ORIGINAL RESEARCH
published: 24 Ma ch 2015
doi: 10.3389/ micb.2015.00220
Edi ed by:
Masayuki Saijo,
Na ional Ins i u e o In ec ious
Diseases, Japan
Re iewed by:
Daniel C. Pe ea ,
Vena oRx Pha maceu icals
Inco po a ed, USA
Hi oaki Ka iwa,
Hokkaido Uni e si y, Japan
*Co espondence:
Ma co Goeijenbie ,
Depa men o Vi oscience, E asmus
MC, d Molenwa e plein 50,
Ro e dam 3015CJ, Ne he lands
m.goeijenbie @e asmusmc.nl
†These au ho s ha e sha e senio
au ho ship.
Special y sec ion:
This a icle was submi ed o Vi ology,
a sec ion o he jou nal F on ie s in
Mic obiology
Recei ed: 15 Janua y 2015
Accep ed: 05 Ma ch 2015
Published: 24 Ma ch 2015
Ci a ion:
Goeijenbie M, Meije s JCM, An asa F,
Roose JM, an de Weg CAM,
Bakh ia i K, Hen onen H, Vahe i A,
Os e haus ADME, an Go p ECM,
Ma ina BEE (2015) E ec o Puumala
han a i us in ec ion on human
umbilical ein endo helial cell
hemos a ic unc ion: pla ele
in e ac ions, inc eased issue ac o
exp ession and ib inolysis egula o
elease.
F on . Mic obiol. 6:220.
doi: 10.3389/ micb.2015.00220
E ec o Puumala han a i us
in ec ion on human umbilical ein
endo helial cell hemos a ic unc ion:
pla ele in e ac ions, inc eased
issue ac o exp ession and
ib inolysis egula o elease
Ma co Goeijenbie 1*,Joos C.M.Meije s
2,3,Fa ih An asa1,4, Je oen M. Roose1,5,
Co nelia A. M. an de Weg1,Kam an Bakh ia i2,3,Heikki Hen onen6,An i Vahe i7,
Albe D. M. E. Os e haus1,5 ,E ic C. M. an Go p1†and By onE.E.Ma ina
1,5†
1Depa men o Vi oscience, E asmus MC, Ro e dam, Ne he lands, 2Depa men o Expe imen al Vascula Medicine,
Academic Medical Cen e , Uni e si y o Ams e dam, Ams e dam, Ne he lands, 3Depa men o Plasma P o eins, Sanquin
Resea ch, Ams e dam, Ne he lands, 4Depa men o In e nal Medicine, Facul y o Medicine, Uni e si as Indonesia, Jaka a,
Indonesia, 5A emis One Heal h Ins i u e, U ech , Ne he lands, 6Me la, Finnish Fo es Resea ch Ins i u e, Van aa, Finland,
7Depa men o Vi ology, Haa man Ins i u e, Uni e si y O Helsinki, Helsinki, Finland
Puumala i us (PUUV) in ec ion causes o e 5000 cases o hemo hagic e e in
Eu ope annually and can in luence he hemos a ic balance ex ensi ely. In ec ion migh
lead o hemo hage, while a ecen s udy showed an inc eased isk o myoca dial
in a c ion du ing o sho ly a e PUUV in ec ion. The mechanism by which his
han a i us in luences he coagula ion sys em emains unknown. The e o e we aimed o
elucida e mechanisms explaining al e a ions seen in p ima y and seconda y hemos asis
du ing PUUV in ec ion. By using low passage PUUV isola es o in ec p ima y human
umbilical ein endo helial cells (HUVECs) we we e able o show al e a ions in he
egula ion o p ima y- and seconda y hemos asis and in he elease o ib inolysis
egula o s. Ou main inding was an ac i a ion o seconda y hemos asis due o
inc eased issue ac o (TF) exp ession leading o inc eased h ombin gene a ion in
a unc ional assay. Fu he mo e, we showed ha du ing in ec ion pla ele s adhe ed
o HUVEC and subsequen ly speci ically o PUUV i us pa icles. In ec ion o HUVEC
wi h PUUV did no esul in inc eased on Willeb and ac o while hey p oduced
mo e plasminogen ac i a o inhibi o ype-1 (PAI-1) compa ed o con ols. The PAI-1
p oduced in his model o med complexes wi h i onec in. This is he i s epo ha
e eals a po en ial mechanism behind he p o-coagulan changes in PUUV pa ien s,
which could be he esul o inc eased h ombin gene a ion due o an inc eased
TF exp ession on endo helial cells du ing in ec ion. Fu he mo e, we p o ide insigh
in o he con ibu ion o endo helial cell esponses ega ding hemos asis in PUUV
pa hogenesis.
Keywo ds: han a i us, hemos asis, pla ele s, endo helial cells, h ombin gene a ion, hemo hagic e e wi h enal
synd ome, HFRS
F on ie s in Mic obiology | www. on ie sin.o g 1Ma ch 2015 | Volume 6 | A icle 220
Goeijenbie e al. Puumala and endo helial hemos a ic unc ion
In oduc ion
Puumala i us (PUUV), a han a i us ca ied by ch onically
in ec ed bank oles, is he causa i e agen o an es ima ed 5000
cases yea ly o i al hemo hagic e e in Eu ope (Vapalah i
e al., 2003;Vahe i e al., 2013a). Han a i uses a e oden -bo ne,
nega i e s anded, RNA i uses belonging o he Bunya i idae
amily, which may cause wo ypes o disease in humans
(Goeijenbie e al., 2013). In Eu ope and Asia, han a i us in ec-
ion causes hemo hagic e e wi h enal synd ome (HFRS),
cha ac e ized by enal ailu e and bleeding complica ions. In
No h and Sou h Ame ica, han a i us in ec ion causes he
han a i us ca diopulmona y synd ome (HCPS) whe e pa ien s
p esen wi h se e e acu e espi a o y dis ess (Sa gianou e al.,
2012). Changing ecological ac o s de e mine fluc ua ions in
han a i us epidemiology esul ing in sudden inc eases in inci-
dence, o ins ance h ough inc eased ood a ailabili y, p olonged
i us su i al and dec eased biodi e si y (Reusken and Heyman,
2013). Recen epidemiological s udies epo ed an o e all inci-
dence inc ease o PUUV in ec ions in Eu ope (Heyman and
Vahe i, 2008).
