A icle
Cellula Impo in-a3 Exp ession Dynamics in he
Lung Regula e An i i al Response Pa hways agains
In luenza A Vi us In ec ion
G aphical Abs ac
Highligh s
dImpo in-a3 is he mos abundan ly exp essed iso o m in he
mammalian lung
dImpo in-a3 is highly conse ed ac oss species
dImpo in-a3 is one o he majo nuclea anspo e s o NF-kB
dImpo in-a3 ac s as an immune senso o in luenza A i us
in ec ions
Au ho s
Swan je Thiele,
S ephanie S anelle-Be am,
Sebas ian Beck, ..., Michael Bade ,
Enno Ha mann, G€
ulsah Gab iel
Co espondence
guelsah.gab i[email p o ec ed]
In B ie
Thiele e al. show ha impo in-a3 is one
o he majo nuclea anspo e s o NF-
kB in he mammalian lung. High-le el
TNF-a-inducing HPAIVs inhibi impo in-
a3 mRNA ansc ip ion by in e e ing wi h
i s p omo e ac i i y. Thus, HPAIVs may
e ade an i i al immuni y in he espi a o y
ac by gene a ing a bo leneck in
impo in-a3 a ailabili y.
Nucleoplasm
TNF-α
TNFR
TNF-α
TNFR
A ian- ype HPAIV
PB2 , PB2
Human- ype HPAIV
PB2 , PB2
RNP
Cy oplasm
Nucleoplasm
Cy oplasm
an i i al genes an i i al genes
α3
p50
p65
α3
p50
p65
α3
p50
p65
α3
p50
p65
p50
p65
−+
RNP
p50
p65
701D
627E 701N627K
Thiele e al., 2020, Cell Repo s 31, 107549
Ap il 21, 2020 ª2020 The Au ho (s).
h ps://doi.o g/10.1016/j.cel ep.2020.107549
Cell Repo s
A icle
Cellula Impo in-a3 Exp ession Dynamics
in he Lung Regula e An i i al Response Pa hways
agains In luenza A Vi us In ec ion
Swan je Thiele,
1
S ephanie S anelle-Be am,
1
Sebas ian Beck,
1
Nancy Mounogou Kouassi,
1
Ma in Zickle ,
1
Ma in M€
ulle ,
1
Be in Tuku,
1
Pa icia Resa-In an e,
1,12
Debby an Riel,
1,3
Malik Alawi,
4,5
Thomas G€
un he ,
5
F anziska Ro he ,
6,7
S e anie H€
ugel,
6
Susanne Reime ing,
8
Alice McHa dy,
8
Adam G undho ,
5
Wol am B une,
9
Albe Os e haus,
10
Michael Bade ,
6,7,11
Enno Ha mann,
7
and G€
ulsah Gab iel
1,2,13,
*
1
Vi al Zoonosis – One Heal h, Hein ich Pe e Ins i u e, Leibniz Ins i u e o Expe imen al Vi ology, Hambu g, Ge many
2
Ins i u e o Vi ology, Uni e si y o Ve e ina y Medicine, Hanno e , Ge many
3
Depa men o Vi oscience, E asmus Medical Cen e , Ro e dam, he Ne he lands
4
Bioin o ma ics Se ice Facili y, Uni e si y Medical Cen e Hambu g-Eppendo , Hambu g, Ge many
5
Vi us Genomics, Hein ich Pe e Ins i u e, Leibniz Ins i u e o Expe imen al Vi ology, Hambu g, Ge many
6
Molecula Biology o Pep ide Ho mones, Max Delb €
uck Cen e o Molecula Medicine, Be lin, Ge many
7
Ins i u e o Biology, Cen e o S uc u al and Cellula Biology in Medicine, Uni e si y o L€
ubeck, L€
ubeck, Ge many
8
Compu a ional Biology o In ec ion Resea ch, Helmhol z Cen e o In ec ion Resea ch, B aunschweig, Ge many
9
Vi us-Hos In e ac ion, Hein ich Pe e Ins i u e, Leibniz Ins i u e o Expe imen al Vi ology, Hambu g, Ge many
10
Resea ch Cen e o Eme ging In ec ions and Zoonoses, Uni e si y o Ve e ina y Medicine Hanno e , Hanno e , Ge many
11
Cha i e
´-Uni e si a
¨ smedizin, Be lin, Ge many
12
P esen add ess: AIDS Resea ch Ins i u e I siCaixa, Ba celona, Spain
13
Lead Con ac
*Co espondence: guelsah.g[email p o ec ed]
h ps://doi.o g/10.1016/j.cel ep.2020.107549
SUMMARY
Impo in-aadap o p o eins o ches a e dynamic nu-
clea anspo p ocesses in ol ed in cellula homeo-
s asis. He e, we show ha impo in-a3, one o he
main NF-kB anspo e s, is he mos abundan ly ex-
p essed classical nuclea anspo ac o in he
mammalian espi a o y ac . Impo in-a3 p omo e
ac i i y is egula ed by TNF-a-induced NF-kBina
concen a ion-dependen manne . High-le el TNF-
a-inducing highly pa hogenic a ian in luenza A i-
uses (HPAIVs) isola ed om a al human cases
ha bo ing human- ype polyme ase signa u es (PB2
627K, 701N) signi ican ly down egula e impo in-a3
mRNA exp ession in p ima y lung cells. Impo in-a3
deple ion is es o ed upon back-mu a ing he HPAIV
polyme ase in o an a ian- ype signa u e (PB2 627E,
701D) ha can no longe induce high TNF-ale els.
Impo in-a3-de icien mice show educed NF-kB-
ac i a ed an i i al gene exp ession and inc eased
in luenza le hali y. Thus, impo in-a3 plays a key
ole in an i i al immuni y agains in luenza. Li ing
he bo leneck in impo in-a3 a ailabili y in he lung
migh p o ide a new s a egy o comba espi a o y
i us in ec ions.
INTRODUCTION
Molecula a icking o p o eins be ween he cy oplasm and he
nucleus is a undamen al p ocess ha is essen ial o main ain
cellula homeos asis (Go
¨ lich and Ma aj, 1996; Imamo o e al.,
1995). Dis u bances in hese highly sensi i e egula o y p o-
cesses may esul in he disequilib ium o cellula and nuclea
p o eins and e en ually cause diseases. Impo in-ap o eins
belong o he majo nuclea anspo ac o s in he cell shu ling
a ious ca go p o eins con aining a nuclea localiza ion signal
(NLS) om he cy oplasm o he nucleus (Gold a b e al., 2004).
Impo in-ap o eins a e acidophilic p o eins ha bind as adap-
o s o ca go p o eins con aining a basic s e ch o mono- o
bipa i e NLS mo i s. This impo in-a/ca go complex o ms a
ime ic complex wi h he ecep o p o ein impo in-b
1
, which
inally ansloca es h ough he nuclea po e by he in e ac ion
o impo in-b
1
wi h nucleopo ins (Hu e al., 1996; Radu e al.,
1995; Rexach and Blobel, 1995). In he nucleus, he guanosine
iphospha e (GTP)-bound o m o Ran binds o impo in-b
1
and media es he dissocia ion and elease o he ca go p o ein
in o he nucleus (Lee e al., 2005; Ma suu a and S ewa ,
2005). A e success ul nuclea impo , impo in-abinds o he
ecycling ac o cellula apop osis suscep ibili y p o ein (CAS)
ha , in conjunc ion wi h RanGTP, anspo s he complex back
in o he cy oplasm whe e impo in-ais again a ailable o he nu-
clea impo o o he ca go p o eins (Ku ay e al., 1997). Impo in-
ap o eins a e highly conse ed h oughou e olu ion (Gold a b
e al., 2004). In euka yo es, se en impo in-aiso o ms we e
desc ibed, whe eas he budding yeas encodes one impo in-
a, and lies and nema odes encode o h ee impo in-aiso o ms.
Howe e , all o hese o ganisms encode o only one impo in-b
1
ecep o p o ein. The speci ici y o impo in-aiso o ms o ca go
p o eins is de e mined by he complexi y and he h ee-dimen-
sional con ex (F ied ich e al., 2006; Gold a b e al., 2004; Sank-
hala e al., 2017). Mo eo e , impo in-ap o eins play a key ole in
cellula di e en ia ion by egula ing he anspo o ansc ip ion
Cell Repo s 31, 107549, Ap il 21, 2020 ª2020 The Au ho (s). 1
This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/).
ac o s h ough lineage-speci ic exp ession p o iles in cells (Ya-
suha a e al., 2007). Howe e , li le is known abou he exp es-
sion p o iles o impo in-aiso o ms in o gans ha could gi e
impo an insigh s in o molecula mechanisms in heal h and dis-
ease. The e is accumula ing e idence ha unde cellula s ess,
such as oxida i e s ess, hea shock, o ul a iole i adia ion, im-
po in-ap o eins may accumula e in he nucleus and dis up
cellula homeos asis (Fu u a e al., 2004; Kodiha e al., 2004; Ya-
suda e al., 2012). In he las decade, a g owing amoun o e i-
dence sugges s ha impo in-ap o eins a e u ilized by a ious
i uses (e.g., in luenza, Ebola, dengue, hepa i is C, and co ona-
i uses) o ci cum en hos de ense mechanisms (Basle and
Ama asinghe, 2009; Can on e al., 2018; F ase e al., 2014; F ie-
man e al., 2007; Gab iel e al., 2011; Hudje z and Gab iel, 2012;
Kopecky-B ombe g e al., 2007; Neu eld e al., 2013; P yo
e al., 2007). Li le e idence is a ailable on hei mode o ac ion
in speci ic o gans unde in lamma o y condi ions such as i us
in ec ions. In his s udy, we iden i ied he ana omic exp ession
p o ile o impo in-aiso o ms in he mammalian lung and s udied
hei ansc ip ional egula ion and impac on in luenza- i us-
induced pneumonia using cell cul u e models, genome-wide
ansc ip ion analyses, and ansgenic mouse models.
RESULTS
Impo in-a3 Is he Mos Abundan ly Exp essed Impo in-
aIso o m in he Mammalian Respi a o y T ac
In o de o analyze he exp ession pa e n o majo impo in-a
iso o ms h oughou he mammalian espi a o y ac (RT), we
ana omically dissec ed he human and mu ine RT in o an uppe
pa de ined by espi a o y cells o he nasal concha (URT) and a
lowe pa de ined by b onchi and al eoli (LRT). Impo in-a1, -a3,
and -a5/7 iso o ms (Kpna2,Kpna4, and Kpna1/6, espec i ely)
we e abundan ly exp essed in epi helial cells, glands, and mac-
ophages o he human (Figu e 1A) as well as he mu ine URT and
LRT (Figu e 1B). We hen measu ed he ela i e mRNA exp es-
sion le els o impo in-aiso o ms along he mammalian RT. In
he human URT and LRT, impo in-a3 was by a he mos abun-
dan ly exp essed iso o m, wi h 4–5 imes inc eased mRNA le els
compa ed o impo in-a1. A e age exp ession le els o o he
iso o ms we e simila o hose o impo in-a1, excep o impo -
in-a7, which was he second-mos abundan iso o m (Fig-
u e 1C). Simila ly, impo in-a3 was also he highes -exp essed
iso o m, ollowed by impo in-a7 in he mu ine URT and LRT.
Lase -based mic odissec ion o epi helial cells was used o
ensu e homogeneous cell popula ions (Figu e 1D). We u he
con i med he exclusi e abundance o impo in-a3 in he mu ine
lung, ollowed by impo in-a7 as he second-mos abundan iso-
o m also on he p o ein le el, using whole-o gan homogena es
(Figu es 1E and 1F). No ably, in immo alized cells (Figu es
S1A–S1D) and unlike in p ima y cells, impo in-a1 is he mos
abundan iso o m and is he e o e used as a p ognos ic cance
ma ke (Ch is iansen and Dy skjø , 2013; Dahl e al., 2006; Kau
e al., 2004). Thus, i is pa icula ly impo an o in es iga e he
ole o impo in-aiso o ms ei he in p ima y cells o in an in i o
animal model. We hen de e mined he exp ession o impo in-a
iso o ms in a ious mu ine o gans on he mRNA and p o ein
le els and con i med ha he lung (Figu e S1E) is unique in exhib-
i ing exclusi ely high impo in-a3 le els, compa ed o o he solid
o gans (Figu es S1F–S1K). Mo eo e , compa able dis ibu ion
and exp ession le els o impo in-aiso o ms wi hin he human
and mu ine RT sugges ha he mouse model may be consid-
e ed adequa e o s udy he impac and mode-o -ac ion o hese
cellula ac o s in heal h and disease. Nex , we di ec ly
compa ed ela i e impo in-aamoun s in lase -mic odissec ed
epi helial samples o he URT e sus he LRT (Figu es 1G–1K).
