIS
RNA-seq da a p ocessing
RNA quali y and in eg i y we e assessed ia an Expe ion RNA HighSense chip (Bio ad). The lib a y o RNA-seq was p epa ed using
he T uSeq RNA sample p epa a ion ki 2 (Illumina) acco ding o he manu ac u e ‘s p o ocol. Quali y con ol analysis was pe -
o med on all samples o he cDNA lib a y by Expe ion DNA Analysis chip (Bio ad) and Qubi Fluo ome ic quan i a ion (Li e Tech).
e7 Immuni y 51, 1074–1087.e1–e9, Decembe 17, 2019
7 o up o 17 samples we e mul iplexed pe lane and un on a 50bp single-end low cell in a HiSeq2000 o HiSeq3000 sequence
(Illumina), espec i ely. Called bases by he Illumina Real ime Analysis so wa e we e con e ed in o BAM o ma using Illumina2bam
and demul iplexed using BamIndexDecode (h ps://gi hub.com/w si-npg/illumina2bam). The RNA-seq analysis pipeline was pe -
o med wi h Tuxedo. Reads we e mapped on he mouse e e ence genome (Mus musculus, Ensembl e87, Decembe 2016) using
TopHa 2 ( 2.0.10). Cu links ( 2.2.1) was employed o assemble ansc ip s om spliced ead alignmen s, using he Ensembl e87
ansc ip ome as he e e ence as well as de no o assembly o ansc ip models. Fu he , di e en ial analysis o gene exp ession
was quan i ied wi h Cu di ( 2.2.1). T ansc ip ome se s o all eplica es o each sample g oup we e combined wi h Cu me ge.
Exp ession alues in g aphs a e epo ed as FPKM ( agmen s pe kilobase p ansc ip pe million). Di e en ial gene exp ession
is a es ed based on exp ession le el R1 FPKM, adjus ed p alue %0.05 and absolu e log2 old-change o 1 (hea map) o 0.6 (ci cos
plo ).
Single cell RNA-seq p ocessing
We pe o med single-cell RNA-sequencing on a 10X Genomics Ch omium Single Cell Con olle wi h he Ch omium Single Cell 30V3
Ki ollowing he manu ac u e ’s ins uc ions. Sequencing was pe o med on an Illumina HiSeq 3000 ins umen in 2x75bp pai ed-end
mode. We used he Cell Range ( 3.0.2) 10X Genomics so wa e o demul iplex he aw sequencing da a and align hem o he mouse
GRCm38 e e ence genome. We p oceeded wi h he analysis o he UMI coun s using he R Bioconduc o package Seu a 3.1.0
(S ua e al., 2019). The wo samples, unin ec ed and 2 days pos in ec ion, we e me ged and p ocessed oge he . Cells wi h
mo e han 30% mi ochond ial con en we e disca ded. Mi ochond ial genes, as well as genes ha we e no de ec ed in mo e
han 1 cell we e disca ded. We used ma ke s o Kup e (Clec4 , Cs 1 ,C1qc,C1qa,C1qb), endo helial (Kd ,Eg l7,Ig bp7,Aqp1)
and hepa ocy e (Apoa1,Apob,Pck1,G6pc,T ) cells o cell iden i ica ion (Halpe n e al., 2017). In o de o ocus on hepa ocy es,
we kep only hose cells ha had a summed exp ession o hepa ocy e ma ke s g ea e han bo h he summed exp ession o Kup e
as well as endo helial cells ma ke s. We also elimina ed cells ha had less han 0.5% o Alb eads. The no maliza ion o he UMI
coun s was pe o med wi h he SCT ans o m om he Seu a Package, wi h he eg ession a iable on he condi ion (unin ec ed
and 2 dpi) (Ha emeis e and Sa ija, 2019). We u he an PCA on he no malized coun s and he 3.000 mos a iable genes. Based
on he Jack S aw me hod we selec ed he i s 20 p incipal componen s explaining mos o he a iabili y and p oceeded wi h a
UMAP low-dimensional p ojec ion. Di e en ial analysis be ween he 2 condi ions was pe o med using he FindMa ke s unc ion
in Seu a , based on he Wilcox es wi h a Bon e oni co ec ion. Genes we e conside ed as di e en ially exp essed i hey had an
absolu e a e age log2 old-change g ea e han 0.2 and an adjus ed p alue smalle han 0.05.
P incipal componen analysis
P incipal componen analysis (PCA) was pe o med on he gene se wi h a minimum a e age exp ession le el ac oss condi ions o 5
FPKM. Only he 10% mos a iable (compu ed on he coe icien o a iance) genes we e conside ed o he PCA analysis.