Al hough PUUV in ec ions ha e a low case a ali y a e (<1%)
and in li e a u e he i us is o en desc ibed as he leas i -
ulen o he pa hogenic i uses wi hin he han a i us genus,
PUUV in ec ions can cause se e e disease in heal hy adul s,
which may equi e a long eco e y pe iod las ing up o 1 yea
(Schmaljohn and Hjelle, 1997). Fu he mo e, se e al epo s
desc ibed cases wi h se e e (hemo hagic) complica ions like
pi ui a y gland hemo hage, hema emesis and gas o-in es inal
bleedings (Ecke le e al., 2012;An onen e al., 2013). In con-
as o hese bleeding complica ions, a ecen s udy om Sweden
epo ed inc eased isk o acu e myoca dial in a c ion sho ly
a e PUUV in ec ion (Connolly-Ande sen e al., 2014). Gi en
he high incidence in No he n Eu ope, acu e myoca dial in a c-
ion as a complica ion o PUUV in ec ion could ha e a majo
impac in endemic a eas. In ligh o bo h bleeding and h ombo ic
e en s ha migh complica e PUUV in ec ions, we hypo he-
sized ha endo helial cells, also he a ge cells o han a i uses
and he majo egula o s o coagula ion and inflamma ion, play
a cen al ole in he pa hogenesis o he disease (Zhou e al.,
2003).
Du ing han a i us in ec ion d as ic al e a ions in he coagu-
la ion sys em ha e been obse ed (Laine e al., 2014). Clinical
s udies ocusing on p ima y and seconda y hemos asis du -
ing han a i us disease showed h ombocy openia in bo h HFRS
and HCPS, a dec eased plasma ac i i y o coagula ion ac o s
II, V, VIII, IX, and X in acu e HFRS pa ien s, p olonga ion
o he p o h ombin and ac i a ed pa ial h omboplas in ime,
inc eased h ombin gene a ion and D-dime le els and a dec ease
in ADAMTS13 ac i i y in acu e PUUV pa ien s (Lee, 1987;
Mackow and Ga ilo skaya, 2009;Laine e al., 2011;Mus onen
e al., 2013). The abili y o in ec endo helial cells by han-
a i uses has been demons a ed bo h in i o and in i o
(Yanagiha a and Sil e man, 1990;Pensie o e al., 1992;To o
e al., 1998). Al hough in ec ion does no lead o cy opa hic
changes, se e al s udies obse ed endo helial cell dys unc ion
du ing han a i us in ec ion (Yanagiha a and Sil e man, 1990;
Pensie o e al., 1992;Mackow and Ga ilo skaya, 2009), ang-
ing om inc eased clinical ma ke s o a s essed endo helium
in i o (sICAM-1, VWF and ci cula ing endo helial cells; Han
e al., 2010;K au k ame e al., 2014), o inc eased pe meabil-
i y and dec eased HUVEC in eg in ligand mig a ion in i o
(Ga ilo skaya e al., 2002;Geimonen e al., 2002;Taylo e al.,
2013).
In eg in ανβ3, expe imen ally p o en o be he ecep o o
han a i us in ec ion, is abundan ly p esen on he su ace o
endo helial cells (Ga ilo skaya e al., 1999;Song e al., 2005).
In ec ion wi h pa hogenic han a i uses is sugges ed o esul in
he loss o unc ion o he ανβ3in eg in(Wang e al., 2012),
bu also an inc eased ανβ3 exp ession on cul u ed endo he-
lial cells and pla ele s has been obse ed (Liu e al., 2008).
Fu he mo e Ga ilo skaya e al. (2010) s udied he adhe ence
o quiescen pla ele s o Sin Nomb e and Han aan i us in ec ed
endo helial cells seems o be he esul o i us binding o
he ανβ3 in eg in p esen on pla ele s (Ga ilo skaya e al.,
2010).
How he abno mali ies in he p ima y ( h ombocy openia)
and seconda y hemos asis (inc ease in h ombin gene a ion
and aised D-Dime le els) a e induced in PUUV in ec ed
pa ien s and he mechanism by which Old-Wo ld han a i uses
cause hemo hage and/o enal ailu e emain la gely elusi e
(Vahe i e al., 2013b). Lack o specific ea men and an effec-
i e accine makes unde s anding o he pa hophysiology o
han a i us in ec ion an impo an medical need, especially wi h
he ecen ly disco e ed associa ion o PUUV wi h ca dio as-
cula disease (Connolly-Ande sen e al., 2014). The e o e, we
ha e used an in eg a ed app oach o s udy changes in p ima y
and seconda y hemos asis using an in i o endo helial cell
model.
Ma e ials and Me hods
Cells
Ve oE6 cells (Ame ican Type Cul u e Collec ion, USA) we e
g own in Dulbecco’s Modified Eagle Medium (DMEM) con-
aining 10% e al bo ine se um (FBS, Lonza, he Ne he lands),
100 U/ml penicillin-s ep omycin solu ion, 1% Hepes buffe and
1% sodium bica bona e (all om Gibco, Li e Sciences, USA).
Human umbilical ein endo helial cells (HUVECs) we e ha -
es ed om umbilical eins, which we e kindly p o ided by
E asmus MC bi h cen e . B iefly, umbilical co ds we e s o ed
in s e ile 500 ml PBS supplemen ed wi h gen amycin (50 μg/ml;
Leo Pha maceu ical, Denma k). Veins we e insed wi h PBS con-
aining 50 U/ml hepa in (Leo Pha maceu ical). Subsequen ly,
cells we e de ached wi h 0.1% collagenase solu ion (C6885,
Sigma Ald ich, USA). Cell suspension was collec ed in a s e ile
50 ml ube ollowed by wo imes cen i uga ion (5 min 300 g).