Fo all impo in-aiso o ms, mRNA le els we e highes in he al e-
ola issue (LRT (A), up o 6 imes) ollowed by epi helial cells o
he b onchi and b onchioles (LRT [b onchiola epi helium, BE]).
The lowes mRNA le els we e de ec ed in epi helial cells o he
nasal concha (URT). Thus, impo in-aiso o ms a e exp essed
as a g adien in he RT wi h g adually inc easing mRNA le els
om he URT owa d he LRT (Figu e 1L). Howe e , impo in-
a3 is he mos abundan ly exp essed impo in-aiso o m in he
mammalian RT, unlike in o he solid o gans.
Impo in-a3 Media es Nuclea T ansloca ion o TNF-
a-Ac i a ed NF-kB
NF-kB subuni s (p50, p65) a e ansloca ed in o he nucleus in an
impo in-a-media ed manne . He ein, pa icula ly impo in-a3
and i s close ela i e, impo in-a4, we e shown o ac as one o
he main iso o ms ha impo s umo nec osis ac o alpha
(TNF-a)-ac i a ed p50/p65 he e odime s in o he nucleus ia
di ec in e ac ion (Fage lund e al., 2005). Since he mammalian
lung mos abundan ly exp esses he impo in-a3 iso o m, we
con inued o dissec he physical and unc ional in e ac ion be-
ween impo in-a3 and NF-kB. We could con i m ha pu i ied
human impo in-a3 p o ein di ec ly in e ac s wi h NF-kB p50
and p65 subuni s as well as wi h i s p ecu so p105 p o ein
upon ac i a ion wi h TNF-aunlike impo in-a1 ha was used
as a nega i e con ol (Figu es 2A and 2B). Consis en ly, ea -
men o mu ine ib oblas s wi h TNF-amos signi ican ly
inc eased NF-kB le els in nuclea cell ac ions along wi h impo -
in-a3(Figu e 2C). In o de o con i m ha impo in-a3is
equi ed o he nuclea impo o NF-kB, we gene a ed mu ine
emb yo ib oblas s (MEFs) wi h a dele ed impo in-a3 gene
(a3
/
). Nuclea ac iona ion analysis in TNF-a- ea ed a3
/
MEFs showed signi ican ly educed nuclea ansloca ion o
he NF-kB subuni p65 compa ed o wild- ype (WT) con ol
(a3
+/+
) MEFs (Figu e 2D). Immuno luo escence analysis u he
con i med ha nuclea localiza ion o NF-kB p65 was also
educed in a3
/
MEFs compa ed o a3
+/+
MEFs (Figu e 2E).
Con e sely, NF-kB p65 accumula ed in he cy oplasm o a3
/
MEFs, albei sligh ly, likely due o he deg ada ion o ine icien ly
anspo ed ca go p o eins (Figu e 2E). The ex en o educed
nuclea localiza ion o impo in-a3 was dependen on he
me hod used, albei he absence o impo in-a3 consis en ly
impai ed e icien nuclea localiza ion o TNF-a-induced NF-kB
subuni p65. These da a clea ly show ha impo in-a3 ac s as
a nuclea ansloca o o TNF-a-ac i a ed NF-kB.
Impo in-a3 P omo e Ac i i y Is Regula ed
by TNF-a-Ac i a ed NF-kB
We u he obse ed an inc ease in gene al impo in-a3 p o ein
le els upon TNF-a ea men in o al cell lysa es, including nu-
clea and cy oplasmic cell ac ions o mu ine as well as p ima y
2Cell Repo s 31, 107549, Ap il 21, 2020
Human URT
A
Epi helial cells Glands
α1
α3
α5/7
Human LRT (Al eoli)
Epi helial cells Mac ophages
Human LRT (B onchi)
Epi helial cells
Impo in
α1α3α5/7
Mu ine URT
B
Epi helial cells
Mu ine LRT (Al eoli)
Epi helial cells Mac ophages
Mu ine LRT (B onchi)
Epi helial cells
Impo in
C
Human LRT
*
**
D
Human URT
α1
0
1
2
3
4
5
6
7
8
α3 α4 α5 α7
Impo in
Rela i e mRNA exp ession le els
(N0 (GOI) / N0 (GAPDH))
*
**
Mu ine URT Mu ine LRT
***
*
**
***
**
Impo in-α1
H
KJ
G
URT LRT (BE) LRT (A)
4
6
8
10
12
Impo in-α3
URT LRT (BE) LRT (A)
Impo in-α5
URT LRT (BE) LRT (A)
Impo in-α7
URT LRT (BE) LRT (A)
IImpo in-α4
URT LRT (BE) LRT (A)
***
***
***
**
***
**
***
***
***
E
0
1
2
3
4
5
6
7
8
0
1
2
3
4
5
6
7
8
0
1
2
3
4
5
6
7
8
Rela i e mRNA exp ession le els
(N0 (GOI) / N0 (GAPDH))
α1 α3 α4 α5 α7
Impo in
α1 α3 α4 α5 α7
Impo in
α1 α3 α4 α5 α7
Impo in
Rela i e mRNA exp ession le els
(N0 (GOI) / N0 (GAPDH))
0
2
4
6
8
10
12
0
2
Rela i e mRNA exp ession le els
(N0 (GOI) / N0 (GAPDH))
4
6
8
10
12
0
2
4
6
8
10
12
0
2
4
6
8
10
12
0
2
URT
LRT
B onchi
Al eoli
Impo in-α
LNasal Concha
_
Rela i e impo in-α p o ein amoun
(α /GAPDH)
Mu ine LRT
0
5
10
15
20
α1 α3 α5 α7
Impo in
*
Impo in-α3
GAPDH
MW
[kDa]
58
37
Mu ine
LRT
Impo in-α1
Impo in-α5
Impo in-α7
60
58
GAPDH
37
GAPDH
37
Dilu ion Impo in-α FLAG
1:5 1:200
1:5 1:100
1:5 1:50 1:5 1:50
F
*
**
Figu e 1. Impo in-aExp ession in he Mammalian Respi a o y T ac
(A and B) Impo in-a1, -a3, o -a5/7 p o eins we e s ained ( ed) in human (A) o mu ine (B) uppe (URT, nasal epi helial cells) and lowe espi a o y ac (LRT,
b onchi and al eoli) sec ions by IHC-P (immunohis ochemis y-pa a in p o ocol) and coun e s ained wi h hema oxylin. The impo in-a7 an ibody c oss- eac s
wi h impo in-a5. 4003(URT and b onchi in A) o 10003(al eoli in A and B) o iginal magni ica ion. Scale ba , 10 mm.
(C and D) Impo in-amRNA le els in he human (C) and mu ine (D) URT and LRT de e mined by qRT-PCR. Rela i e impo in-a1 exp ession alues we e se o 1
a e no maliza ion agains GAPDH (Glyce inaldehyd-3-phospha -dehyd ogenase). Each da a poin ep esen s an indi idual sample (n = 6–12).
(E and F) Impo in-ap o ein amoun s in mu ine LRT. Wes e n blo analyses using impo in-aiso o m-speci ic an ibodies and GAPDH adjus men we e pe o med
o de e mine endogenous impo in-ap o ein amoun s (a1, a3, a5, a7) in mu ine o gans. Fo ela i e quan i ica ion (E), s anda d cu es o a ini y-pu i ied, FLAG-
agged impo in-ap o eins we e used. Rela i e impo in-a1 p o ein amoun in mu ine LRT was se o 1. Da a shown ep esen means ±SD (n = 3 biological
eplica es; echnical eplica es: n = 1–2 pe o gan, n = 1–2 wes e n blo analyses). Rep esen a i e wes e n blo s o he endogenous impo in-aiso o ms and he
s anda d cu es a e shown in (F). The gaps in (F) depic c opping o he ele an bands un on he same gel.
(G–K) Compa ison o impo in-amRNA exp ession le els be ween mu ine URT and LRT: a1 (G), a3 (H), a4 (I), a5 (J), o a7 (K). A, al eoli; BE, b onchiola
epi helium; n = 8–12.
(L) Schema ic pa i ion o he RT wi h inc easing impo in-amRNA le els om URT o LRT.
*p < 0.05; **p < 0.01; ***p < 0.001.
Cell Repo s 31, 107549, Ap il 21, 2020 3
F
G
K
LM
D
E
H
IJ
N
ABC
O
(legend on nex page)
4Cell Repo s 31, 107549, Ap il 21, 2020

ai way epi helial cells (Figu es 2F and 2G). These indings migh
sugges ha impo in-a3 gene exp ession i sel could also be
a ec ed by TNF-a ea men . In o de o unde s and how impo -
in-a3 gene ansc ip ion is egula ed, we sc eened he p omo e
egion o he impo in-a3 gene o he p esence o po en ial an-
sc ip ion ac o binding si es. We ound ha he p omo e egion
o he impo in-a3 gene con ains a binding si e o NF-kB, as
shown by ea lie ch oma in immunop ecipi a ion s udies (Raska-
o e al., 2012)(Figu e 2H). Consis en ly, using a epo e
cons uc encoding o enilla luci e ase unde he con ol o
he impo in-a3 p omo e , we could show ha impo in-a3
gene exp ession is di e en ially a ec ed upon TNF-a ea men
in a cell- ype- and dose-dependen manne (Figu es 2I–2K). In
clinically ele an p ima y human ai way epi helial cells as well
as in gene a ed MEFs he ein, impo in-a3 p omo e ac i i y
was highes upon low-dose TNF-a ea men , while high-dose
TNF-a ea men esul ed in a educ ion o impo in-a3 gene
exp ession (Figu es 2J and 2K). Howe e , mul i-passage
cance ous A549 cells p esen ed high esis ance o TNF-a ea -
men (Figu e 2I), sugges ing a delayed dose- esponse cu e.
T ea ing MEFs wi h o he cy okines, such as in e leukin (IL)-10
o ans o ming g ow h ac o b(TGF-b) did no signi ican ly al e
impo in-a3 p omo e ac i i y, unlike TNF-a ea men (Figu es
S2A and S2B), albei IL-10 p esen ed a sligh induc ion (Fig-
u e S2A). Nex , we wan ed o assess whe he TNF-a-ac i a ed
NF-kB is di ec ly in ol ed in egula ing impo in-a3 gene an-
sc ip ion. The e o e, we used cells de icien in he NF-kB essen-
ial modula o (NEMO), he egula o y subuni o he IKK (IkB ki-
nase), which is equi ed o NF-kB phospho yla ion and hus i s
ac i a ion. Impo in-a3 mRNA le els we e educed in TNF-
a- ea ed NEMO
/
cells compa ed o un ea ed con ols (Fig-
u e 2L). As expec ed, educed impo in-a3 mRNA le els also e-
sul ed in educed impo in-a3 p o ein exp ession in NEMO
/
cells ea ed wi h TNF-aas compa ed o un ea ed cells (Fig-
u e 2M). Reduced impo in-a3 exp ession in NEMO
/
cells
could be es o ed upon ans ec ion o a NEMO-exp essing
plasmid back in o NEMO
/
cells (Figu e 2M). Con ol ea men
o NEMO
/
cells wi h IL-6, IL-10, in e e on (IFN)-go TGF-b
showed simila ly educed impo in-a3 mRNA exp ession le els
as upon TNF-a ea men (Figu e 2N). I is impo an o no e
ha IL-6 ansc ip ion is di ec ly dependen on NF-kB, unlike
IL-10, IFN-go TGF-b. These indings suppo he concep ha
TNF-a-ac i a ed NF-kB is c ucial o impo in-a3 gene exp es-
sion. Howe e , o he cy okines migh also educe impo in-a3
exp ession a high, bu no low, cy okine concen a ions (Fig-
u e S2C). Ou indings suppo he concep ha impo in-a3
gene exp ession is con olled by TNF-a-ac i a ed NF-kBina
cell- and dose-dependen manne . This is in line wi h gene al
con ol mechanisms in cells, such as nega i e o posi i e eed-
back loops, whe e gene ansc ip ion is egula ed (e.g., ei he
by ac i a ing o blocking he p omo e , espec i ely, depending
on demand) (Figu e 2O). Howe e , o he , no ye de ined, NF-kB-
independen pa hways migh also a ec impo in-a3 gene
exp ession. Fu u e s udies a e equi ed o dissec he impo in-
a3 p omo o o ansc ip ion ac o binding si es o he han
NF-kB.