Hie a chical clus e ing
Hie a chical clus e ing o di e en gene se s FPKM o CPM exp ession as well as me aboli e abundance alues we e pe o med
using a Pea son dis ance measu e wi h an a e age clus e ing me hod. A k-means ++ (z-no m) clus e ing app oach using
Exp essClus e so wa e 1.3 (h p://cbdm.hms.ha a d.edu/LabMembe sPges/SD.h ml) was pe o med on he union o di e en-
ially modula ed genes o me aboli es.
In e ac ion model
Fo he in e ac ion model (2x2 ac o ial design) o nai e e sus LCMV-in ec ed I na 1
D/D
and I na 1
+/+
, we quan i ied gene exp ession
as he numbe o eads co e ing each gene. The gene exp ession on he mouse Ensembl e87 ansc ip s was quan i ied om he
p e iously TopHa 2 mapped eads wi h ea u eCoun s (Liao e al., 2014). Raw ead coun s we e no malized wi h he oom (Law
e al., 2014) unc ion o he limma package (Smy h, 2004). No malized exp ession alues, epo ed as log2 coun s pe million
(CPM), we e hen p ocessed h ough limma’s empi ical Bayes models. We implemen ed limma’s in e ac ion model as a wo
(I na 1
+/+
,I na 1
D/D
) by wo (nai e, LCMV-in ec ed) ac o ial design. Genes di e en ially modula ed in he in e ac ion model ha e
been selec ed based on a minimum log2 CPM o 0, an adjus ed p alue %0.05 and a minimum log2 old-change absolu e alue
o 1 (hea map) o 0.6 (ci cos plo ).
En ichmen s and pa hway analyses
En ichmen analyses on he union o di e en ial modula ed en i ies ( ansc ip s and/o p o eins) speci ic clus e s we e done in Cy o-
scape ClueGO (Bindea e al., 2009) 2.3.3, based on GO (Biological P ocesses, Molecula Func ions, Immune Sys em P ocess), In-
e P o, KEGG, Reac ome and Wiki Pa hways. Te ms we e called en iched based on a maximum p alue o 0.05 and a minimum o 2%
gene o e lap. GO Te m Fusion and g ouping was applied. En iched g oups whe e u he anked acco ding o he g oup Bon e oni
s ep-down adjus ed p alue.
Fo me abolic pa hway en ichmen analyses, we ook he union o di e en ially exp essed genes (exp ession R1 FPKM, absolu e
log2 old-change R0.7, adjus ed p alue %0.05) and p o eins (absolu e log2 old-change R0.25 and adjus ed a io p alue < 0.05
and sample p alue < 0.05) ac oss all ime poin s. We ex ac ed me abolism-associa ed genes om KEGG (Kanehisa and Go o, 2000)
me abolic pa hways da abase. Only he pa hways wi h q alue o en ichmen < 0.05 we e conside ed. The en iched pa hways a e
Immuni y 51, 1074–1087.e1–e9, Decembe 17, 2019 e8
ep esen ed as a ci cos plo (K zywinski e al., 2009), wi h he wid h o each ibbon in a gi en ca ego y ep esen ing he pe cen age o
genes a each ime poin among all genes leading o en ichmen s in he espec i e ca ego y. The colo g adien ( om ligh e o da ke )
ep esen s he pe cen age o pa hways in each ca ego y ha we e ound as en iched a a speci ic ime poin .
Mass spec ome y da a p ocessing
The acqui ed aw MS da a iles we e p ocessed wi h mscon e (P o eoWiza d Lib a y 2.1.2708) and con e ed in o MASCOT
gene ic o ma (mg ) iles. Pep ides we e iden i ied by sea ching he esul an peak lis s agains he SwissP o mouse da abase
e sion 2013.01_20130110 (24615 sequences; 14280050 esidues) wi h he sea ch engines MASCOT ( 2.3.02, Ma ixScience, Lon-
don, UK) and Phenyx ( 2.5.14, GeneBio, Gene a, Swi ze land). Submission o he sea ch engines was done ia a Pe l sc ip ha pe -
o ms an ini ial sea ch wi h ela i ely b oad mass ole ances (MASCOT only) on bo h he p ecu so and agmen ions (±10 ppm and ±
0.6 Da, espec i ely). High-con idence pep ide iden i ica ions we e used o ecalib a e all p ecu so s and agmen ion masses p io
o a second sea ch wi h na owe mass ole ances (±4 ppm and ± 0.025 Da). T ypsin was chosen as clea age speci ici y wi h he
maximum o 1 misclea age si e allowed. Ca bamidome hyl cys eine, N- e minal and lysine-modi ied TMT 6-plex we e se as ixed
modi ica ions, whe eas oxidized me hionine was se as a a iable modi ica ion.