The cell pelle was e-suspended in HUVEC medium (human
endo helial-SFM medium; In i ogen, Li e Sciences, USA) con-
aining 10% human se um (Lonza), 20% fil a ed FBS (Lonza);
penicilin/s ep omycin 100 U/ml, 20 ng/ml fib oblas g ow h ac-
o (Pep o ech, USA) and 10 ng/ml o endo helial cell g ow h
ac o (Pep o ech). HUVEC cell suspensions we e cul u ed in
F on ie s in Mic obiology | www. on ie sin.o g 2Ma ch 2015 | Volume 6 | A icle 220
Goeijenbie e al. Puumala and endo helial hemos a ic unc ion
flasks p e-coa ed wi h 20 μg/ml o fib onec in (Roche, he
Ne he lands). Only cells up o passage ou , om one specific
dono , we e used o his s udy. Iden i y o he endo helial cells
was confi med by flow cy ome y using Ulex eu opeus lec in,
an i-CD31 and Von Willeb and Fac o (VWF) s aining and
immunoblo .
An i-Se a
We made use o he ollowing an ibodies and conjuga es: poly-
clonal abbi an i- VWF, HRP labeled polyclonal goa an i-
abbi IgG and polyclonal abbi an i-mouse (All om Dako,
he Ne he lands). FITC labeled monoclonal an i-CD31 (Sigma
Ald ich, USA), polyclonal abbi an i-CD41 (Pe bio Science,
he Ne he lands), polyclonal abbi an i-CD3 (Dako), poly-
clonal abbi an i-PUUV nucleop o ein (BEI Resou ces, USA),
monoclonal an i-PUUV glycop o ein (HY Tes , Finland), mon-
oclonal an i-ανβ3 in eg in (Abcam, UK), monoclonal an i-
i onec in (No us Bio, USA), polyclonal abbi -an i PAI-1 (Bio
Connec , he Ne he lands), polyclonal abbi an i- issue ac o
(TF; Bio Connec ), human se um om a eco e ed PUUV case
desc ibed in Goeijenbie e al. (2011) e ie ed a e in o med
consen and e hical boa d app o al. An ibodies and conjuga e
we e dilu ed in dilu ion buffe , which consis ed o PBS wi h
0.5% bo ine se um albumin, 2% NaCl and 1% no mal goa
se um.
Vi us In ec ion
Vi us Isola ion
Lungs o Myodes gla eolus om Konne esi, Finland, in ec ed
wi h PUUV we e homogenized in DMEM (10% w/ ) and 100 μl
was added on o a 70–80% monolaye o Ve oE6 cells and incu-
ba ed o 60 min a 37◦Cin5%CO
2. The supe na an was
disca ded and cells we e washed h ee imes and incuba ed wi h
esh e oE6 medium o an addi ional 5 days. Vi us s ocks up
o passage ou we e c ea ed by cen i uga ion (10 min 400 g)
o he supe na an o c ea e a cell ee i us s ock. Vi us i e
was de e mined using immune pe oxidase eac ion (IPOX) and
TCID50 was calcula ed using he Ka be o mula (Kä be , 1931).
In ec ious i us was inac i a ed using be a-p opiolac one (BPL;
Sigma Ald ich, USA; 1:4000 / ) a 4◦C o 24 h. Subsequen ly,
BPL was inac i a ed o 1 h a 37◦C. All i us s ocks we e
s o ed a −80◦un il use. All expe imen s we e conduc ed unde
biosa e y ins uc ions equi ed ega ding wo k wi h li e PUUV.
Vesicula s oma i is i us (VSV) s ain Indiana, p opaga ed also
on Ve oE6 cells, was kindly p o ided by D . Ba Haagmans
(E asmus MC).
In ec ion Kine ics and Dynamics
Human umbilical ein endo helial cell we e seeded in o 24-
well- (2.4 ×105cells) o 96-well pla es (4 ×104cells; Co ning,
USA) depending on he expe imen . Confluen monolaye s we e
in ec ed wi h a mul iplici y o in ec ion (MOI) o 0.5 o 3, wi h
in ec ious and inac i a ed (BPL-inac i a ed) i us o a no mal
medium con ol o 60 min a 37◦Cin5%CO
2. A e incuba-
ion, he supe na an was disca ded and cells we e washed h ee
imes wi h RPMI 1640 (Gibco, Li e Sciences). F esh medium was
added as desc ibed ea lie . Fo VWF and plasminogen ac i a o
inhibi o ype-1 (PAI-1) quan ifica ion, medium did no con ain
FCS bu was supplemen ed wi h 4% s e ile fil e ed bo ine se um
albumin (BSA, Sigma Ald ich, USA) o a oid addi ion o e al cal
VWF and PAI-1 (Zoellne e al., 1996). To quan i y he pe cen -
age o in ec ed cells we used an in house de eloped IPOX p o-
cedu e. HUVEC we e washed h ee imes wi h PBS. Cells we e
fixed wi h absolu e −20◦C me hanol and incuba ed a −20◦C o
30 min. A e fixa ion, me hanol was disca ded and cells we e
incuba ed o 30 min a 37◦C wi h 100 μl o 0.05% H2O2in
PBS, o block endogenous pe oxidases. Subsequen ly, cells we e
washed h ee imes wi h PBS and incuba ed o 60 min wi h poly-
clonal abbi an i-PUUV nucleop o ein an ibody (1:500). Cells
we e washed wi h PBS 0.05% ween ollowed by incuba ion wi h
HRP-labeled goa an i- abbi IgG conjuga e (1:500). Colo de el-
opmen was achie ed by addi ion o 3-amino-9-e hylca bazole
(AEC) subs a e (AEC dissol ed in dime hyl o mamide buffe ed
wi h ace a e buffe o pH 5). Pe cen age o in ec ed cells was
de e mined by manual coun ing.
Fo he quan ifica ion o i al eplica ion we used a s an-
da d line o in house gene a ed PUUV RNA un-off an-
sc ip s, as desc ibed o Wes -Nile i us (Lim e al., 2013).
B iefly, RNA un-off ansc ip s we e gene a ed using a seg-
men amplified wi h pan-han a i us degene a i e PCR p ime s
om (Johansson e al., 2010). PCR p oduc s we e sepa a ed
on 1% aga ose gel and bands o co ec size we e collec ed
o DNA gel ex ac ion using he MinElu e Gel Ex ac ion
Ki P o ocol (Qiagen, USA). DNA agmen s we e cloned in o
he pCR4 ec o using he TOPOTA Cloning KIT (Li e
Technologies) and One Sho TOP10 chemically compe en
Esche ichia coli we e ans o med wi h he ecombinan ec-
o (QIAGEN) acco ding o manu ac u e ’s p o ocol. A leas
fi e colonies we e collec ed o u he analyses. Plasmid DNA
was pu ified using MinElu e DNA pu ifica ion ki (QIAGEN).
