Vitamin D Deficiency Causes Defective Resistance to Aspergillus fumigatus in Mice via Aggravated ...

12 Pages·2014·1.56 MB·English

Vitamin D Deficiency Causes Defective Resistance to
Aspergillusfumigatusin Mice via Aggravated and
Sustained Inflammation
Pei Li1, Xiaoyong Xu1, Ehong Cao1, Bo Yu2, Wanchun Li2, Ming Fan3, Mei Huang3, Lining Shi4,
Rong Zeng4, Xin Su1.*, Yi Shi1.*
1DepartmentofRespiratoryandCriticalCareMedicine,JinlingHospital,NanjingUniversitySchoolofMedicine,NanjingJiangsuProvince,People’sRepublicofChina,
2DepartmentofPathology,JinlingHospital,NanjingUniversitySchoolofMedicine,NanjingJiangsuProvince,People’sRepublicofChina,3DepartmentofClinical
Microbiology,JinlingHospital,NanjingUniversitySchoolofMedicine,NanjingJiangsuProvince,People’sRepublicofChina,4DepartmentofClinicalLaboratory,Jinling
Hospital,NanjingUniversitySchoolofMedicine,NanjingJiangsuProvince,People’sRepublicofChina
Abstract
Background:VitaminDplaysanimportantroleinpulmonaryresistanceandimmunity,anditsdeficiencyhasbeenlinkedto
various respiratory infections. Little is known about the effect of vitamin D deficiency on host pulmonary defense to
Aspergillus fumigatus (A.fumigatus).
Methods:Mice raised on vitamin D sufficient or deficient diets were infected intratracheally with A. fumigatus conidia.
Mortality,fungalgrowth,weightlossandlunghistologyweremonitored.Alveolarmacrophages(AMs)werestimulatedwith
A.fumigatusconidiainvitro.Thekineticsofpro-inflammatorycytokines(TNF-a,IL-1bandIL-6),chemokines(CXCL1,CCL3),
and pattern recognition receptors (Toll-like receptor [TLR] 2, TLR 4 and dectin-1) expression in the lungs and AMs were
measured.
Results:Upon A. fumigatus infection, vitamin D deficient mice showed higher mortality, greater fungal load, and more
weightlossthanitssufficientcounterparts.VitaminDdeficientmicedemonstratedaggravatedandprolongedhistological
evidence of lung inflammation as well as enhanced BAL cell counts, dominated by neutrophils after A. fumigatus
inoculation.Increasedbasallevelsofpro-inflammatorycytokinesinthelungsandAMsfromna¨ıvevitaminDdeficientmice
wereobserved.UponA.fumigatusexposure,vitaminDdeficiencyledtoenhancedandsustainedexpressionofTNF-a,IL-1b,
IL-6,CXCL1andCCL3bothinvivoandinvitro.Up-regulationofTLR2,TLR4anddectin-1wasobservedinthelungsandAMs
from vitaminD deficientmiceboth atbaseline andafter A.fumigatus exposure.
Conclusions:VitaminDdeficiencycausesdefectivepulmonaryresistancetoA.fumigatusinmice,possiblybytheenhanced
basalexpressionofpatternrecognitionreceptors andpro-inflammatorycytokines,whichinducedexcessiveinflammatory
responseinresponse toA. fumigatus challenge.
Citation:LiP,XuX,CaoE,YuB,LiW,etal.(2014)VitaminDDeficiencyCausesDefectiveResistancetoAspergillusfumigatusinMiceviaAggravatedandSustained
Inflammation.PLoSONE9(6):e99805.doi:10.1371/journal.pone.0099805
Editor:JoySturtevant,LouisianaStateUniversity,UnitedStatesofAmerica
ReceivedDecember12,2013;AcceptedMay19,2014;PublishedJune13,2014
Copyright:(cid:2)2014Lietal.Thisisanopen-accessarticledistributedunderthetermsoftheCreativeCommonsAttributionLicense,whichpermitsunrestricted
use,distribution,andreproductioninanymedium,providedtheoriginalauthorandsourcearecredited.
Funding:ThisworkwassupportedbyNationalNaturalScienceFoundationofChina(NSFC81270064).Thefundershadnoroleinstudydesign,datacollection
andanalysis,decisiontopublish,orpreparationofthemanuscript.
CompetingInterests:Theauthorshavedeclaredthatnocompetinginterestsexist.