High-Le el TNF-a-Inducing Human-Type HPAIV
In ec ions Reduce Impo in-a3 P o ein Exp ession
Le els in Human Respi a o y Cells
Since he NF-kB anspo ing nuclea impo ac o impo in-a3
was he only iso o m in he lung ha al e ed i s exp ession le els
upon TNF-a ea men , we wan ed o know how impo in-a3is
egula ed unde espi a o y s ess associa ed wi h cellula cy o-
kine changes. The e o e, we used an in luenza in ec ion model.
We in ec ed MEFs wi h highly pa hogenic a ian in luenza A i us
(HPAIV) mu an s ha we e known o ac as low- o high-le el
TNF-ainduce s (Figu e 3A). Then, we in ec ed MEFs ei he
wi h he low-le el TNF-a-inducing HPAIV con aining an a ian-
ype polyme ase signa u e (SC35-PB2
701D
), common in in lu-
enza A i us (IAV) ha ing a low eplica i e i ness in mammals,
o i s high-le el TNF-a-inducing HPAIV coun e pa con aining
a human- ype polyme ase signa u e (SC35M-PB2
701N
), com-
mon in IAV possessing a high eplica i e i ness and i ulence
Figu e 2. Molecula Func ion and Regula ion o Impo in-a3
(A) Schema ic o e iew o impo in-a3-media ed NF-kB p65 nuclea ansloca ion.
(B) Pull-down o NF-kB om lysa es o un ea ed o TNF-a- ea ed A549 cells wi h Sepha ose-immobilized GST-impo in-a1 and -a3. Exp ession o GST-im-
po in-aiso o ms was e i ied wi h Coomassie s aining (uppe panel), and bound NF-kB p105, p65, and p50 we e de ec ed ia wes e n blo (lowe panel). GAPDH
p o ein de ec ion se ed as loading con ol. One ep esen a i e expe imen is shown ou o h ee independen expe imen s.
(C) E ec o TNF-a ea men on exp ession and localiza ion o NF-kB p65 and impo in-a3 in WT MEFs (n = 3).
(D and E) Nuclea ac ion (NF; con ol p84) and cy oplasmic ac ion (CF; con ol GAPDH) localiza ion o NF-kB p65 in a3
/
MEFs o WT con ols he eo (a3
+/+
)
a e TNF-a ea men using cell ac iona ion assay (D) o immuno luo escence (E) as well as quan i ica ions he eo (n = 3–5). Scale ba , 20 mm.
(F and G) A e seeding, cells we e se um s a ed o 24 h. Subsequen ly, cells we e con ol ea ed (w/o) o ea ed wi h TNF-a. Endogenous impo in-a3 and NF-
kB p65 p o ein le els we e de ec ed and quan i ied o whole-cell lysa es (WCLs), CFs and NFs o WT MEFs (n = 3) (F), and WCLs o HSAEpCs (n = 3) (G). Con ol-
ea ed samples o each ac ion we e se o 100%.
(H) NF-kB binding si es in he p omo e elemen o he human impo in-a3 gene in A549 cells (Raska o e al., 2012).
(I–K) E ec o TNF-a ea men on impo in-a3 p omo e ac i i y in HSAEpC (n = 2) (I), MEFs (n = 3) (J), and A549 cells (n = 6) (K) upon con ol (C ) o impo in-a3
p omo e epo e cons uc (a3) ans ec ion. Rela i e impo in-a3 p omo e ac i i y in con ol- ea ed samples was se o 100%.
(L) Impo in-a3 mRNA le els in mock o TNF-a- ea ed MEF NEMO
/
cells (n = 6). Rela i e exp ession alues o impo in-a3 in samples we e no malized o
GAPDH, and impo in-a1 mRNA exp ession was se o 1. Each da a poin ep esen s an indi idual sample.
(M) MEF NEMO
/
cells we e ans ec ed wi h emp y plasmid (Mock) o NEMO and un ea ed o ea ed wi h 50 ng TNF-a. Impo in-a3 and NEMO p o ein
exp ession le els we e de ec ed (n = 3). GAPDH p o ein de ec ion se ed as loading con ol.
(N) MEF NEMO
/
cells we e ea ed wi h 10 ng TNF-a, IL-6, IL-10, IFN-gand TGF-band impo in-a3 mRNA le els we e measu ed as desc ibed abo e (n = 3)
(O) Impo in-a3 gene exp ession is con olled by TNF-a-ac i a ed NF-kB in a dose-dependen manne . Low amoun s o TNF-ainduce ac i a ion o NF-kB and
high exp ession o impo in-a3 (uppe panel). In con as , high amoun s o TNF-ainduce high ac i a ion o NF-kB and low exp ession o impo in-a3 (lowe panel).
Da a shown ep esen means ±SD o a leas h ee independen biological expe imen s. *p < 0.05; **p < 0.01; ***p < 0.001.
Cell Repo s 31, 107549, Ap il 21, 2020 5
SC35-
PB2
701D
SC35M-
PB2
701N
0
500
1,000
1,500
TNF-α concen a ion [pg/ml]
TNF-α
***
A
0
50
100
150
Rela i e impo in-α3 amoun
[(α3/GAPDH) %]
SC35-
PB2701D
SC35M-
PB2701N
Mock
***
*
0
50
100
150
Rela i e impo in-α3 amoun
[(α3/NP) %]
***
SC35-
PB2701D
SC35M-
PB2701N
Mock
Mock SC35-
PB2701D
SC35M-
PB2701N
Impo in-α3
NP
GAPDH
MW
[kDa]
58
56
37
BC D
E
F
Mock
SC35M-
PB2701N
G
α1
0
50
100
150
Rela i e impo in-α amoun
[(α/GAPDH) %]
α7α5α4
Impo in
SC35M-PB2701N
L
J
Time p. i. [h]
Vi us i e [log (p. .u. • ml-1)]
024487296
0
2
4
6
8
*** *** ***
SC35-PB2701D – MOI 0.1
024 48 72 96
0
2
4
6
8
Vi us i e [log (p. .u. • ml-1)]
Time p. i. [h]
***
*** ***
SC35-PB2701D – MOI 1
Vi us i e [log (p. .u. • ml-1)]
Time p. i. [h]
0 24487296
0
2
4
6
8
*** ***
*** ***
SC35M-PB2701N – MOI 0.01
WT
α3-/-
α7-/-
Vi us i e [log (p. .u. • ml-1)]
Time p. i. [h]
024487296
0
2
4
6
8
***
***
SC35M-PB2701N – MOI 0.001
HI
WT
α3-/-
α7-/-
WT
α3-/-
α7-/-
WT
α3-/-
α7-/-
K
24 48
α3+/+
α3-/- -TNF-α
α3+/+
α3-/-
0
0
2
4
6
8
10
Vi us i e [log (p. .u. • ml-1)]
Time p. i. [h]
+TNF-α
(legend on nex page)
6Cell Repo s 31, 107549, Ap il 21, 2020
in mammalian species. Impo in-a3 p o ein le els we e s ongly
educed in human- ype HPAIV compa ed o a ian- ype HPAIV-
in ec ed MEFs (Figu es 3B–3E). In con as , p o ein exp ession
le els o o he impo in-aiso o ms (a1, a4, a5, and a7) we e
no a ec ed (Figu es 3F and 3G). Nex , we wan ed o assess
whe he eplica ion o a ian- and human- ype HPAIV is a ec ed
in cells de icien in impo in-a3. Cells de icien in impo in-a3
we e ob ained om ib oblas s o mice gene a ed o his s udy
wi h a dele ed impo in-a3 gene (a3
/
;a3
/
MEFs; Figu es
S2D and S2E). MEFs wi h a de icien impo in-a7 gene we e ob-
ained om mice wi h a dele ed impo in-a7 gene (a7
/
),
desc ibed p e iously (Gab iel e al., 2011). Replica ion kine ics
in MEFs e ealed ha a ian- ype bu no human- ype HPAIV
eplica ion is impai ed in cells wi h a dele ed impo in-a3 gene
(a3
/
), unlike in con ol cells, wi h a dele ed impo in-a7 gene
(a7
/
) ha ac s as a posi i e ac o o human- ype HPAIV epli-
ca ion in mammalian cells (Hudje z and Gab iel, 2012)(Figu es
3H and 3I). In e es ingly, a ian- ype HPAIV eplica ion in a3
/
MEFs is inc eased by >1,000 imes compa ed o WT con ol
cells. Human- ype HPAIV eplica ion was compa able be ween
WT and a3
/
cells (Figu es 3J and 3K). In line, human- ype
SC35M-PB2
701N
HPAIV eplica ion was no a ec ed in human
H1299 a3
+/+
cells upon TNF-a ea men , likely due o he s ong
inhibi ion o impo in-a3 p o ein exp ession (Figu e 3L). In human
H1299 a3
/
cells, SC35M-PB2
701N
HPAIV eplica ion sligh ly
inc eased, sugges ing some esidual inhibi o y ac i i y o
impo in-a3 in human cells (Figu e 3L). These indings show
ha low-le el TNF-a-inducing HPAIVs wi h a ian- ype polyme -
ase signa u es a e highly sensi i e o impo in-a3-media ed i us
es ic ion. In con as , high-le el TNF-a-inducing HPAIVs wi h
human- ype polyme ase genes a e esis an o impo in-a3-
media ed i us es ic ion.
Impo in-a3 Belongs o he Mos Conse ed Nuclea
T anspo Fac o s be ween A ian and Mammalian
Species
Di e en ial egula ion o impo in-a3 exp ession le els upon
a ian- and human- ype HPAIV in ec ions p omp ed us o
analyze he sequence iden i y o impo in-aiso o ms in a ious
a ian and mammalian species. Among all impo in-aiso o ms
analyzed, impo in-a1 was he mos di e se nuclea impo ac-
o , wi h amino acid homology o 83.52% be ween chicken and
humans (Figu e 4A; Table S1). The iso o ms impo in-a3and
impo in-a4 displayed he highes amino acid homologies be-
ween chicken and humans o 99.04% and 98.08%, espec-
i ely (Figu es 4B and 4C; Table S1). Impo in-a5, -a6, and
-a7 p esen ed amino acid homologies be ween chicken and
humans o 95.91%, 94.22%, and 94.4%, espec i ely (Figu es
4D–4F; Table S1). These indings sugges ha pa icula ly im-
po in-a3andimpo in-a4, bo h known nuclea anspo e s
o NF-kB(Fage lund e al., 2005), a e highly conse ed ac oss
species kingdoms. Howe e , impo in-a3 is exp essed up o
~5- imes-highe le els han impo in-a4in hemu ineand
human lung (Figu es 1C and 1D), highligh ing pa icula ly
he impo ance o impo in-a3in espi a o yimmune e-
sponses. Thus, acqui ing he abili y o down egula e impo in-
a3 in he lung migh se e as an impo an basis in a ian-
mammalian i us ansmission. In o de o see whe he
di e en ial impo in-a3 egula ion is also p esen in human iso-
la es, we analyzed human- ype HPAIV s ains ha caused
poul y ou b eaks wi h human in ec ions in he Ne he lands
(H7N7) o Thailand (H5N1). In ec ion o p ima y lung cul u es
wi h cy okine s o m, including TNF-a-inducing H5N1 and
H7N7 HPAIV isola es wi h human- ype polyme ase signa u es
(PB2
627K
;PB2
701N
), showed signi ican ly educed impo in-a3
exp ession le els compa ed o hei coun e pa s wi h a ian-
ype polyme ase signa u es (PB2
627E
;PB2
701D
)(Figu es 4G
and 4H). These indings highligh he c ucial ole o impo in-
a3 as a espi a o y ac o conse ed ac oss animal kingdoms
in HPAIV in e species ansmission.