Plasmid DNA was linea ized by es ic ion diges ion (No I o
he nega i e s and RNA and Ps I o he posi i e s and RNA).
Run-off ansc ip s (in i o ansc ip s) we e syn hesized using
T3 RNA polyme ase o nega i e s and and T7 RNA poly-
me ase o posi i e s and (MEGAsc ip T3 and T7 an-
sc ip ion ki s, Li e echnologies), ollowed by DNase ea men
(AmbionTURBO DNA- eeTM Li e Technologies), acco ding
o manu ac u e ’s manual. The amoun o RNA in he s ock
was de e mined using NanoD opand se ially dilu ed. Copy
numbe s in he s anda ds we e calcula ed using RNA concen-
a ion and sequence leng h wi h help o an online calcula o
(h p://endmemo.com/bio/dnacopynum.php).
Pla ele Binding Assay
Pla ele Collec ion
To s udy in e ac ion be ween PUUV and pla ele s, pla ele s we e
collec ed acco ding o he p o ocol desc ibed in Ga ilo skaya
e al. (2010) wi h mino modifica ions. B iefly, blood was col-
lec ed in 0.105 M (end concen a ion) sodium ci a e ubes
(BD-plymou h, UK) supplemen ed wi h 1 μM p os aglandin E1
(Cayman Chemical, USA) o block pla ele ac i a ion. Pla ele -
ich plasma (PRP) was p epa ed by cen i uga ion o 15 min a
700 ×ga 25◦C. Subsequen ly, pla ele s we e pelle ed o 15 min
a 1300 ×ga 25◦C. Pla ele s we e washed wice and esuspended
F on ie s in Mic obiology | www. on ie sin.o g 3Ma ch 2015 | Volume 6 | A icle 220
Goeijenbie e al. Puumala and endo helial hemos a ic unc ion
wi h modified hepes buffe (25 mM Hepes, 137 mM NaCl, 0.1%
Albumin and 1 μM p os aglandin E1 pH 7.4) and coun ed by a
hema ocy ome e .
Pla ele HUVEC Binding
Pla ele s (108pe ml) we e incuba ed wi h in ec ed HUVEC (96
wells pla e; PUUV, BPL inac i a ed PUUV o mock con ol)
o 30 min a 37◦C. A e incuba ion, monolaye s we e washed
h ee imes wi h RPMI and cells we e fixed wi h o malin. A e
fixa ion, o malin was disca ded and cells we e incuba ed o
30 min a 37◦C/5% CO2wi h PBS 0.05% H2O2.A e h ee
washing s eps cells we e incuba ed wi h abbi polyclonal an i-
human CD41a an ibody (1:500). The ollowing s eps we e as
desc ibed ea lie o IPOX. A e incuba ion wi h HRP-labeled
goa an i- abbi conjuga e (1:1000) TMB was added o he wells
o subs a e eac ion. A e 10 min eac ion was s opped by
addi ion o 0.5 M sulphu ic acid and op ical densi y (OD) was
measu ed a 450 nm using Tecan ELISA eade . CD41a exp es-
sion OD was calcula ed by sub ac ing he blanc OD alue (wells
incuba ed wi hou pla ele s bu wi h de ec ion an ibody and con-
juga e). Rabbi polyclonal an i-CD3 (1:500) se ed as an iso ype
con ol.
Pla ele PUUV Binding
To es i changes in CD41a exp ession was ela ed o di ec
binding be ween PUUV and pla ele s, a mechanism shown
in Han aan and Andes i us in ec ion (Ga ilo skaya e al.,
2010), a pull down assay was designed. To his end, we fi s
coa ed ELISA pla es wi h PUUV o a con ol i us (VSV;
100 μlo 10
6 i us pa icles in DMEM a 4◦Co e nigh )
ollowed by pla ele incuba ion (107pla ele s). Subsequen ly,
cells we e washed fi e imes wi h PBS and he bound pla ele s
we e quan ified by using a pla ele de ec ion an ibody (an i-
CD41a 1:500 in dilu ion buffe ) ollowed by a conjuga e sub-
s a e eac ion. CD41a exp ession was calcula ed by sub ac-
ion o he OD measu ed in he wells wi hou pla ele incu-
ba ions (blanco) o co ec o di ec (a-specific) an i-CD41a
an ibody binding o PUUV and an i-CD3 was used as iso ype
con ol. Subsequen ly, ELISA pla es we e coa ed wi h mouse
monoclonal an i-PUUV glycop o ein- o iso ype con ol an i-
body (IgG2 co ona i us; 1:500 in PBS a 4◦Co e nigh )
ollowed by incuba ion wi h PUUV o cap u e he i us ol-
lowed by pla ele incuba ion, de ec ion an ibody and conjuga e
subs a e eac ion. Thi dly, o u he confi m pla ele PUUV
binding, ELISA pla es we e coa ed wi h an an i-pla ele an i-
body (1:500 in PBS a 4◦C o e nigh ), ollowed by incuba ion
wi h esh isola ed pla ele s and e en ually an incuba ion s ep
wi h PUUV o VSV ollowed by a han a i us de ec ion an i-
body (mouse mAB an i-PUUV-glycop o ein 1:500). A e wash-
ing subs a e eac ion was achie ed by conjuga e addi ion and
TMB eac ion s eps. As a final s ep we s udied he po en ial
blocking o pla ele binding by PUUV pa icles by he addi-
ion o a blocking s ep wi h polyclonal human an i PUUV
se um. PUUV coa ed pla es and pla es coa ed wi h 5 days old
i us ee Ve oE6 medium we e incuba ed wi h a polyclonal
PUUV se um (1:50) om a case desc ibed in Goeijenbie e al.
(2011) o wi h a PUUV IgG nega i e con ol human se um
om a heal hy olun ee (also 1:50). The ollowing pla ele bind-
ing s eps and CD41 de ec ion we e he same as in he ea lie
expe imen s.