*E-mail:[email protected](XS);[email protected](YS)
.Theseauthorscontributedequallytothiswork.
Introduction activation, and alteration of T-cell activation [6]. These cellular
effects are important for host responses against infection and
ItisincreasinglyrecognizedthatvitaminDplaysanimportant impairedvitaminDstatuswouldcompromisetheimmuneactivity
roleinpulmonarydefense,immunity,andinflammatoryprocesses of vitaminD.
[1,2].TheactivevitaminDgeneratingenzyme,1a-hydroxylase,is
Vitamin D deficiency and insufficiency is a global issue which
expressed by the airway epithelium[3], alveolar macrophages
hassignificantimplicationsforhealth[7,8].Multiplestudieshave
(AMs)[4] and dendritic cells[5], indicating that active vitamin D
identifiedacorrelationbetweenvitaminDdeficiencyandvarious
can be produced locally within the lungs, that it acts in an
respiratory infections, especially increased risk of tuberculosis,
autocrine or paracrine fashion, and is responsible for many
influenza and viral respiratory tract infections [9,10,11,12]. A.
immunomodulatoryactions.ImmuneeffectsofvitaminDinclude
fumigatusremainsthemostcommonAspergillusspeciesthatcause
increased secretion of the antimicrobial peptide cathelicidin,
respiratory infections in humans, in particular in the immuno-
decreased chemokine production, inhibition of dendritic cell
compromised patient population [13]. However, little has been
PLOSONE | www.plosone.org 1 June2014 | Volume 9 | Issue 6 | e99805
RoleofVitDonHostResistancetoA.fumigatus
donetoassesstheimpactofimpairedvitaminDstatusonthehost 30hrs after infection. Survival, clinical appearance and body
control of pulmonary A. fumigatus infection. VitaminD deficiency weightweremonitoreddailyforeachgroup.Forsurvivalstudies,
hasbeenproposedasariskfactorforAllergicBronchopulmonary miceweremonitoredevery6 hrpost-A.fumigatuschallenge.Mice
Aspergillosis (ABPA) in cystic fibrosis (CF) patients[14]. Based on wereobservedonaregularbasisduringthedayandwereweighed
thefactthatlocalsynthesisofactivevitaminDinlungcanactto each morning. Body temperature was taken in the morning and
support immunomodulation at this barrier site, we hypothesized eveningwithadigitalthermometerinsertedintothevagina.Time
that vitaminDdeficiencymaycompromisepulmonaryresistance of death or premature euthanasia was recorded, and deaths that
to A. fumigatus. In the current study, we sought to test this occurredatnightwereassignedatimeofdeathmidwaybetween
hypothesis by inducing pulmonary A. fumigatus infection in mice the last evening observation and the first morning observation.
raised on vitamin D sufficient (VitD+) or vitamin D deficient Criteriaforprematureeuthanasiawerelaboredbreathing,a25%
(VitD-) diets. weight loss, and severe hypothermia (body temperature ,32uC).
Fungal burden in lungs was determined by a quantitative colony
Materials and Methods formingunits(CFUs)assay.Atselected timepointspostinfection
(p.i.) lungs were aseptically removed and their wet masses
Animals determinedusingaprecisionbalance(60.01g).Tissuespecimens
This study was carried out in strict accordance with the werehomogenizedusingamechanicalhomogenizerfor1 minata
recommendations in the Guide for the Care and Use of 1:5 dilution (w/v) in sterile PBS. Primary homogenate dilutions
Laboratory Animals of the National Institutes of Health. All were quantitatively cultured by serial dilution, plated in triplicate
mouse experiments were approved by the Animal Care and Use onSDAplates,incubatedat37uCfor24–48hrs,andthenumber
Committee at Nanjing University and Jingling Hospital under of CFUs per gram of tissue was enumerated. Results were
university-approvedstandards.Three-weekoldfemaleC57BL/6J expressed as log CFU per lung. Sections of formalin fixed lung
10
mice (Institute of Laboratory Animal Sciences, the Chinese were stained with hematoxylin and eosin (HE) for histological
Academy of Medical Sciences) were housed in groups of five analysis, with Grocott’s methenamine silver (GMS) for the
together on hardwood chip bedding in micro-isolator cages in detection of conidia and hyphae. Histopathologic scoring (HPS)
specific pathogen-free enviroment within small animal care was performed by a pathologist in a blinded manner. This HPS
facilities, at a room temperature of 22.561uC, and a relative system assigned values of 0–26 (the higher the score, the greater
humidity of 5265%. Mice were exposed to a 12hr light/dark the inflammatory changes) as described by Cimolai et al.[15].