Human-Type Polyme ase Signa u es in HPAIV Media e
Reduced Impo in-a3 mRNA Exp ession Le els in he
Mu ine BE
Nex , we wan ed o assess whe he di e en ial egula ion o
impo in-a3 can also be de ec ed in a mu ine in ec ion model.
The e o e, we assessed impo in-a3 exp ession le els in lung
samples o HPAIV-in ec ed WT mice. In ec ion wi h a ian- ype
SC35-PB2
701D
led o inc eased impo in-a3 mRNA le els in he
lung compa ed o unin ec ed con ols on day 3 pos in ec ion
(p.i.) (Figu es 5A and 5B). In con as , in ec ion wi h human-
ype SC35M-PB2
701N
esul ed in signi ican ly educed impo -
in-a3 mRNA le els (Figu e 5C). Since he human- ype SC35M
HPAIV con ains mainly wo hos adap i e signa u es (PB2
701N and NP 319K), we analyzed ecombinan i uses wi h
hese single mu a ions o dissec hei indi idual impac on he
dys egula ion o impo in-amRNA ansc ip ion. He ein, he
PB2 D701N hos adap i e mu a ion was p edominan ly espon-
sible o educed impo in-a3 ansc ip ion, while i s e ec
was u he enhanced when combined wi h he NP N319K hos
Figu e 3. HPAIV Replica ion Kine ics in Impo in-a3
/
Cells
(A) TNF-ainduc ion in MEFs in ec ed wi h a ian- ype SC35-PB2
701D
o human- ype SC35M-PB2
701N
H7N7 HPAIV (MOI = 0.1) measu ed by ELISA (n = 3).
(B–E) P o ein ex ac s he eo we e analyzed and quan i ied by wes e n blo a 48 h p.i. Impo in-a3 and i al NP p o ein we e de ec ed (B and D) and quan i ied (C
and E) wi h GAPDH (B and C) o NP adjus men (D and E) (n = 4). Dashed lines in (D) depic c opping o he ele an bands un on he same gel.
(F and G) Addi ionally, impo in-a1, -a4, -a5, and -a7 and i al NP p o ein we e de ec ed and quan i ied wi h GAPDH adjus men o con i m ha obse ed e ec s
we e speci ic o impo in-a3. Con ol- ea ed, unin ec ed samples o SC35-PB2
701D
-in ec ed samples we e se o 100%. C oss- eac i i y o impo in-aan i-
bodies wi h NP esul ed in a band (#) in (B) and (F).
(H–K) G ow h kine ics o a ian- ype SC35-PB2
701D
(H and I) and human- ype SC35M-PB2
701N
(J and K) H7N7 HPAIV in impo in-a
/
MEFs (n = 3). WT (black),
impo in-a3
/
(a3
/
, ed), and impo in-a7
/
(a7
/
, g een) MEFs we e in ec ed wi h low (H and J) o high (I and K) mul iplici ies o in ec ion (MOI). Vi us i e s
gi en in plaque- o ming uni s (p. .u.) we e de e mined by plaque assay on MDCKII cells (0, 24, 48, 72, and 96 h p.i.).
(L) H1299 a3
+/+
(black) and a3
/
( ed) cells we e in ec ed wi h human- ype SC35M-PB2
701N
HPAIV (MOI = 0.001) and eplica ion kine ics measu ed in he
p esence o absence o TNF-a(n = 3). Vi us i e s we e de e mined by plaque assay on MDCKII cells (0, 24, 48 h p.i.). Da a shown ep esen means o ela i e
impo in-aamoun s o loga i hmic i us i e s ±SD o a leas h ee independen biological expe imen s (*p ˂0.05; ***p < 0.001).
Cell Repo s 31, 107549, Ap il 21, 2020 7
adap i e signa u e (Figu es 5D–5F). Compa ed o all o he im-
po in-aiso o ms, impo in-a3 con inuously showed he mos
p ominen al e a ions. In e es ingly, he mos signi ican al e -
a ions in impo in-a3 mRNA le els we e de ec ed a 3 days p.i.
compa ed o 1 day p.i. (Figu es S2F–S2K), consis en wi h he
concep ha accumula ing cy okine le els h ough mul iple epli-
ca ion kine ics is equi ed o a p o ound cy okine-dependen
ep ession o impo in-a3 ansc ip ion. A ian- ype SC35-
PB2
701D
i us eplica ion, which caused an inc ease in impo -
in-a3 mRNA le els, was es ic ed o he espi a o y ac (Fig-
u es S3A and S3B). In con as , human- ype SC35M-PB2
701N
i al eplica ion led o a signi ican educ ion o impo in-a3
mRNA le els in he lung, inc eased i us load in he lung, and
sys emic i us sp ead (Figu e S3C). The ecombinan i uses
wi h he single o combined human- ype PB2 701N and NP
319K signa u es showed s epwise sys emic sp ead along wi h
hei abili y o educe impo in-a3 mRNA le els in he lung (Fig-
u es S3D–S3F). These indings u he highligh ha hos
0
100
200
300
400
500
48 h p.i.24 h p.i. 48 h p.i.24 h p.i.
Rela i e impo in-α3 amoun
[(α3/NP) %]
HSAEpC
H5N1 H7N7
PB2701D
PB2701N
PB2627E
PB2627K
** ***
0
200
400
600
800
1,000
1,200
Rela i e impo in-α3 amoun
[(α3/NP) %]
48 h p.i.24 h p.i. 48 h p.i.24 h p.i.
HBEpC
H5N1 H7N7
* *
*
H
Anas pla y hynchos
Gallus gallus
Mus musculus
Sus sc o a
Macaca ascicula is
Homo sapiens
Homo sapiens
Macaca ascicula is
Sus sc o a
Mus musculus
Gallus gallus
Anas pla y hynchos
Anas pla y hynchos
Gallus gallus
Mus musculus
Sus sc o a
Macaca ascicula is
Homo sapiens
Homo sapiens
Macaca ascicula is
Sus sc o a
Mus musculus
Gallus gallus
Anas pla y hynchos
Anas pla y hynchos
Gallus gallus
Mus musculus
Sus sc o a
Macaca ascicula is
Homo sapiens
Homo sapiens
Macaca ascicula is
Sus sc o a
Mus musculus
Gallus gallus
Anas pla y hynchos
85
90
95
100
iden i y
AB C
Impo in-D1Impo in-D3Impo in-D4
Anas pla y hynchos
Gallus gallus
Mus musculus
Sus sc o a
Macaca ascicula is
Homo sapiens
Homo sapiens
Macaca ascicula is
Sus sc o a
Mus musculus
Gallus gallus
Anas pla y hynchos
D
Impo in-D5
Anas pla y hynchos
Gallus gallus
Sus sc o a
Macaca ascicula is
Homo sapiens
Homo sapiens
Macaca ascicula is
Sus sc o a
Gallus gallus
Anas pla y hynchos
EImpo in-D6
G
Gallus gallus
Mus musculus
Sus sc o a
Macaca ascicula is
Homo sapiens
Homo sapiens
Macaca ascicula is
Sus sc o a
Mus musculus
Gallus gallus
85
90
95
100
iden i y
FImpo in-D7
Figu e 4. High Impo in-a3 Homology ac oss Species and I s Role in HPAIV In e species T ansmission om Bi ds o Humans
(A–F) Impo in-ahomologies ac oss animal kingdoms. Homolgies o impo in-a1 (A), -a3 (B), -a4 (C), -a5 (D), -a6 (E), and -a7 (F). Pai wise amino acid sequence
iden i ies we e calcula ed o impo in-aac oss species.
(G and H) P ima y human b onchial (HBEpC; G) o p ima y human small ai way epi helial cells (HSAEpC; H) we e in ec ed wi h clinical a al case A/Thailand/
1(KAN-1)/2004 (H5N1, MOI = 1) o A/Ne he lands/219/2003 (H7N7, MOI = 10) isola es ha bo ing he o iginal human- ype signa u es (PB2
701N
o PB2
627K
,
espec i ely) o hei ecombinan a ian- ype coun e pa s (PB2
701D
o PB2
627E
, espec i ely). Impo in-a3 was quan i ied in NP-adjus ed wes e n blo s o HBEpC
(H) and HSAEpC (I) in ec ed wi h H5N1 (MOI = 0.1) o H7N7 (MOI = 1). Impo in-a3/NP a ios o human- ype i us in ec ed samples a 24 h p.i. we e se o 100%.
Da a shown ep esen means ±SD o a leas h ee independen biological expe imen s. *p < 0.05; **p < 0.01; ***p < 0.001.
8Cell Repo s 31, 107549, Ap il 21, 2020
STAR+METHODS
KEY RESOURCES TABLE
LEAD CONTACT AND MATERIALS AVAILABILITY
Fu he in o ma ion and eques s o esou ces and eagen s should be di ec ed o and will be ul illed by he Lead Con ac , Guelsah
Gab iel ([email p o ec ed]). Ma e ials, eagen s and mouse lines o his s udy a e a ailable wi h a comple ed Ma e ial
T ans e Ag eemen .
REAGENT o RESOURCE SOURCE IDENTIFIER
An ibodies
impo in-a1 Abcam ab6036; RRID: AB_305245, ab54489;
RRID: AB_2249727
impo in-a3 Abcam ab6039; RRID: AB_1977565
impo in-a3 No us Biologicals #NB100-93345; RRID: AB_1237130,
eplaced by #NB100-81651;
RRID: AB_1146049
impo in-a4 Abcam ab6038; RRID: AB_305247
impo in-a5/7 Uni e si y o L€
ubeck Ko
¨hle e al., 1999
NF-kB p65 Cell Signaling #3987; RRID: AB_2341215
Vi us S ains
SC35-PB2
701D
Uni e si y o Ma bu g Gab iel e al., 2011
SC35M-PB2
701N
Uni e si y o Ma bu g Gab iel e al., 2011
A/Thailand/1(KAN-1)/2004 (H5N1) Uni e si y o Ma bu g Pu ha a hana e al., 2005
A/Ne he lands/219/2003 E asmus Medical Cen e Fouchie e al., 2004
Biological Samples
Human biopsy ma e ial om nasal concha
and he lung
E asmus Medical Cen e
Ro e dam, he Ne he lands
#MEC-2011-129, MEC 2008-207,
MEC 2009-128
Chemicals, Pep ides and Recombinan P o eins
Human TNF-aR&D Sys ems #210-TA
Mouse TNF-aBioLegends #575204
C i ical Comme cial Assays
Dual-Luci e ase Repo e Assay Sys em P omega E1910
Ligh Swi ch Luci e ase Assay Reagen Swi chGea Genomics LS010
Deposi ed Da a
Eu opean Nucleo ide A chi e (ENA) Hein ich Pe e Ins i u e, Hambu g h ps://www.ebi.ac.uk/ena/da a/
iew/PRJEB8023
Expe imen al Models: Cell Lines
Impo in-a3
/
MEFs Max Delb €
uck Cen e , Be lin This s udy
Impo in-a3
/
H1299 Hein ich Pe e Ins i u e, Hambu g This s udy
NEMO
/
MEFs Hein ich Pe e Ins i u e, Hambu g Mak is e al., 2000
HBEpC P omoCell C-12640
HSAEpC P omoCell C-12642
Expe imen al Models: O ganisms/S ains
Impo in-a3
/
mice Max Delb €
uck Cen e , Be lin This s udy
Recombinan DNA
pLigh Swi ch_P om-impo in-a3 Swi chGea Genomics This s udy
pLigh Swi ch_P om-con ol Swi chGea Genomics This s udy
So wa e and Algo i hms
Muscle e sion 3.8.1551 No applicable Edga , 2004
Clus al Omega e sion 1.2.4 No applicable Sie e s e al., 2011
Cell Repo s 31, 107549, Ap il 21, 2020 e1

EXPERIMENTAL MODEL AND SUBJECT DETAILS
Human Subjec s
Biopsy ma e ial om human unin ec ed URT (nasal concha) and LRT (lung, mainly consis ing o al eola issue) was ob ained om he
E asmus Medical Cen e in Ro e dam, he Ne he lands, om dono s a he E asmus Medical Cen e in Ro e dam. App o al o use
he clinical samples o scien i ic use was ob ained om he Du ch Medical E hical Commi ee unde he pe mi numbe s #MEC-
2011-129, MEC 2008-207 and MEC 2009-128. Dono sex and age is unknown.