Von Willeb and Fac o and Plasminogen
Ac i a o Inhibi o Type-1 Quan i ica ion
A e in ec ion, in a 24-well pla e, supe na an s (500 μl) we e
subsequen ly emo ed and cells we e lysed, a e h ee washing
s eps, using a 15 min incuba ion wi h 500 μl PBS 1% T i onX-100
ollowed by cen i uga ion (10 min 400 ×g). Cell- ee supe -
na an s and supe na an om cell lysa es we e measu ed using
a PAI-1 an igen and VWF ELISA ki s acco ding he manu-
ac u e ’s ins uc ions (bo h om Zymugen, Hyphen Biomed,
F ance).
Tissue Fac o Exp ession and Ac i i y
Tissue Fac o Cell Su ace Exp ession
Human umbilical ein endo helial cell in 96 wells pla es we e
fixed wi h 4% o malin and incuba ed wi h abbi polyclonal an i-
TF an ibody (1:500) ollowed by incuba ion wi h he espec i e
conjuga e (1:500). A e washing TMB was added o subs a e
eac ion and eac ion was s opped a e 10 min by addi ion o
0.5 M o sulphu ic acid. OD 450 nm alue was measu ed on Tecan
ELISA eade .
Tissue Fac o Cell Lysa e Concen a ion
Cell lysa es we e p epa ed as desc ibed in Sec ion “Von
Willeb and Fac o and Plasminogen Ac i a o Inhibi o Type-
1 Quan ifica ion.” ELISA pla es we e coa ed wi h a mix u e o
50 μlcelllysa eand50μlPBSo e nigh a 4
◦C oge he
wi h a s anda d cu e o ecombinan TF (Inno in; Siemens
Heal hca e Diagnos ics, Ge many). A e blocking wells we e
incuba ed wi h abbi polyclonal an i-TF an ibody (1:500) ol-
lowed by incuba ion wi h he espec i e conjuga e (1:500).
A e washing TMB was added o subs a e eac ion and eac-
ion was s opped a e 10 min by addi ion o 0.5 M o sul-
phu ic acid. OD 450 nm alue was measu ed on Tecan ELISA
eade .
Th ombin Gene a ion
Th ombin gene a ion in pla ele -poo plasma (PPP) was mea-
su ed di ec ly on HUVEC su ace, in a 96 well pla e, by ecal-
cifica ion o 80 μl o pooled ci a ed plasma om heal hy
dono s added o he monolaye o in ec ed and unin ec ed
cells. In summa y, cells we e washed h ee imes wi h RPMI
and 80 μl eshly hawed plasma was added o he mono-
laye oge he wi h60μl o HEPES buffe (25 mM Hepes,
137 mM NaCl, 0.1% albumin). On he same pla e a se ial
dilu ion o ecombinan TF (Inno in; Siemens Heal hca e
Diagnos ics, Ge many) in he absence o cells. Finally, 60 μlo
HEPES calcium [25 mM Hepes, 137 mM NaCl, 0.1% Albumin,
38 mM CaCl(2)] was added o plasma. Di ec ly a e ecal-
cifica ion, OD 450 nm alue was measu ed using a Tecan
ELISA eade inakine iccyclemeasu inge e y45s o 1h.
Th ombin gene a ion ime was defined as he ime a hal -
maximal OD.
F on ie s in Mic obiology | www. on ie sin.o g 4Ma ch 2015 | Volume 6 | A icle 220
Goeijenbie e al. Puumala and endo helial hemos a ic unc ion
Vi onec in – PAI-1 Complex Le els
ELISA pla es we e coa ed wi h an i- i onec in an ibody (1:500
in PBS a 4◦C o e nigh ), incuba ed wi h supe na an om
PUUV in ec ed o non-in ec ed HUVEC ollowed by incuba-
ion wi h polyclonal an i-PAI-1 an ibody (1:500) and subsequen
conjuga e-subs a e eac ion. PBS incuba ion was used as a blanc
con ol and an i-CD3 an ibody (1:500) incuba ion as an iso ype
con ol.
S a is ics
All s a is ical analyses we e pe o med using G aphPad P ism
5.01 o Windows. When compa ing wo g oups we made
use o a S uden ’s - es o Mann–Whi ney U, depending on
he dis ibu ion o he da a. Fo he compa ison be ween
mul iple g oups non-pa ame ic K uskal-Wallis es was used
wi h Dunn’s mul iple compa ison es o a one-way ANOVA
wi h Tukey’s mul iple compa isons es , depending on he
dis ibu ion o he da a. P≤0.05 we e conside ed signifi-
can .
Resul s
PUUV In ec s and Replica es in P ima y
Endo helial Cells
To p e en PUUV om in i o loss o i ulence, i us s ocks
o no mo e han ou passages we e p epa ed. F eshly isola ed
HUVEC we e in ec ed wi h MOI 0.5 and 3. The PUUV in ec ed
and eplica edinHUVEC,asissumma izedinFigu e 1.Non-
in ec ed cells (Figu es 1A,B) showed no ed pe oxidase s aining,
confi ming specifici y o he PUUV-s aining. F om 24 h pos
in ec ion wi h a low MOI in ec ion (0.5) onwa d (Figu e 1C)
FIGURE 1 | Puumala i us (PUUV) immune pe oxidase (IPOX) s aining
and i al copy numbe s a e in ec ion wi h a mul iplici y o in ec ion
(MOI) o 0.5 and 3. Mock in ec ed cells (A,B) showed no ed pe oxidase
s aining o PUUV nucleop o ein a e 24 (A) and 48 (B) hou s. Twen y- ou
hou s a e in ec ion wi h MOI 0.5 (C) a small numbe o cells s ained posi i e,
which, inc eased a 48 h (D). In ec ion wi h MOI 3 esul ed in mo e in ec ed
cells a e 24 h (E), which inc eased sligh ly a 48 h pos in ec ion (F).Analysis
o i al eplica ion showed a mo e han 2-log inc ease o i al copy numbe s in
bo h supe na an (G) and cell lysa es (H), sugges ing ac i e i al eplica ion.