cycleandhadadlibitumaccesstotheassigneddietandtapwater. Lungs were evaluated for inflammatory infiltrates and assigned a
Cage maintenance was performed twice weekly, and the animals cumulative score in the following five categories: (1) peribronchi-
were handled by the same individuals at all times. Mice were olar and bronchial infiltrates, (2) bronchiolar and bronchial
weighedweeklyandmonitoredforchangesinhealthstatus.After luminal exudates, (3) perivascular infiltrates, (4) the number of
oneweekacclimatization,themicewererandomizedtoVitD-diet neutrophils, and(5) parenchymal pneumonia.
(AIN-93G/No vitamin D) or VitD+ diet (AIN-93G) for at least
eight weeks. The vitamin D replete diet contained 1000IU Bronchoalveolar lavage (BAL) for cell recovery
vitamin D3/kg of food. All diets were procured from Research Airwaycontentswererecoveredby4roundsoffillingthelungs
Diets,Inc.(NewBrunswick,NJ).Everytwoweeks,threemicefrom with0.8 mlPBSandwithdrawingasmuchoftheliquidaspossible
each group were bled from the retro-orbital venous plexus. The througha20-gaugeneedleinsertedintothetracheaaftercervical
blood was centrifuged at 1,000–1,500rpm for 10min and the dissection. The bronchoalveolar lavage fluid (BALF) was centri-
serum collected for 25-OH VitD3 (25-(OH)D3) analysis. Serum fuged, thesupernatant wasremovedandimmediately frozenand
25-(OH)D3 levels were measured by ELISA kit (IDS) to confirm stored at 280uC. The recovered BAL cells were resuspended in
vitamin D deficiency. Ketamine-xylazine anesthesia was used for PBSandenumeratedunderalightmicroscopeinthepresenceof
retro-orbital venous plexus puncture. Under ketamine-xylazine trypan blueusing ahemocytometer.
anesthesia,animalsweresacrificedbycervicaldislocationforeach
experiment. All efforts were madetominimize animalsuffering. Morphological leukocyte differential analysis
Cells were visually counted by standard morphological criteria
Preparation of A. fumigatus
in Wright-Giemsa-stained samples of BAL cell suspensions
The strain of A. fumigatus was obtained from a patient with a cytospunontoglassslides. For Wright-Giemsastaining,theslides
fatal case of pulmonary aspergillosis at the Department of were fixed for 2 min with a one-step methanol-based Wright-
RespiratoryandCriticalCareMedicine,JinlingHospital,Nanjing Giemsa stain (Sigma) followed by steps 2 and 3 of the Diff-Quik
University School of Medicine. Conidia were harvested by wholebloodstainkit(BaxterScientific).Atotalof200to300cells
washing a 7 day-old slant culture on Sabouraud dextrose agar werecountedfromrandomlychosenhigh-powermicroscopefields
withPBSsupplementedwith0.1%Tween20.Thesuspensionwas foreach sample.
filteredthrougha40mmcellstrainer(Falcon)toseparateconidia
fromthecontaminatingmyceliumandtheabsenceofmyceliumin Isolation and stimulation of AMs
thefiltrate was verified microscopically.
AMswereharvestedfrommouselungswith0.8mlice-coldPBS
(10 times) through a 20-gauge needle inserted into the trachea
In vivo A. fumigatus infection, survival and tissue burden after cervical dissection. For stimulation assays, cells were
assessment, and histology separated from BALF by centrifugation at 3006g for 8 min at
Mice were lightly anesthetized with ketamine-xylazine and 4uCandsuspendedataconcentrationof26106/mlofRPMI1640
administered intratracheally with 2–76107 per 50ml viable supplemented with penicillin (100 U/ml), streptomycin (100 mg/
conidia while being held in a vertical position. Afterword, mice ml),and5%heat-inactivatedFBS.Aliquotsof250ml(56105cells)
wereplacedontheirbacksduringrecoveryfromanesthesia.After were added to 24-well plates, and allowed to adhere for 2 h at
inoculation, all animals fully recovered within 1–2hrs and were 37uCunderahumidifiedatmospherewith5%CO .Allwellswere
2
normalinappearanceuntilsignsofdiseasebecameapparent24– then washed three times with RPMI1640. Fresh RPMI 1640
PLOSONE | www.plosone.org 2 June2014 | Volume 9 | Issue 6 | e99805
RoleofVitDonHostResistancetoA.fumigatus
supplemented with 10% FBS was added prior to challenge with specificmouseELISAkits(R&D)accordingtothemanufacturers
conidia.AMswereincubatedinmediumaloneorinthepresence recommendations.