Cells and Vi uses
Cell lines o human emb yonic kidney cells (HEK293T, ATCC), human al eola adenoca cinoma cells (A549, H1299, ATCC), a ican
g een monkey kidney cells (Ve oE6) and human ce ical adenoca cinoma cells (HeLa, ATCC) we e g own in Dulbecco’s modi ied
Eagle’s medium (DMEM, PAA) supplemen ed wi h 10% e al cal se um (FCS; PAA), 1% penicillin/s ep omycin (P/S, PAA), and
1% L-Glu amine (PAA). Mouse emb yonic ib oblas s (MEF, see desc ip ion o hei gene a ion in a sepa a e pa ag aph below)
we e g own in DMEM supplemen ed wi h 10% FCS, 1% P/S, 1% L-Glu amine, 1% sodium py u a e, and 1% non-essen ial amino
acids (NEAA, PAA). Immo alized NEMO
/
MEFs (Mak is e al., 2000) we e kindly p o ided by Michael Ka in (Uni e si y o Cali o nia,
San Diego). Madin Da by canine kidney cells (MDCKII, ATCC) we e g own in Minimal Essen ial Medium (MEM, PAA) supplemen ed
wi h 10% FCS, 1% P/S, and 1% L-Glu amine. P ima y human b onchial (HBEpC, P omoCell, C-12640) and p ima y human small
ai way epi helial cells (HSAEpC, P omoCell, C-12642) we e g own in Ai way Epi helial Cell Media o Small Ai way Epi helial Cell Media
(P omoCell), espec i ely, supplemen ed wi h 1% P/S. HBEpC cell dono was a 52 yea old emale subjec . HSAEpC cell dono was a
57 yea old emale subjec . Cells we e cul i a ed a 5% CO
2
, 96% H and 37C. Recombinan H7N7 HPAIV i uses SC35-PB2
701D
,
SC35M-PB2
701N
, he single-poin mu an s SC35-PB2
701N
and SC35-NP
319K
o he double mu an SC35-PB2
701N
-NP
319K
(Gab iel
e al., 2011) as well as ecombinan A/Thailand/1(KAN-1)/2004 (H5N1) wild- ype s ain (H5N1-PB2
701N
)(Gab iel e al., 2011; Pu ha-
a hana e al., 2005), he a ian- ype single poin mu an he eo (H5N1-PB2
701D
) and H7N7 HPAIV i us isola es A/Ne he lands/219/
2003 (H7N7-PB2
627K
, a al case) o he ecombinan i us he eo (H7N7-PB2
627E
)(Czudai-Ma wich e al., 2014; Fouchie e al., 2004)
we e used.
All expe imen s wi h H5N1 and H7N7 i uses we e conduc ed a he Hein ich Pe e Ins i u e, Leibniz Ins i u e o Expe imen al
Vi ology in Hambu g, Ge many, in biosa e y le el 3 acili ies, excep o he SC35M-PB2
701N
s ain ha was classi ied as biosa e y
le el 2 agen , and app o ed by he ele an Ge man au ho i ies (Gen echnikbeho
¨ de Hambu g).
Animals
Impo in-a3
/
mice in he C57BL/6 gene ic backg ound and WT li e ma es he eo we e b ed and housed unde speci ic pa hogen-
ee condi ions a he animal acili y o he Hein ich Pe e Ins i u e, Leibniz Ins i u e o Expe imen al Vi ology, Hambu g, Ge many. We
used 6-12 week old emale and male mice. Expe imen al g oups we e ma ched ega ding equal numbe s o emale and male mice as
well ma ched in compa able age g oups.
Animal expe imen s we e pe o med in s ic acco dance wi h he guidelines o he Ge man Animal Wel a e Regula ions. All animal
p o ocols we e app o ed by he ele an Ge man au ho i y (Beho
¨ de €
u S ad en wicklung und Umwel Hambu g, license numbe s:
29/09, 54/11). Mice we e humanely killed upon > 25% weigh loss o upon achie emen o de ined body index sco es (e.g., on phys-
ical appea ance, clinical signs, beha io , ea ing and d inking e c.) acco ding o he Ge man Animal Wel a e Regula ions.
METHOD DETAILS
T ans ec ion and Vec o s
T ans ec ions we e pe o med using Lipo ec amine 2000 (In i ogen) o je PRIME(Polyplus-T ans ec ion) acco ding o he manu-
ac u e ’s ins uc ions. Vec o cons uc s used we e pHW2000-(SC35-PB2
701D
, SC35-PB2
627K
, SC35-PB1, SC35-PA, SC35-NP,
SC35M-PB2
701N
, SC35M-PB1, SC35M-PA, SC35M-NP) (Gab iel e al., 2005), pcDNA-impo in-a1-N-FLAG, pcDNA-impo in-a3-
N-FLAG, pcDNA-impo in-a5-N-FLAG and pcDNA-impo in-a7-N-FLAG, pPol-I-NP-Luc-human (Gab iel e al., 2005), pRL-TK
(P omega), pLigh Swi ch_P om-con ol and pLigh Swi ch_P om-impo in-a3 ( enilla luci e ase epo e cons uc s; Swi chGea Ge-
nomics), pFL-TK and pCMVTAG-NEMO (Addgene). The pcDNA-impo in-a-N-FLAG cons uc s we e desc ibed p e iously (Gab iel
e al., 2011; Wang e al., 1997). pFL-TK was gene a ed by ampli ica ion o he i e ly luci e ase gene om he pPol-I-NP-luc-human
cons uc (Gab iel e al., 2005) using p ime s p5_NheI_Fluc_ o (50-GCGCTAGCCACCATGGAAGACGCCAAAAACATAAAGAA
AGGCCCGGC-30) and p3_Fluc_No I_ e (50-GCGCGGCCGCTTACAATTTGGACTTTCCGCCCTTCTTGG-30). Inse ion o he i e ly
luci e ase gene in o pRL-TK wi hou he enilla luci e ase gene was pe o med ia he NheI and No I es ic ion si es.
An ibodies
P ima y an ibodies used o wes e n blo and immunohis ochemical analyses include goa an i-impo in-a1 (Abcam, ab6036), abbi
an i-impo in-a1 (Abcam, ab54489), goa an i-impo in-a3 (Abcam, ab6039), goa an i-impo in-a4 (Abcam, ab6038), abbi an i-im-
po in-a3 (No us Biologicals, #NB100-93345, eplaced by #NB100-81651), abbi an i-impo in-a3(Ko
¨hle e al., 1997), abbi an i-
impo in-a5/a7(Ko
¨hle e al., 1999), abbi an i-NF-kB p65 (Cell Signaling, #3987), abbi an i-LSD1 (Cell Signaling, #2139), abbi
e2 Cell Repo s 31, 107549, Ap il 21, 2020
an i-GAPDH (Cell Signaling, #2118), mouse an i-nuclea ma ix p o ein p84 (Abcam, ab487), mouse an i-FLAG M2 (Sigma-Ald ich,
#F3165), abbi an i-HA (Sigma-Ald ich, #H908), mouse an i-in luenza A i us nucleop o ein (HB65; Ame ican Type Cul u e Collec-
ion, Manassas, VA) an ibodies and abbi an i-FPV se um (Gab iel e al., 2008). An i- abbi -HRP, an i-goa -HRP, an i-mouse-HRP
(Sigma-Ald ich, #A8275, #A5420, and #A4416, espec i ely), and IRDye-conjuga ed an i- abbi (Li-Co ) an ibodies we e used as sec-
onda y an ibodies o wes e n blo .
Impo in-asequence analysis
Nucleo ide sequences o he coding egions o impo in-a1, impo in-a3, impo in-a4, impo in-a5, impo in-a6 and impo in-a7 we e
downloaded om he NCBI nucleo ide da abase o Homo sapiens (GenBank: NM_001320611.1, NM_002268.5, NM_002267.4,
CR456743.1, NM_001366307.2, NM_012316.5), Macaca ascicula is (GenBank: NM_001283904.1, XM_005546256.2,
XM_005585866.2, XM_005547987.2, XM_005551660.2, XM_005544119.2), Sus sc o a (GenBank: NM_001163404.1,
XM_021069735.1, NM_001193575.1, NM_001163405.1, XM_005654393.3, XM_021095806.1), Mus musculus (GenBank:
NM_010655.3, NM_008467.4, NM_008466.5, BC006771.1, NM_008468.4), Gallus gallus (GenBank: NM_001006209.1,
NM_001007963.1, NM_001193575.1, NM_001030774.1, XM_015284586.2, NM_001012841.2) and Anas pla y hynchos (GenBank:
XM_005012314.3, XM_027464723.1, XM_027471420.1, XM_027449713.1, XM_027454663.1). Fo Mus musculus, no impo in-a6
sequence was a ailable, while he impo in-a7 sequence was missing o Anas pla y hynchos. The nucleo ide sequences we e sub-
sequen ly ansla ed in o amino acid sequences using he s anda d gene ic code. Fo each Impo in-a, he sequences we e aligned
using Muscle e sion 3.8.1551 (Edga , 2004) wi h s anda d pa ame e s. Pai wise nucleo ide and amino acid sequence iden i ies we e
hen calcula ed using Clus al Omega e sion 1.2.4 (Sie e s e al., 2011).
Gene a ion o Impo in-a3
/
Mice and Impo in-a
/
-MEFs
To gene a e an impo in-a3 a ge ing cons uc , a 900 bp-long sequence ups eam and a 5100 bp-long sequence downs eam o
exon 1 o he impo in-a3 gene we e cloned in o a a ge ing ec o desc ibed be o e (Wal he e al., 1998). A e homologous ecom-
bina ion in emb yonic s em (ES) cells, exon 1 bea ing he ansla ional s a si e o he impo in-a3 p o ein, was dele ed. ES cells we e
elec opo a ed wi h he linea ized cons uc and clones we e picked a e double selec ion wi h neomycin and gancyclo i . Posi i e
clones we e iden i ied by PCR and one o hem was chosen o blas ocys injec ion. F om he injec ed ES cell clone, ge mline chi-
me as we e ob ained and b ed wi h C57BL/6 mice. Impo in-a3 de icien mice we e backc ossed o eigh gene a ions o he
C57BL/6 gene ic backg ound; he colony was main ained by b eeding he e ozygous mice. Fo geno yping, he ollowing p ime s
we e used o PCR on genomic DNA o ail o ea biopsies: Impa3 o 5’-CCCTCAGTGTGAATTACTTCC-3’; Impa3 e 5’-GAGTC
CAAAGCCACTTCGAG-3’; Impa4neo e 5’-GCCCAGTCATAGCCGAATAG-3’. To con i m he gene dele ion, absence o impo in-
a3 mRNA and p o ein was de e mined by RT-PCR and wes e n blo o o gans. Absence o impo in-a3 mRNA wi h b-Ac in as
e e ence was e i ied using he ollowing p ime s: Impa3Ex1_ o 50-GACAACGAGAAATTGGACAACC-30; Impa3Ex7_ e
50-CGTTCCAGATGCAATGTTTGTC-30; be a-ac in_ o 50-TACAATGAGCTGCGTGTG-30; be a-Ac in_ e 50-CACAGCCTGGATGGC
TAC-30. Wild- ype (WT) and impo in-a
/
mu ine emb yonic ib oblas s (MEFs) we e p epa ed om mu ine emb yos ha es ed
on emb yonic day 13.5 om p egnan emales. To ob ain immo alized cell lines, p ima y MEFs we e passaged a leas 25 imes
acco ding o an adap ion o he 3T3 p o ocol (Toda o and G een, 1963).