Ba s ep esen SE o he mean. BPL inac i a ion o he i us lead o no
inc ease in i al copy numbe s in he supe na an and a nega i e IPOX s aining
(da a no shown). Fu he mo e in ec ion was con i med by wes e n blo o he
p esence o he PUUV nucleop o ein in he cell lysa e and Von Willeb and
Fac o (VWF) o con i m he cha ac e o he endo helial cells (I). The i s ou
lanes show con ol wells wi h only one band p esen a he uppe side o he
blo (VWF). The las o lanes show he p esence o bo h VWF and he PUUV
nucleop o ein (∼55 kDa 10ug/lane). Da a a e ep esen a i e o h ee
independen expe imen s.
F on ie s in Mic obiology | www. on ie sin.o g 5Ma ch 2015 | Volume 6 | A icle 220

Goeijenbie e al. Puumala and endo helial hemos a ic unc ion
only a small pe cen age (±10%) o he cells we e in ec ed, which
s ongly inc eased a e 48 h (Figu e 1D), esul ing in 50% o
in ec ed s ained cells. Twen y- ou hou s a e in ec ion a a
MOI o 3 abou 40–50% o cells we e in ec ed (Figu e 1E),
which inc eased u he o 80% by 48 h (Figu e 1F). Compa able
kine ics we e seen when i al RNA copy numbe s we e de e -
mined. To his end, i al RNA numbe s we e es ima ed bo h in
supe na an (Figu e 1G) and cell lysa e (Figu e 1H). A bo h
MOIs he numbe o i al RNA inc eased significan ly (2-Log)
a e 48 h (K uskal-Wallis; p=0.0028), in he supe na an as
well as in he cell lysa e, confi ming ac i e i al eplica ion. Vi al
eplica ion eached a pla eau a 72 h pos in ec ion.
Fu he mo e wes e n-blo analysis o he cell lysa e o PUUV
nucleop o ein confi med in ec ion o HUVEC (Figu e 1I). The
i al copy numbe s in he supe na an o cell lysa e o HUVEC
incuba ed wi h BPL inac i a ed i us se ed as a con ol o non-
eplica ing i us. Consis en ly, he RNA copy numbe s did no
inc ease o e ime indica ing he efficien inac i a ion o he i us
by BPL ea men . Efficien inac i a ion was confi med by nega-
i e IPOX s aining o he HUVEC incuba ed wi h BPL inac i a ed
PUUV (da a no shown).
Inc eased CD41a Exp ession a e
Incuba ion o Pla ele s on HUVEC upon
PUUV In ec ion
Ga ilo skaya e al. (2010) epo ed he po en ial o Han aan and
Andes i us o bind quiescen pla ele s ia ανβ3 in eg in. Since
his obse a ion is o much impo ance in u he unde s and-
ing he al e a ions in p ima y hemos asis and i s ole in disease
mechanisms in HFRS we decided o confi m his mechanism o
PUUV using a diffe en app oach. Fi s we assessed he abili y o
quiescen pla ele s o bind o PUUV in ec ed HUVEC (Figu e 2).
Binding o pla ele s was de e mined by measu ing he in ensi y
o CD41a (pla ele glycop o ein IIb), a he e odime ic in eg al
memb ane p o ein p esen only on pla ele s and megaka y-
ocy es. CD41a exp ession was significan ly highe on he HUVEC
monolaye a e in ec ion wi h a MOI o 0.5, o 3 compa ed o
he con ol. De ec ion o CD41a exp ession did no diffe when
pla ele s we e no added o he HUVEC monolaye s, sugges ing
ha he e was no non-specific an i-CD41a binding o in ec ed
cells. Fu he mo e he e was no diffe ence in OD alues when an
iso ype con ol (an i-CD3) was used o de ec pla ele s. Based on
CD41a exp ession, s a is ically significan diffe ences we e mea-
su ed be ween in ec ed wells and wells incuba ed wi h i us ee
Ve oE6 medium (nega i e con ol) a 24 h pos in ec ion (one
way ANOVA, MOI 3 s. NEG p<0.01, MOI 0.5 s. Neg p<0.05).
A e 48 h o in ec ion his diffe ence in CD41a exp ession was
also significan be ween MOI 3 in ec ed wells and he BPL inac i-
a ed i us con ol (one way ANOVA, MOI 3 s. NEG p<0.001
andMOI3 s.BPLp<0.001; MOI 0.5 s. NEG p<0.05). Taken
oge he , he da a indica e ha pla ele s bind o cul u es incu-
ba ed wi h PUUV. HUVEC incuba ed wi h BPL did show a end
o inc eased pla ele CD41a exp ession (Figu e 2), bu his was
no s a is ically significan .
Von Willeb and Fac o (VWF) is no
Inc eased Du ing PUUV In ec ion o HUVEC
We measu ed VWF an igen in cell ee supe na an and VWF
exp ession on he su ace o in ec ed HUVEC. Inc eased VWF
p oduc ion may be a gene al inflamma o y esponse o endo he-
lial cells ha could be e oked as a esul o PUUV in ec ion.
Howe e , a he ime poin s whe e pla ele binding inc eased,
HUVECs in ec ed wi h PUUV showed no al e a ion in VWF
ac i i y, as de e mined by ELISA, in nei he he supe na an no
he cell lysa e (da a no shown) compa ed o BPL o nega i e
con ol.
Pla ele s Bind Di ec ly o PUUV
Nex we looked whe he he pla ele s could bind di ec ly o
PUUV pa icles. Fo his pu pose we pe o med an in-house
de eloped pla ele pull down-assay using quiescen pla ele s. To
his end, se e al expe imen s we e conduc ed o demons a e
specifici y o his binding. Figu e 3 shows esul s o binding o
pla ele s o i us-coa ed ELISA pla es (Figu e 3A). Mo e pla ele s
FIGURE 2 | Inc eased pla ele binding o PUUV in ec ed HUVEC.
Inc eased op ical densi y (OD) o CD41a was measu ed bo h 24 (A) and 48 (B)
hou s a e in ec ion wi h bo h low and mode a e MOI and pla ele incuba ion on
HUVEC su ace. p alues (*p<0.05, **p<0.01, ***p≤0.001) a e he esul o
one way ANOVA es ing wi h Tuckey’s mul iple compa ison pos es . Ba s
ep esen he SE o he mean. Incuba ion wi h an iso ype con ol an ibody
(polyclonal an i-CD3) did no lead o inc eased OD on in ec ed o con ol
HUVEC. Da a a e ep esen a i e o h ee independen expe imen s.