To alida e he p o eins, MASCOT and Phenyx ou pu iles we e p ocessed by in e nally de eloped pa se s. P o eins wi h R2
unique pep ides abo e a sco e T1, o wi h a single pep ide abo e a sco e T2, we e selec ed as unambiguous iden i ica ions. Addi-
ional pep ides o hese alida ed p o eins wi h sco e > T3 we e also accep ed. Fo MASCOT and Phenyx, T1, T2, and T3 pep ide
sco es we e equal o 16, 40, 10 and 5.5, 9.5, 3.5, espec i ely (P alue < 10
3
). The alida ed p o eins e ie ed by he wo
algo i hms we e me ged, any spec al con lic s disca ded, and g ouped acco ding o sha ed pep ides. A alse posi i e de ec ion
a e (FDR) o < 1% and < 1% was de e mined o p o eins and pep ides, espec i ely, by applying he same p ocedu e agains a
e e sed da abase.
The log2 old-change cu o s o di e en ial p o ein modula ion we e de e mined based on pai wise compa isons o p o ein abun-
dances be ween he wo eplica es o unin ec ed con ol samples ac oss he 3 independen uns. The 2.5% and 97.5% qua iles o
he in e - eplica pai wise log2 old-change was compu ed. Based on he qua ile alues, we ha e se he cu o a 0.25 and 0.25 o
up- and down-modula ed p o eins, espec i ely. Addi ional o he log2 old-change, s a is ical signi icance o obse ed changes was
calcula ed by Isoba so wa e (B ei wiese e al., 2011). Adjus ed P alue a io as well as he P alue samples as calcula ed by Isoba
we e asked o be less han 0.05.
Ta ge ed LC-MS based me aboli e quan i ica ion
Me aboli e measu emen s using he Absolu eIDQ p180 Ki (Bioc a es Li e Science AG) we e alida ed wi h he supplied so wa e
Me IDQ, Ve sion 5-4-8-DB100-Bo on-2607 (Bioc a es Li e Sciences). All me aboli e abundances in e io o he limi o de ec ion
(LOD) we e eplaced wi h a alue equal o hal he LOD. We elimina ed all me aboli es ha did no ha e an a e age abundance su-
pe io o LOD in a leas one condi ion. Modula ion o me aboli es was assessed h ough a es be ween condi ions. The cu o s o
he modula ion we e ob ained based on all he in e - eplica pai wise compa isons o wild- ype unin ec ed samples. The 2.5 and 97.5
qua iles o hese in e - eplica log old-change we e used o de ine he cu o s. The cu o s we e 0.44 and 0.65 o he whole li e
dynamics se um me aboli e measu emen s. Fo se um measu emen s o I na 1
D/D
and I na 1
+/+
mice a 1.5 days pos in ec ion,
based on he same app oach, we ha e se he cu o s a 0.73 and 0.4. Signi icance was in e ed based on a p alue in e io o
0.05. Fo he in e ac ion model (2x2 ac o ial design) o nai e e sus LCMV-in ec ed I na 1
D/D
and I na 1
+/+
, we implemen ed limma’s
in e ac ion model as a wo (I na 1
+/+
,I na 1
D/D
) by wo (nai e, LCMV-in ec ed) ac o ial design. Signi ican ly modula ed me aboli es in
he in e ac ion model we e iden i ied by an absolu e log2 old change o 0.7 and a p alue in e io o 0.05.
Me aboli e acing quan i ica ion
Me aboli e acing da a we e p ocessed using he T aceFinde 4.1 so wa e (The mo Scien i ic).
Immunohis ochemis y quan i ica ion
Images we e pho og aphed using an Olympus BX 53 mic oscope, and we e quan i ied using His oQues TM so wa e
(TissueGnos ics GmbH, Vienna, Aus ia).
S a is ical in o ma ion
Da a a e p esen ed as a i hme ic mean ± SEM. S a is ical signi icances we e calcula ed using a S uden ’s es when compa ing wo
g oups o using wo-way ANOVA wi h Bon e oni co ec ion when compa ing longi udinal changes. * p < 0.05 ** p < 0.01 *** p < 0.001.
DATA AND CODE AVAILABILITY
P o eomic da a (PRIDE: PXD011122) and ansc ip omic da a (GEO: GSE123688, which includes GEO: GSE118703, GSE123684,
GSE118819 and GSE137082 and PRIDE: PXD011122) a e deposi ed in he PRIDE and GEO da abases, espec i ely.
e9 Immuni y 51, 1074–1087.e1–e9, Decembe 17, 2019