of viable A. fumigatus conidia at a microbe to cell ratio of 1:10
(MOI=0.1). After 6, 10, 14, and 18hr of co-culture, total RNA Statistical analyses
fromAMswasextracted,reverse-transcribedtocDNA,andstored Statistical analysis was performed using the Graph Pad Prism
at 220uC. 5.0software.Differencesofsurvivalrateswereanalyzedusingthe
Log-rank and Wilcoxon tests. Parametric data were expressed as
Flow cytometry analysis mean6SE and analyzed by t-tests. Non-parametric data were
CellularpopulationinBALFwerewashedandresuspendedata presented as median and analyzed using the Mann-Whitney test.
concentration of 0.56106 to 16106/50ml FA buffer (Difco) plus P-values,0.05wereconsideredsignificant.Allexperimentswere
0.1% NaN , and Fc receptors were blocked by the addition of performed intriplicate.
3
unlabeled anti-CD16/32(Fc blocking, eBioscience Inc.). After Fc
receptor blocking, immunostaining for cell surface molecules was Results
performed for 30min at 4uC. Cells were washed twice with FA
Vitamin D deficiency defected host resistance to A.
buffer,resuspendedin200mlof4%formalin(Sigma).Aminimum
fumigatus
of 10000 events were acquired on a FACSCanto flow cytometer
(BDPharmingen)usingCellQuestsoftware(BDPharmingen).The To investigate whether vitamin D deficiency alters host
data acquired were analyzed with FlowJo software (Tree Star, resistancetoA.fumigatus,wegeneratedC57BL/6Jmicethatwere
nutritionally deficient in vitamin D as previously described [16].
Stanford, CA). Fluorochrome-conjugated antibodies directed
After 8 wk on the diet, VitD- mice had 25-(OH)D3 levels of
against the following antigens were obtained: CD45-PE, Gr-1-
5.160.25ng/mLcomparedtoVitD+micewith25-(OH)D3levels
FITC, CD11c-APC, TLR2-FITC, TLR4-PE, (eBioscience Inc.),
of 82.0461.74 ng/mL (Table 1, p,0.0001). No significant
and Dectin-1-PE (R&D). Numbers of relevant cell types were
changes were observed in serum calcium and phosphorus
determined by combining flow cytometry data (percentage of a
concentrations at 8wk on the diet (data not show). Average
givencell type).
weight was similar in mice on VitD- and VitD+ chow over the
8 wk period (Table 1). Then, mice were infected intratracheally
Gene expression
with viable A. fumigatus conidia and parameters of resistance to
Gene expression was assessed by quantitative real-time PCR
infectionintermsofmortality,fungalgrowthandweightlosswere
with TaqMan probes specific for mouse GAPDH (GU214026.1),
evaluated.
TNF-a (NM_013693.2), IL-1b (NM_008361.3), IL-6 Infection with 56107 conidia resulted in significantly higher
(NM_031168.1), CXCL1 (NC_000004.12), CCL3
mortality in VitD- mice than in VitD+ mice (45%; 9/20 mice
(NC_000017.11), TLR2 (NM_011905.3), TLR4(NM_021297.2),
versus 15%; 3/20 mice, respectively) (Fig. 1A). Increasing the
and dectin-1 (NM_020008.2) obtained from Applied Biosystems. inoculumto76107resultedin61%mortalityinVitD-miceby4
RNA was isolated from lung tissue or cultured cells, and cDNA days (Fig.1B). Although infection with 26107conidia resulted in
synthesis was performed using the RevertAid First Strand cDNA nodeathineithergroupofmiceina28-daymortalityassay(data
Synthesis Kit (Thermo Scientific) according to manufacturer’s notshow),vitaminDdeficiencyimpairedpulmonaryclearanceof
instructions.Datawereexpressedusingthedelta-deltaCtmethod A. fumigatus, as determined by a significantly higher pulmonary
and normalized to the housekeeping gene GAPDH. Each PCR fungalloadinVitD-miceat24,48and72hrafterinfectionwith
run included a no-template control. 26107 conidia (Fig. 1C). Upon challenge with 26107 conidia,
weightlosswasmorepronouncedinVitD-mice,andweightgain
Western blot analysis inVitD-micewasfarlessthaninVitD+mice(Fig.1D).Allthese
Equal amounts of protein from tissue homogenates were date suggested that vitamin D deficiency resulted in a defective
electrophoresed through SDS-PAGE 8–18% gradient gels. The host pulmonaryresistance to A.fumigatus.