F on ie s in Mic obiology | www. on ie sin.o g 6Ma ch 2015 | Volume 6 | A icle 220
Goeijenbie e al. Puumala and endo helial hemos a ic unc ion
FIGURE 3 | Puumala i us and pla ele s bind o each o he . In a pull
down assay pla ele s adhe e be e o PUUV i us pa icles compa ed o
esicula s oma i is i us (VSV; A; Mann–Whi ney Up=0.0022). When i us
was cap u ed wi h a PUUV glycop o ein an ibody (B), pla ele s we e able o
bind o he cap u ed i us, in con as o wells coa ed wi h an IgG2 con ol
an ibody (an i-co ona i us glycop o ein; Mann–Whi ney Up=0.0022),
esul ing in no cap u e o PUUV du ing he incuba ion p ocess, con olling o
po en ial o he ac o s p esen in he i us s ock medium. When pla ele s
we e bound o pla es coa ed wi h an an i-CD41a an ibody (C), he PUUV
pa icles we e able o bind o pla ele s based on he signi ican inc ease in
PUUV de ec ion OD compa ed o wells incuba ed wi h VSV pa icles, hus no
PUUV p esen (Mann–Whi ney Up=0.0043). The binding be ween PUUV
and pla ele s could be blocked by he addi ion o a blocking s ep wi h
human an i-PUUV se um (D) which show a dec eased CD41 exp ession
when compa ed o he PUUV coa ed wells incuba ed wi h a PUUV nega i e
con ol se um. In all expe imen s no di e ence in OD was measu ed when an
iso ype con ol an ibody was used. Da a a e ep esen a i e o h ee
independen expe imen s. ∗∗p<0.01, ∗∗∗p≤0.001.
(Mann–Whi ney U;p=0.0022) adhe ed o pla es di ec ly coa ed
wi h PUUV compa ed o pla es coa ed wi h a i us con ol
(VSV), which was cul u ed unde he same condi ions as PUUV.
Subsequen ly, o con ol i he binding o pla ele s was di ec ly
o he PUUV pa icles and no due o ano he ac o p esen in
he Ve oE6 supe na an we made use o a sandwich ELISA p in-
ciple. PUUV was incuba ed on ELISA pla es wi h wells coa ed
wi h a monoclonal IgG2 specific o he glycop o ein o PUUV o
wi h a IgG2 con ol an ibody. By his app oach significan ly mo e
pla ele s bound o he wells whe e PUUV was cap u ed compa ed
o wells wi h no PUUV cap u e (Mann–Whi ney U;p=0.0022;
Figu e 3B).
To confi m di ec binding be ween pla ele s and PUUV,
pla ele s we e cap u ed o an i-CD41 coa ed ELISA pla es and
incuba ed wi h i us ollowed by de ec ion wi h a PUUV specific
an ibody. To con ol o binding be ween PUUV de ec ion an i-
body and cap u ed pla ele s, con ol wells we e incuba ed wi h
VSV. PUUV de ec ion was significan ly highe in he wells
incuba ed wi h PUUV compa ed o VSV (Mann–Whi ney U;
p=0.0043; Figu e 3C). These expe imen s collec i ely sugges
ha pla ele s can specifically bind o PUUV.
Finally, we show in Figu e 3D ha he binding o pla ele s
o PUUV pa icles could be blocked by addi ion o a blocking
s ep wi h human se um om a eco e ed PUUV case. When
wells coa ed wi h PUUV we e incuba ed wi h human se um
wi h p o en PUUV neu alizing IgG an ibodies significan ly less
pla ele s adhe ed o he wells compa ed o wells incuba ed wi h
a PUUV nega i e human con ol se um (Figu e 3D;p<0.001).
The an i-CD41 exp ession in he wells wi h a blocking s ep was
compa able o ha o he nega i e con ol, which consis ed ou o
F on ie s in Mic obiology | www. on ie sin.o g 7Ma ch 2015 | Volume 6 | A icle 220
Goeijenbie e al. Puumala and endo helial hemos a ic unc ion
FIGURE 4 | Puumala i us in ec ion o HUVEC induces issue ac o
exp ession esul ing in enhanced h ombin gene a ion. MOI 3 in ec ion
esul ed in inc eased TF exp ession (A, Mann–Whi ney Up=0.0002) on
HUVEC su ace a e 48 h (A) and in he cell lysa e (B). Th ombin gene a ion
ime (TGT) was signi ican ly dec eased, indica ing mo e h ombin o ma ion, o
cells in ec ed wi h PUUV (MOI 3) a 24 (C) and 48 (D) hou s pos in ec ion (one
way ANOVA; 24 h p<0.01 and 48 h p<0.001). Resul s a e shown wi h
whiske s om minimum o maximum. MOI 0.5 in ec ion esul ed in sho ened
TGT 48 h pos in ec ion (one way ANOVA p<0.001). Mean issue ac o (TF)
concen a ion, calcula ed om TGT s anda d cu e, inc eased mo e han
nine old a e MOI 3 in ec ion (one way ANOVA 24 and 48 h p<0.01). MOI 0.5
in ec ion inc eased TF concen a ion h ee old a e 48 h (one way ANOVA
p<0.05; E, F; ba s ep esen SE o he mean). Da a a e ep esen a i e o h ee
independen expe imen s. ∗p<0.05, ∗∗ p<0.01, ∗∗∗ p≤0.001.
pla es coa ed wi h 5 days old e o E6 i us ee medium. Binding
o he neu alizing IgG an ibodies was confi med by incuba ion
wi h a goa -an i human HRP labeled conjuga e and subsequen
TMB eac ion.
Inc eased Th ombin Gene a ion and Tissue
Fac o Exp ession a e PUUV In ec ion o
HUVEC
To es he hypo hesis whe he inc eased h ombin gene a ion
obse ed in acu e PUUV pa ien s is he esul o inc eased TF
exp ession on endo helial cells we incuba ed HUVEC, in ec ed
wi h PUUV a a MOI o 3 o wi h a i us ee 5 days old Ve o
E6 medium (con ol) wi h a polyclonal an i-TF an ibody. By
his app oach we showed ha TF exp ession was significan ly
inc eased wi h an almos wo old inc ease in OD alue 48 h
pos in ec ion (Figu e 4A, Mann–Whi ney U;p=0.0047). Cells
in ec ed wi h PUUV also showed an inc eased TF concen a ion
when he cell lysa es o PUUV in ec ed wells we e compa ed o
he lysa es o con ol wells (Figu e 4B; bo h mock and BPL).