Gene a ion o Impo in-a3
/
Cell Lines
To gene a e H1299 impo in-a3
/
and con ol impo in-a3
+/+
cell lines, he cells we e seeded in 12 wells and ans ec ed using Lip-
o ec amine 2000 (In i ogen) wi h ei he a con ol CRISPR-Cas9 plasmid o an impo in-a3
/
CRISPR-Cas9 plasmid (San a C uz
Bio echnology). 48 hou s pos ans ec ion posi i e cells we e so ed ia FACS in o 96 well pla es wi h help o a g een luo escen
p o ein selec ion ma ke . Clones we e es ed ia wes e n blo o impo in-a3 exp ession and posi i e clones we e cul i a ed u he
o c yop ese a ion.
Animal Expe imen s
Mice we e anaes he ized wi h ke amine-xylazine (70 mg/kg and 7 mg/kg, espec i ely) and in anasally in ec ed wi h he espec i e
i us dose in 50 ml 1x PBS. WT mice we e in ec ed wi h 6*10
4
p. .u. (100- old MLD
50
o SC35M-PB2
701N
in BALB/c mice
9
) o SC35-
PB2
701D
, SC35M-PB2
701N
, he single-poin mu an s SC35-PB2
701N
o SC35-NP
319K
o he double mu an SC35-PB2
701N
-NP
319K
.
Con ol g oups ecei ed PBS. On day 3 p.i., i e animals pe ime poin we e anaes he ized wi h an o e dose o ke amine-xylazine
(140 mg/kg and 14 mg/kg, espec i ely) and sac i iced a e in aca dial pe usion wi h DEPC- ea ed 1x PBS. Pe used o gans (lung,
achea, li e , spleen, b ain) we e emo ed and dissec ed in o pieces (< 5 mm) and subme ged in RNAla e RNA S abiliza ion Reagen
(QIAGEN). Mu ine URT samples we e ob ained by decapi a ion o he head abo e he a las. A e skinning and sagi al di ision o he
head, he issue o he nasal concha was excised and subme ged in RNAla e .
Fo ela i e mRNA-Exp ession le els de e mina ion o an i i al genes unde an NF-kB p omo e by RT-qPCR, WT (n = 5) and im-
po in-a3
/
(n = 5) mice we e in ec ed wi h 10
6
p. .u. o SC35M. Con ol g oups ecei ed PBS. On day 3 p.i., i e animals we e anaes-
he ized wi h an o e dose o ke amine-xylazine and sac i ice. Whole lungs we e emo ed and dissec ed in o pieces and subme ged
Cell Repo s 31, 107549, Ap il 21, 2020 e3
in RNAla e . O gans (lung, spleen, b ain) o unin ec ed WT mice (n = 3) o wes e n blo lysa e p epa a ion we e ob ained a e sho
inhala i e iso lu ane na cosis ollowed by ce ical disloca ion. A e dissec ion in o small pieces, he o gans we e immedia ely ozen
on d y ice.
Fo i us i e de e mina ion o o gans, WT (n = 3) and impo in-a3
/
(n = 3) mice we e in ec ed wi h 6*10
4
p. .u. o SC35-PB2
701D
(~0,1x MLD
50
o SC35-PB2
701D
in C57BL/6 mice) o a low and high dose o SC35M-PB2
701N
(10
3
p. .u.; ~0,5x MLD
50
o 6*10
4
p. .u.;
~30x MLD
50
o SC35M-PB2
701N
, espec i ely). On days 3 and 6 p.i., h ee animals pe ime poin we e sac i iced, o gans (lung,
achea, b ain) emo ed, i us i e s de e mined by plaque assay
64
, and he URT and LRT s ained immunohis ochemically agains
in luenza i us and impo in-aan igen. Fo his, mu ine o malin- ixed, pa a in-embedded (FFPE) lungs o skinned, decalci ied
FFPE heads we e hin sec ioned (3 mm). Heads we e decalci ied in 10% EDTA (pH 7,4) o 8 days, ollowed by sagi al di ision
and incuba ion in same solu ion o 3 mo e days. Mouse le hal dose 50 (MLD
50
) was assessed by in ec ing WT (n = 4-9) and impo -
in-a3
/
(n = 4-9) mice wi h se ial i us dilu ions o SC35-PB2
701D
(10
5
, 5*10
5
, and 10
6
p. .u.) o SC35M-PB2
701N
(10
2
,10
3
, and 10
4
p. .u.). Animals we e moni o ed o weigh and su i al o 14 days p.i. and he MLD
50
was calcula ed. Humane endpoin s we e
de ined acco ding o he Ge man Animal Wel a e Regula ions as desc ibed abo e.
Immunohis ochemical Analysis (IHC-P)
Vi al and impo in-a1, -a3, and -a7 an igen we e de e mined by IHC-P wi h an igen-speci ic p ima y and HRP-conjuga ed seconda y
an ibodies. URT and lungs o in ec ed WT and impo in-a3
/
mice we e excised on days 1 and 3 p.i.. FFPE hin-sec ions (3 mm) o
mu ine samples and human biopsies we e p epa ed as desc ibed (Gab iel e al., 2009). Vi al an igen was s ained using he mouse
an i-in luenza A nucleop o ein (HB65; Ame ican Type Cul u e Collec ion, Manassas, VA) o he abbi an i-FPV-se um, a bio in-con-
juga ed an i- abbi seconda y an ibody (Jackson ImmunoResea ch, #711-066-152), he A idin/Bio in Blocking Ki SP2001 (Biozol),
and he Zy oChemPlus (HRP) B oad Spec um (DAB) Ki (Zy omed) acco ding o he manu ac u e ’s ins uc ions. Fo isualiza ion o
impo in-aiso o ms in mu ine and human FFPE samples, he desc ibed abbi an i-impo in-a1, abbi an i-impo in-a3 (No us
Biologicals), and abbi an i-impo in-a5/a7 an ibodies we e used oge he wi h a bio in-conjuga ed an i- abbi seconda y an ibody
(DAKO; #E0432), a pe oxidase-conjuga ed a idin-bio in-complex- eac ion ki (DAKO), and he HRP subs a e AEC (DAKO). Coun e -
s aining was pe o med wi h hema oxylin. Since he impo in-a7 an ibody, c oss- eac s wi h impo in-a5, lung issues o mice wi h a
dele ion o ei he he impo in-a5 o he -a7 gene we e addi ionally used as con ols (da a no shown). To exclude alse-posi i e s ain-
ing due o c oss- eac i i y o he an ibodies, FFPE hin sec ions we e addi ionally s ained using impo in-aiso o m-speci ic blocking
pep ides o impo in-a1 (aa518-529, QVQDGAPGTFNF, GenSc ip ), impo in-a3 (aa509-521, NSSANVPTEGFQF, Abcam,
#ab23144), impo in-a5 (aa3-16, TPGKENFRLKSYKN, GenSc ip ) and impo in-a7 (aa3-12, MASPGKDNYR; aa526-536, PEAP-
MEGFQL, GenSc ip ) in di e en concen a ions (0.2 mg/ml, 2 mg/ml, and 20 mg/ml) (Table S5). P ima y an ibodies we e incuba ed
wi h he espec i e pep ides o 30 min p io o IHC-P. No unspeci ic s aining was obse ed wi h 2 mg/ml and 20 mg/ml o pep ide
(da a no shown). Images we e acqui ed using a NIKON AZ100 wide ield ligh mic oscope wi h a NIKON DS-Ri1 high- esolu ion mi-
c oscope came a, AZ Plan Fluo 5x and AZ Plan Apo 1x objec i es, and he NIKON Digi al Sigh DS-L2 Imaging Con olle . Image
p ocessing was pe o med using AdobePho oshopCS4 Ex ended .11.0.2.
Lase Mic obeam Mic odissec ion (LMM)
LMM was pe o med on mu ine lung c yosec ions o ob ain b onchiola epi helium and al eola issue samples using he PALM
Mic oBeam Sys em (Axio e 200 mic oscope incl. PALM CapMo e ,PALM RoboS age I,PALM RoboSo wa e 2.2.; P.A.L.M.
Mic olase Technologies GmbH/ZEISS). A e ho aco omy o WT C57BL/6 mice, lungs we e illed wi h Tissue TekCompound
(50% in 1x PBS, Saku a Fine ek) and we e shock- ozen in mel ing isopen ane. C yosec ions (10 mm) we e gene a ed a he C yos a
CM3050 S (LEICA), le d y o 5 min a 20C and s ained wi h c esyl iole (2 min 70% e hanol, 20C; 2 s 1% c esyl iole in 50%
e hanol, RT; 30 s 70% e hanol, 20C; 30 s 100% e hanol).
RNA Isola ion
To al RNA om cells o issue samples was isola ed using he innup ep RNA Miniki (Analy ik Jena). Homogeniza ion o ~50 mg o
RNAla e - ixed o gans was pe o med in 450 ml lysis bu e RL wi h 10 s e ile, s ainless s eel beads (Ø 2 mm, #22.455.0010, Re sch)
a 30 Hz and 4C o 6 min in he mixe mill MM400 (Re sch). RNase ee DNase Se (QIAGEN) was used o on-column DNase I- ea -
men . A e elu ion in RNase- ee wa e , 1 U/mlRibolock RNase Inhibi o (The mo Scien i ic) was added. RNA concen a ion and pu i y
we e de e mined using he Nanod op 1000 (Peqlab). Mic odissec ed mu ine issue was o exed o 1 min in lysis bu e RL, incu-
ba ed o 30 min a RT and o exed again (1 min). A e sho cen i uga ion, he lysa e was s o ed a 80C o e nigh . A e hawing,
1 min o o exing, and cen i uging (8 s), he sample was mixed en imes wi h 700 ml o 70% e hanol using a needle and sy inge. RNA
was isola ed acco ding o he innup ep RNA Miniki ’s ins uc ions. To al RNA om human FFPE URT was isola ed om 1-2 sec ions
(20 mM) a e xylene depa a iniza ion using he QIAGEN RNeasyFFPE Ki (QIAGEN). Isola ed RNA was ei he p ocessed
immedia ely o s o ed a 80C.
cDNA Gene a ion
cDNA om issues o cells was gene a ed using andom p ime s and he Supe Sc ip III Re e se T ansc ip ase (In i ogen) acco ding
o he manu ac u e ’s ins uc ions. Random noname p ime (Gene Link, pd(N)9, 26-4000-06, inal concen a ion: 5 mM) and dNTP
e4 Cell Repo s 31, 107549, Ap il 21, 2020
mix (QIAGEN, inal concen a ion: 500 mM each dNTP) we e incuba ed o 5 min a 65C wi h 5 mg o o al RNA in a inal olume o 13 ml
and cooled o a leas 1 min on ice. Then, 1x Fi s S and Bu e , DTT ( inal concen a ion: 5 mM), RiboLock RNase Inhibi o (The mo
Scien i ic, inal concen a ion: 2 U/ml), and 10 U/ml Supe Sc ip III Re e se T ansc ip ase we e added ( inal eac ion olume: 20 ml).
cDNA was gene a ed using he GeneAmpPCR Sys em 9700 (Applied Biosys ems; cycle: 25C o 5 min, 50C o 60 min, 70C
o 15 min, 4C o N).
cDNA om mic odissec ed samples was gene a ed using he Re e Aid H Minus Re e se T ansc ip ase (The mo Scien i ic). 2 mlo
o al RNA we e ea ed wi h 1 U/ml DNase I and 1x DNase I bu e (Sigma-Ald ich, #AMPD1) o 15 min a RT ( eac ion olume: 10 ml).