material in the gels was blotted onto a polyvinylidene difluoride
(PVDF) membrane (Bio-Rad), and the membrane was blocked Aggravated and sustained inflammatory response in A.
with 3% BSA in TBS buffer pH 7.5 for 2 hr. After washes (TBS fumigatus-challenged VitD- mice
pH 7.5,0.02%Tween20),immunolabelingwasperformedusing Histological evidence of aggravated and sustained
the primary antibody at 4uC overnight. The following primary inflammatory response in A. fumigatus-challenged VitD-
antibodies were used: anti-dectin-1 at 1:1000, anti-TLR2 at mice. HE staining of lung tissue sections revealed the presence
1:1500,andanti-TLR4at1:500(Abcam).Thenmembraneswere of inflammatory cells, mainly neutrophils, centered on bronchi/
incubated with the appropriate secondary antibodies (Cell bronchiolesandbloodvesselswithsignsofperibronchialnecrosis
Signaling, 1:2000) for 2h at room temperature. Anti-GAPDH inbothgroupsofmice(Fig.2A,upper,72hrafterchallengewith
(1:1000 Santa Cruz Biotechnology) was used as loading control. 56107 A. fumigatus conidia). The inflammatory lesions in VitD-
Afterwashing,proteinbandswerevisualizedwithChemilumines- miceweremoresevereanddiffused,withperibronchiolaredema,
cent HRP Substrate (Millipore) for 5 min at room temperature consolidationoftheinflammatoryresponse,anddisappearanceof
and exposed to X-ray film (Fuji Hyperfilm). Relative changes in alveolar structure (Fig. 2A). GMS staining of lung tissue sections
protein expression were estimated from the mean pixel density revealed A. fumigatus germinating conidia and septate-branching
using Quantity One software 4.6.2 (Bio-Rad), normalized to hyphae in VitD- mice as opposed to the relatively fewer
GAPHD andpresented asrelative density units. germinating conidia and no hyphae in VitD+ mice (Fig. 2A,
below).
Cytokine assays Infection with 26107 conidia resulted in no death in either
group of mice. So lung histopathology was analyzed sequentially
TheproteinlevelsofTNF-a,IL-1b,IL-6,CXCL1,andCCL3
(Fig.2B)andlunginflammationwasquantifiedbyHPS(Fig.2C)
in BALF were measured by commercially available cytokine-
afterchallengewiththisamountofinocula.By6 hr,lungsections
PLOSONE | www.plosone.org 3 June2014 | Volume 9 | Issue 6 | e99805
RoleofVitDonHostResistancetoA.fumigatus
e
er remained without significant histological changes. By 12hr and
w
ml) 24hr, early scattered areas of inflammation were observed, and
g/ were more apparent in VitD- mice. By 48hr and 72hr,
n
VitD- ***65.10.25 620.750.16 of25-(OH)D3( diitnnoiffrfltyaehrmeceenmVlclaiettisDnorf+iywltrleruarenetsipgoonnsneosactenatibdoinlnefs,iVbariwptiDnpitoa-hurmesnmietclxoeyu.redsBuaybtpisor7idonedfod,wu,tenhwrdeehiislnateinfllliadnmnfloamintmeatdemtnioaisnne-
ns VitD- mice. The HPS first increased at 12hr, peaked at 48hr,
o
ati and declined thereafter, and it was significantly higher in VitD-
ntr mice than in VitD+ mice at 48hr, 72hr and 7d p.i. (Fig. 2C).
e
eight VitD+ 682.041.74 620.650.17 heserumconc01(***). Twscuoihftufhuilcdsi,dehhenalitasrtydcoeolpyduanbtrhteeeosrfolopoluaugrnitctidos,aninanlaitnblhyoosVtiuhsigtrDhgerv-ogeeumarplmiescdieonmfacltoouinmrneggpsscaeoraventediardeltilaoitnhatfelhnaedimnirhdmyiVcapaithtitDoeande-
ndt,0.0 timesp.i. (no shown).