Subsequen ly we wan ed o p o e ha he inc ease in TF exp es-
sion on he endo helial cell su ace was o biological significance
and would led o inc eased h ombin gene a ion. Th ombin
gene a ion was quan ified di ec ly on in ec ed endo helial cells by
incuba ing no mal plasma on cells and ini ia ing coagula ion by
he addi ion o calcium ions. In ec ed cells induced plasma clo -
ing as e due o inc eased h ombin gene a ion (Figu es 4C,D).
F on ie s in Mic obiology | www. on ie sin.o g 8Ma ch 2015 | Volume 6 | A icle 220
Goeijenbie e al. Puumala and endo helial hemos a ic unc ion
FIGURE 5 | Plasminogen ac i a o inhibi o ype 1 p oduc ion, su ace
exp ession o ανβ3-in eg in and PAI-1 i onec in complex o ma ion in
HUVEC in ec ed wi h PUUV. To al plasminogen ac i a o inhibi o ype-1
(PAI-1) concen a ion was signi ican ly inc eased on ime poin s 48 pos PUUV
in ec ion (black; one way ANOVA, p<0.05; A) compa ed o BPL inac i a ed
con ol (g ay) o nega i e medium con ol (whi e). Panel B shows he PAI-1
i onec in complex le els in HUVEC in ec ed wi h PUUV (black) o non-in ec ed
con ols (whi e). A e 48 h o in ec ion PAI-1 i onec in le els we e inc eased
compa ed o he mock in ec ion (Mann–Whi ney Up=0.03; B). Da a a e
ep esen a i e o h ee independen expe imen s. ∗p<0.05.
Using a calib a ion cu e wi h pu ified TF in he absence o
endo helial cells, we quan ified HUVEC TF p oduc ion a e i us
in ec ion and a e incuba ion wi h a i us ee medium con ol.
TF concen a ion showed a s a is ically significan inc ease o
MOI 3 a 24 h compa ed o he nega i e con ol and he HUVEC
in ec ed wi h MOI 0.5 (one way ANOVA; p<0.01) and a
48 h pos in ec ion compa ed o he nega i e con ol (one way
ANOVA; p<0.001). The MOI 0.5 in ec ion led o highe le els
o TF on he HUVEC su ace only a e 48 h pos in ec ion (one
way ANOVA; p<0.05) when compa ed o mock.
Inc eased Le els o PAI-1 and
PAI-1-Vi onec in Complexes in HUVEC
Supe na an a e In ec ion wi h PUUV
Impo an p o eins in he egula ion o fib inolysis show close
in e ac ions wi h he pa hogenic han a i us ecep o ανβ3in e-
g in. Fo ins ance i onec in, a s abilize o PAI-1 ac i i y
in plasma, is la gely egula ed by his ecep o (Mackow and
Ga ilo skaya, 2009;Flo o a e al., 2013). To s udy po en ial
changes in egula o s o fib inolysis we fi s measu ed PAI-
1 le els in he cell- ee supe na an and supe na an o cell
lysa e om 24-well pla es in ec ed wi h PUUV o con ol in ec-
ions. The o al PAI-1 an igen ( he combina ion o le els in
he supe na an and cell lysa e) was significan ly inc eased
48 h pos in ec ion (K uskal–Wallis; p<0.05) wi h MOI 3
(Figu e 5A). Subsequen ly, we es ed i in ou model PAI-1
would bind o i onec in, since his binding is associa ed wi h
inc eased/p olonged PAI-1 ac i i y (Seiffe and Losku off, 1991),
and i his in e ac ion is al e ed du ing in ec ion. ELISA pla es
coa ed wi h a monoclonal an ibody agains i onec in, incu-
ba ed wi h supe na an om ou expe imen s (pooled, con ol o
om MOI 3 in ec ed wells) ollowed by incuba ion wi h PAI-1
an ibody sugges ed o ma ion o PAI-1 i onec in complexes
due o an inc ease in OD compa ed o incuba ion wi h PBS
(mean exp ession in medium 490 mOD (±100) s. 370 mOD
(±70) p =0.02). I supe na an s we e es ed sepa a ely (PUUV
s. mock) le els o PAI-1 i onec in complexes we e inc eased
a e PUUV in ec ion (Mann–Whi ney U;p=0.03; Figu e 5B).
Discussion
The p esen s udy add esses pla ele binding o PUUV in ec ed
cells and ac i a ion o seconda y hemos asis a e endo helial cell
PUUV in ec ion. Wi h he lack o a alid and accessible animal
model o old-wo ld han a i us in ec ion, we emain dependen
on ex i o cell cul u e models o add ess ques ions ega ding i u-
lence and pa hogenesis (Vahe i e al., 2013b). Taking in o accoun
he ecen ly ound associa ion o PUUV in ec ion wi h ca dio as-
cula disease (Connolly-Ande sen e al., 2014) and hemo hagic
complica ions ha may occu du ing in ec ion, he in e ac ion
be ween PUUV and he coagula ion sys em especially wa an s
u he a en ion. Since PUUV ends o apidly lose i ulence
upon in i o cell passages he use o low passaged isola es is o
i al impo ance (Nemi o e al., 2003). The e o e, we ha e pu
a lo o emphasis on ob aining low passage PUUV isola es and
op imisa ion o he hemos a ic assays unde he igh biosa e y
egula ions using p ima y cell cul u es.
Based on hemos a ic changes seen in se e al clinical s ud-
ies, mos om No he n Eu ope, we decided o s udy specific
pa s o he coagula ion sys em in i o.Wes a edbys udy-
ing he effec s o PUUV in ec ion on o ma ion o a pla ele
plug, he majo e en in p ima y hemos asis. Binding o pla ele s
by PUUV in ec ed cells could explain h ombocy openia in
F on ie s in Mic obiology | www. on ie sin.o g 9Ma ch 2015 | Volume 6 | A icle 220