DNA es ic ion was s opped by adding 1.5 ml 50 mM EDTA and incuba ion o 10 min a 70C. Then, 1x Reac ion Bu e , DTT ( inal
concen a ion: 500 mM), dNTP-Mix (Peqlab; inal concen a ion: 500 mM each dNTP), 250 ng andom hexame p ime (pd(N)6,
Eu o ins MWG Ope on), and 100 U Re e Aid H Minus Re e se T ansc ip ase we e added ( inal eac ion olume: 20 ml). cDNA
was gene a ed using he cycle 42C o 50 min, 70C o 15 min, and 4C o N.
Fo cDNA om human FFPE URT, 250 ng o o al RNA we e incuba ed wi h Random P ime s (P omega, 500 mg/ml, # C1181; inal
concen a ion: 500 ng) and dNTP mix ( inal concen a ion: 500 mM each dNTP) o 5 min a 65C in a olume o 13 ml. A e cooling (a
leas 1 min on ice), 1x Fi s S and Bu e , DTT ( inal concen a ion: 5 mM), RNasinRibonuclease Inhibi o (P omega, #N251B; inal
concen a ion: 2 U/ml) and 10 U/ml Supe Sc ip III Re e se T ansc ip ase we e added ( inal eac ion olume: 20 ml). cDNA was gene -
a ed using he ollowing cycle: 25C o 5 min, 50C o 45 min, 70C o 15 min, and 4C o N. All cDNA samples we e p ocessed
immedia ely o s o ed a 20C.
Design o RT-qPCR P ime s
DNA oligonucleo ides (Sigma-Ald ich; Table S6 o RT-qPCR we e designed using Clone Manage 9 P o essional Edi ion o P ime -
BLAST (h ps://www.ncbi.nlm.nih.go / ools/p ime -blas /index.cgi).
Real-Time Quan i a i e PCR (RT-qPCR)
Gene o in e es (GoI) mRNA exp ession le els we e de e mined using speci ic p ime pai s (Table S6). Singleplex eac ions (10 ml)
we e se up manually in Ul aPu e DNase/RNase-F ee Dis illed Wa e (GIBCO) in Mic oAmpOp ical 96-Well Reac ion Pla es (In i-
ogen, #4306737): 5 ml Pla inumSYBRG een qPCR Supe Mix-UDG (2x, In i ogen), 0.02 ml ROX Re e ence Dye (25 mM, In i o-
gen, inal concen a ion: 50 nM), 300 nM o o wa d and e e se p ime each, and 1 ml cDNA empla e. RT-qPCR uns we e conduc ed
on he ABI 7500 Fas Sys em (Applied Biosys ems) in he S anda d 7500 mode wi h endpoin luo escence de ec ion: 2 min a 50C,
3 min a 95C, 50 ampli ica ion cycles (15 s a 95C, 10 s a 65C, and 30 s a 72C). Analysis was pe o med in iplica e o quad u-
plica e o each GOI and GAPDH in each sample. Subsequen ly, mel ing cu e analysis was pe o med on he ABI 7500 Fas Sys em
(15 s a 95C, 1 min a 60C, 15 s a 95C). Co ec amplicon size was checked by aga ose gel elec opho esis. Da a o eac ions wi h
alse p oduc s was excluded om da a analyses. Rela i e exp ession alues we e de e mined using he E
-DDCT
-me hod (Ramake s
e al., 2003; Ruij e e al., 2009). The Rn- alues we e expo ed om he SDS So wa e 1.3.1 (Applied Biosys ems) o Mic oso O ice
Excel 2007 and N
0
- alues o he s a ing concen a ion o he ansc ip in he o iginal sample we e ob ained wi h LinReg PCR So -
wa e 11.1 (Ruij e e al., 2009). The a e aged N
0
- alue o he GOI (e.g., N
0 (impo in-a3)
, n = 3–4 echnical eplica es) was hen no mal-
ized wi h he a e aged N
0
- alue o GAPDH (N
0 (GAPDH)
) o he espec i e sample. The ela i e N
0(GOI)
/N
0 (GAPDH)
-exp ession alues o
he biological eplica es a e p esen ed. Fo easie compa ison o di e en ial exp ession, he a e aged ela i e N
0
- alue was se 1.
Rela i e mRNA exp ession le els o NF-kB egula ed genes we e de e mined using speci ic p ime pai s (Table S6) o he genes
o in e es (GOIs) and he e e ence gene HPRT. Singleplex eac ions (20 ml) we e se up manually in H
2
O PCR g ade (ROCHE)
in Ligh Cycle 480 Mul iwell Pla e 96 (ROCHE, #04729692001): 10 ml Fas S a Essen ial DNA G een Mas e (2x, ROCHE),
300 nM o o wa d and e e se p ime each, and 2 ml cDNA empla e. RT-qPCR uns we e conduc ed on he Ligh Cycle 96
Real-Time PCR Sys em (ROCHE) wi h endpoin luo escence de ec ion: 10 min a 95C, 50 ampli ica ion cycles (15 s a 95C,
10 s a 65C, and 20 s a 72C). Analysis was pe o med in iplica e o each GOI and HPRT in each sample. Subsequen ly, mel ing
cu e analysis was pe o med (15 s a 95C, 15 s a 60C, 1 s a 95C). Co ec amplicon size was checked by aga ose gel elec o-
pho esis. The ela i e N
0(GOI)
/N
0 (HPRT)
-exp ession alues o he biological eplica es a e p esen ed.
P o ein Ex ac ion and Pu i ica ion
Whole cell lysa es (WCL) o cells we e ob ained by lysis in HEPES lysis bu e (50 mM HEPES (pH 8.0), 200 mM NaCl, 0.5% IGEPAL,
25% glyce ol, 1 mM PMSF, 0,07 ml/ml b-me cap oe hanol, 1x HALT P o ease and Phospha ase Inhibi o Cock ail, 1x EDTA Solu ion)
o samples we e ac iona ed in o cy oplasmic (CF) and nuclea ac ions (NF) using he NE-PER Nuclea and Cy oplasmic Ex ac ion
Reagen s (The mo Fishe Scien i ic Inc.). Whole o gan lysa es o unin ec ed WT mice we e ob ained by weighing ~30 mg o issue in o
O- ing ubes illed wi h 8 s ainless s eel beads (Ø 2 mm, #22.455.0010, Re sch). A e washing he o gan pieces once wi h 800 ml1x
PBS and cen i uga ion a 500 xg and 4C o 5 min, he supe na an was disca ded and 10 ml HEPES lysis bu e pe mg o gan we e
added. Homogeniza ion was ca ied ou a 4C and 20 Hz o 2 min ollowed by 30 Hz o 1 min in he mixe mill MM400 (Re sch). The
homogena es we e hen incuba ed o 5 min on ice and subsequen ly cen i uged o 20 min a 21,000 xg and 4C. The supe na an
was ans e ed in o a esh ube, mixed wi h 4x p o ein loading dye and boiled o 5 min a 95C be o e being subjec ed o wes e n
blo analysis. Whole cell lysa es o HEK293T cells ans ec ed wi h 10 mg o each o he pcDNA-impo in-a-N-FLAG cons uc s we e
subjec ed o a ini y pu i ica ion using he EZ iew Red ANTI-FLAG M2 a ini y gel (Sigma-Ald ich) and elu ed using a 3x FLAG pep ide
Cell Repo s 31, 107549, Ap il 21, 2020 e5
(Sigma-Ald ich) acco ding o he manu ac u e ’s ins uc ions. Thus, pu i ied FLAG- agged impo in-aiso o ms we e ob ained and
used o gene a e s anda d cu es o de e mine he a ini ies o he di e en iso o m-speci ic impo in-aan ibodies. The p o ein
amoun s o each iso o m we e quan i ied in mu ine o gan homogena es using wes e n blo analyses and s anda dized using he in-
di idual s anda d cu es as desc ibed p e iously (Ko
¨hle e al., 2002, 1999).
Wes e n Blo
Success ul knockdown o impo in-aiso o ms was con i med by wes e n blo analysis using he desc ibed goa an i-impo in-a1,
goa an i-impo in-a3, goa an i-impo in-a4, and abbi an i-impo in-a5/a7 an ibodies, espec i ely, wi h no maliza ion o GAPDH
using he abbi an i-GAPDH an ibody. The same an ibodies we e used o de ec ion and quan i ica ion o endogenous impo in-a
iso o ms in cell o o gan lysa es. Fo de ec ion and quan i ica ion o FLAG- agged impo in-aiso o ms, he mouse an i-FLAG M2 an i-
body was used. Amoun s o i al NP p o ein o NF-kB p65 in cell lysa es a e i al in ec ion o TNF-a ea men we e quan i ied using
he abbi an i-FPV se um and he abbi an i-NF-kB p65 an ibodies, espec i ely. Fo de ec ion o HA- agged NEMO, he abbi an i-
HA an ibody was used. Equal amoun o p o ein was loaded o each sample using GAPDH, p84 o LSD1 as loading con ol. Quan-
i ica ion o p o ein was pe o med wi h he Bioimage Image Quan LAS 4000 a non-sa u a ed le els wi h GAPDH o LSD1 adjus -
men o WCL, GAPDH adjus men o CF, and LSD1 o p84 adjus men o NF. Al e na i ely, NP le els we e adjus ed and used o
no maliza ion.
Immuno luo escence
WT MEFs as well as MEFs de icien o impo in-a3 we e seeded in glass bo om dishes (#81218-200, Ibidi GmbH) using egula
g ow h medium (10% FBS, 1% P/S, 1% L-Glu amine, 1% NEAA, 1% sodium py u a e). A e 8h, he g ow h medium was eplaced
wi h s a a ion medium (se um educed o 0.5% FBS) and cells we e s a ed o ano he 16h. Subsequen ly, cells we e ei he con ol-
ea ed o ea ed wi h mu ine TNF (10 ng/ml; #575204, BioLegend, Inc) o 15 min a 37C. Then, immuno luo escence s aining was
pe o med. All washing s eps we e pe o med wi h phospha e bu e ed saline (PBS), and all incuba ion s eps we e pe o med a oom
empe a u e, i no o he wise s a ed. B ie ly, cells we e washed and ixed wi h 4% pa a o maldehyde (in PBS) o 10 min a 37C.
Fixed cells we e washed wice and pe meabilized wi h 0.1% T i on X-100 (in PBS) o 20 min. A e washing, cells we e blocked
wi h 3% bo ine se um albumin (BSA) solu ion (in PBS) and hen incuba ed wi h a p ima y an ibody ecognizing mu ine NF-kB p65
subuni (1:300 in 3% BSA/PBS; #8242S, Cell Signaling) o 1h. Cells we e washed again and incuba ed o 1h wi h a donkey an i- ab-
bi IgG an ibody coupled o Alexa-555 dye (1:500 in 3% BSA/PBS; A31572, Li e Technologies GmbH). This s ep also included s ain-
ing o nuclei wi h Hoechs dye (1:1000 in 3% BSA/PBS; #62249, Li e Technologies GmbH). A e u he washing, s ained cells we e
s o ed in PBS a 4C. Images we e acqui ed on a con ocal Nikon Eclipse Ti-E spinning disc mic oscope (Nikon, Japan) a 100- old
magni ica ion. Image p ocessing was pe o med using Nikon NIS-Elemen s Ad anced Resea ch ( e sion 4.51; Nikon, Japan) and
ImageJ so wa e (Na ional Ins i u e o Heal h, NIH). Semi-au oma ic quan i ica ion o nuclea and cy oplasmic NF-kB p65 signals
was ca ied ou using a cus om-designed plugin o he ImageJ so wa e which is a ailable on eques . Da a a e based on h ee in-
dependen biological eplica es, each pe o med wi h echnical duplica es.
HPAIV G ow h Kine ics
WT MEFs and MEFs wi h dele ed impo in-a3o -a7 genes we e in ec ed wi h di e en mul iplici ies o in ec ion (MOI) o SC35-
PB2
701D
(H7N7, MOI 1 o 0.1) o SC35M-PB2
701N
i us (H7N7, MOI 0.01 o 0.001). 7*10
5
cells we e inocula ed wi h 2 mL i us dilu ion
in DMEM (con aining 1% P/S, 1% L-Glu amine, 1% NEAA, 1% sodium py u a e) and incuba ed o 30 min a 5% CO
2
, 96% H and
37C. A e washing wice wi h acidic incuba ion medium (DMEM, pH 5.0, con aining 1% FCS, 1% P/S, 1% L-Glu amine, 1% NEAA,
1% sodium py u a e), 2 mL o pH-neu al incuba ion medium we e added and he in ec ed cells we e incuba ed a 37C. Vi us i e s
o supe na an s aken a 0, 24, 48, 72, and 96 hou s pos in ec ion (p.i.) we e de e mined as plaque o ming uni s pe ml (p. .u./ml) by
plaque assay
17
on MDCKII. Fo de e mina ion o impo in-ap o ein le els a e i al in ec ion, WT MEFs we e in ec ed wi h SC35-
PB2
701D
o SC35M-PB2
701N
i us a an MOI o 0.1. o WT MEF. Simila ly, H1299 WT and H1299 wi h dele ed impo in-a3 gene
we e ea ed wi h TNFa(10ng/ml) 24h p io o inocula ed wi h SC35M-PB2
701N
a MOI 0.001 (see abo e). TNFa ea men was e-
applied and supe na an aken a 0, 24 and 48h p.i. Vi us i e s we e de e mined by plaque assay on MDCK cells and gi en in
p. .u/ml. Addi ionally, p ima y human lung cells (HBEpC and HSAEpC) we e in ec ed wi h H5N1-PB2
701N
and H5N1-PB2
701D
(MOI = 0.1 and 1) o H7N7-PB2
627K
and H7N7-PB2
627E
(MOI = 1 and 10). Cells we e lysed 24 and/o 48 hou s p.i. in lysis bu e
(50 mM HEPES (pH 8.0), 200 mM NaCl, 0.5% IGEPAL, 25% glyce ol (In i ogen), 1 mM PMSF, 0,07 ml/ml b-me cap oe hanol, 1x
HALT P o ease and Phospha ase Inhibi o Cock ail, 1x EDTA Solu ion (bo h: 100x, Pie ce/The mo Scien i ic). Whole cell p o ein ex-
ac s we e subjec ed o wes e n blo analysis.
Lib a y P epa a ion and Nex Gene a ion Sequencing
Lungs o WT and a3
/
mice in ec ed wi h 6*10
4
p. .u. o SC35-PB2
701D
o SC35M-PB2
701N
(H7N7) we e subjec ed o gene exp es-
sion analysis. Con ol mice ecei ed PBS. On day 3 p.i., whole lungs we e pe used, ha es ed and subme ged in RNAla e and o al
RNA was isola ed using he innup ep RNA Mini Ki (Analy ik Jena). RNA samples om h ee animals pe mouse s ain and i us we e
pooled and used o gene exp ession analysis. Fo each sample, 1mg o o al RNA was used o gene a ing sequencing lib a ies
wi h he Illumina T uSeq RNA Sample P epa a ion Ki 2 as ecommended by he manu ac u e (Illumina Inc.). Size and quali y o
e6 Cell Repo s 31, 107549, Ap il 21, 2020

he lib a ies we e isualized on a BioAnalyze High Sensi i i y DNA Chip (High Sensi i i y DNA Ki , Agilen Technologies). Dilu ed li-
b a ies (2 nM) we e mul iplex-sequenced on he Illumina HiSeq 2500 ins umen . Fo each sample be ween 78 and 113M, pai ed-end
eads o 101 bp leng h we e gene a ed. The eads we e aligned o he mu ine e e ence ansc ip ome (UCSC mm10) using Bow ie2
( 2.2.2) (Langmead and Salzbe g, 2012). DESeq (Ande s and Hube , 2010) was employed o assess di e en ial exp ession based on
ead coun s pe gene. Genes we e hie a chically clus e ed acco ding o hei exp ession p o iles using he R unc ion hclus (h p://
www.R-p ojec .o g/). DAVID ( 6.7) (Huang e al., 2009) was employed o unc ionally anno a e he esul ing clus e s. Ne wo ks o p o-
ein-p o ein in e ac ions we e compu ed and isualized using Cy oscape (Shannon e al., 2003).
TNF-aELISA
TNF-ap o ein amoun was de e mined in lung homogena e supe na an s o WT mice in ec ed in anasally wi h 6*10
4
p. .u. o SC35-
PB2
701D
o SC35M-PB2
701N
i uses (H7N7). Whole lungs we e ha es ed on day 1 p.i. and ~50–100 mg o issue homogenized wi h
~300 ml glass beads (RETSCH #22.222.0003) in 1 mL 1x PBS (PAA) in he mixe mill MM400 (Re sch; 20 Hz, 4C, 10 min). A e cen i-
uga ion (6000 xg, 4C, 5 min), lung homogena e supe na an s we e s o ed a 80C. Supe na an s o h ee animals pe i us we e
pooled. Then, he Enzyme-linked Immunoso ben Assay (ELISA) Ki o mu ine TNF-awas pe o med acco ding o he manu ac-
u e ’s ins uc ions (#E90133Mu, USCN Li e Science Inc.).
TNF-aT ea men , Impo in-a3 mRNA Exp ession, Subcellula Localiza ion and P omo e Ac i a ion
HSAEpC, WT MEFs and NEMO
/
MEFs we e seeded in 24-well pla es un il con luen o in 10 cm-dishes o 24 h, espec i ely, and
se um s a ed o 24 h. Then, cells we e con ol- ea ed (w/o) o ea ed wi h human TNF-a(50 ng/ml; #210-TA, R&D Sys ems) o 6 h
o wi h mu ine TNF-a(50 ng/ml; #575204, BioLegend, Inc.) o 6 h o 24 h, espec i ely. WCL, CF, and NF we e subjec ed o wes e n
blo analysis o impo in-a3 and NF-kB p65 de ec ion. To al RNA we e isola ed 6h pos ea men o ela i e impo in-a3 exp ession,
using RT-qPCR.
Fo de e mina ion o he ela i e impo in-a3 mRNA exp ession le els a e cy okine ea men , NEMO
/
MEFs we e seeded in
10cm cell cul u e dishes (5x10
6
cells pe dish). A e se um s a a ion o 24 h, cells we e con ol- ea ed (mock) o ea ed 10 min
wi h 10ng/ml o mu ine TNF-a, IL-6, IL-10, IFN-gand TGF-b. A e 6h pos ea men , cells we e ha es ed, RNA isola ed and
cDNA gene a ed be o e RT-qPCR analysis.
Fo he de e mina ion o nuclea impo o NF-kB p65 a e TNF-as imula ion, WT MEF and Impo in a3 KO MEF we e seeded in
10cm cell cul u e dishes (1x10
6
cells pe dish). A e se um s a a ion o 18 h, cells we e ea ed wi h 10ng/ml TNF-a. Subsequen ly,
cells we e ha es ed by sc aping in ice-cold PBS, cen i uged and a e emo ing he supe na an , he cells we e esuspended in 50ml
o Bu e A (10mM T is pH7,9; 1,5mM MgCl
2
; 10mM KCl). A e adding 6ml o 2% NP-40 and o exing, he samples we e incuba ed
o 10min on ice and cen i uged €
u 15min a 6,0000 pm a 4C. The supe na an (cy osolic ac ion) was placed in a esh ube, sub-
sequen ly he pelle was washed again wi h 100ml o bu e A, cen i uged again o 15min and he supe na an was emo ed. Fo
nuclea ac ions, he pelle was esuspended in 50 ml o bu e B (20 mM T is pH 7.9; 10mM KCl; 1.5mM MgCl
2
; 20% Glyce ol)
and o exed; hen 33 ml o bu e C (20mM T is pH 7.9; 1.2M KCl; 1.5 mM MgCl
2
; 20% Glyce ol) was added and he samples
we e incuba ed on a wheel a 4C o 45 min. A e wa d, samples we e cen i uged o 30 min a 12,000 pm a 4C and he supe -
na an (nuclea ac ion) was ans e ed o a esh eac ion ube. Bo h ac ions we e mixed wi h 4x p o ein loading dye and boiled o
5 min a 95C be o e being subjec ed o wes e n blo analysis.
Fo de e mina ion o impo in-a3 p omo e ac i a ion in MEF cells, cells we e seeded in 96-well pla es (5000 cells/well) o each
app oxima ely 60% con luency a e 24h. Then, cells we e ans ec ed using FuGene HD ans ec ion eagen (Swi chGea Geno-
mics), ei he wi h pLigh Swi ch-P om-Con ol (Swi chGea Genomics; 100 ng/well) as a con ol, o pLigh Swi ch-P om-impo in-
a3 (Swi chGea Genomics; 100 ng/well) o impo in-a3 p omo e ac i i y eadou . A e 24h, ans ec ed cells we e ea ed wi h mu-
ine TNF-a, IL-10 o TGF-ba 0, 1 o 50 ng/ml, o 15 min. P omo e ac i i y was measu ed a 3h pos ea men using Ligh Swi ch
TM
Luci e ase Assay Reagen (Swi chGea Genomics, LS010) acco ding o manu ac u e ’s ins uc ions.
Fo de e mina ion o impo in-a3 p omo e ac i a ion, A549 o HSAEpC we e seeded in 96-well pla es o 24 h o un il 80% con-
luency, espec i ely, and ans ec ed wi h pLigh Swi ch_P om-con ol (50 ng; Con ol) o pLigh Swi ch_P om-impo in-a3 (50 ng;
a3) and pFL-TK (250 ng; i e ly luci e ase ans ec ion con ol) using Lipo ec amine-2000 o je PRIME, espec i ely. 14 h a e
ans ec ion, se um s a a ion was ini ia ed o 30 h. Subsequen ly, cells we e ea ed wi h 0, 10, 50 o 100 ng/ml (A549) o wi h 0,
1, 10 o 100 ng/ml (HSAEpC) o human TNF-aand lysed 20 h la e . Fi e ly and Renilla luci e ase ac i i ies we e assessed using
he Dual-Luci e aseRepo e Assay Sys em (P omega, E1910).
NEMO Complemen a ion Assay
NEMO
/
MEFs we e ans ec ed in suspension wi h 5 mg emp y ec o pcDNA3.1 o pcDNA-HA-human NEMO in 6-well pla es using
Lipo ec amine 2000 (In i ogen) o 6 h. A 48 h pos ans ec ion, cells we e se um s a ed o 24 h. Then, cells we e con ol- ea ed
(w/o) o ea ed wi h mu ine TNF-a(50 ng/ml; #575204, BioLegend, Inc.) o 6 h. WCL we e subjec ed o wes e n blo analysis o
impo in-a3 and NEMO de ec ion.
Cell Repo s 31, 107549, Ap il 21, 2020 e7
QUANTIFICATION AND STATISTICAL ANALYSIS
P alues o RT-qPCR da a we e ob ained by K uskal-Wallis one-way ANOVA using PAST .2.17c (h p:// olk.uio.no/ohamme /pas /).
RT-qPCR da a we e conside ed as ele an when mo e han ±1.5- old change was de ec ed in ela i e mRNA exp ession le els.
S a is ical signi icance o animal su i al a es was de e mined wi h G aphPad P ism 5 .5.03 (G aphpad So wa e, Inc.) using he
Log Rank (Man el-Cox) es and he Kaplan-Meie su i al cu es. All o he mean alues, s anda d de ia ions (SD), and P alues
we e calcula ed wi h G aphPad P ism 5 ( 5.03) o he G aphPad So wa e, Inc. ool (h ps://www.g aphpad.com/quickcalcs/
es 1/?Fo ma =SD) using he unpai ed, wo- ailed S uden ’s es o he one-sample, wo- ailed S uden ’s es when all alues
in one g oup we e iden ical, e.g., se 100%. S a is ical signi icance was de ined as p < 0.05 (* p < 0.05, ** p < 0.01, *** p < 0.001).
DATA AND CODE AVAILABILITY
T ansc ip ome sequence da a epo ed in his publica ion is submi ed o he Eu opean Nucleo ide A chi e (ENA) and is a ailable
unde ollowing link: h ps://www.ebi.ac.uk/ena/da a/ iew/PRJEB8023.
e8 Cell Repo s 31, 107549, Ap il 21, 2020