plexusat-testp lguantIengdcsrcoeefallsAne.udmfubimnerfiisgltaarntaudtsioc-conhmapololfesinitingofenldaimnVmiBtADaLt-oFmryaifctecere.lWAls.efiuinmnvigetashttuie-s
erdietaryrestriction. six VitDVitD-+ ***6675.61.218.560.21 6619.90.2719.20.2 erebledfromtheretro-orbitalvenousmicewasevaluatedusingaStudent’s ahscoocicntyhhnffeaftaamdAAlioBanllllmtMmoAieetncbnnhLemsgyaggettr(seeaoFcywe.taemi.loigimlMrtnlr.ehseyeot3siefcuuAWrc(relnno.et).eotmlrwBlsidNsgeeAhooVrixnLeitfpn-idtGocBciDCfshefiAus+eeelailrmglrLslleelnsaeFssnnii)nawnf,cidg(ceFetseashBrtiVdngaeewA.iitnwteBiLDd3r,iAiFeBetn-hsnL)uco,mtwrFbb2iefw6esisiaceecerhseroqde1eivncu0(ehbFe7sediiAnynisngwtu.tie.nlmmyafdut3stohmehBspreroaeip)ragcr.eahtbtcnedteAoaoududo.lssuommanfcgbulutbioymcysneeniodnairagifundglafonftttioeoualwtayrrssf,
aft pwof byanincreaseinneutrophils(Fig.3A,arrowindicated).BALcells
twodietarygroups VitD- ***612.30.32 618.70.28 hreemicefromeachgroubetweenthetwogroups wespV(iFnmoeeiisatgrteiDhke.naee3+rodsBpdB(ham)oFA.timilWiLg4sci.8eneas3manhfAatodrete)u.daplnaTry.idslm.lhb,bnepayteohnhtteowdodntcieawefiyenfluteandentrsrsiueoVcgnmacpricoetthbeeDauisdeltl+sredinrtoaihfifmtnrnhaoidnereVmdfVplilpatyrDimoot1Dibp-nm2eom-tashrmftitorciopouierc.nnyitec.od,oocamfeetil7nnxlpas2ecantueirhnhypteerrtdBooBpfAw6phAtLi.hhitiLlFh.ers,
htsofmiceinthe four VitD+ 663.21.02 618.30.26 twoweeksthereafter,tStatisticalsignificance fTisafneltnoalhdlwtBieinibwcaicaoGnsiyettdghertdio-dies1numtsr2tpaaoieComtgtnenraDegeyitsni1heoss1tpsfetac.pfiCmi2no.rDuieo(ndpvrr4gioapod5tphoua,iuosfsCtlllnicayonDeotgcirltiol1etcsnsp1aphocolobio,ronpttwabaeuntsdnilhneda)re.[etv1GidoaB7nrtb,Ai-1soy1L8noB]wFn,oAaafcsaLBopu.nGAasCsneLridse-Dt1slefm4/odo5Cfre+ocaDdferBis1lfllAfy1aesmcLrnuedrppncfhaltetohicolaltees-.l
SerumconcentrationsofVitaminDandweigTable1. twoweeks VitDVitD-+ ***6625-(OH)D3(ng/ml)43.40.6722.30.89 66Weight(g)16.80.3816.70.26 C57BL/6micewereraisedonVitDandVitD-dietsfromweaning.Every+6analyzed.ValuesrepresentmeanSE(n=3,repeatedthreetimeseach).doi:10.1371/journal.pone.0099805.t001 ipntcm,lmdn9c1tnw2penohey45+ieeoheuiftdeai2ltCtruu%Atfcailekhrceculh%ettlsssepDorlrierrrc,.htnoeBocoonooae1onypafppwAtpotdgVG1wattnchChufheheaLacgeidar)ilitstisehDn.slea-nhtlgasi1oslbytL4ifeeem4oi(naC+itiyo8Gosuryla5fnCtt,intfDmikl+r(n6nrmhhrGDsoBat-ac4eepBrmh1tcherApre1u5ihDey+Aep-porlai1t+L1Csltoptd,rr.nscLeeucilocoooa.rDs2(plrdyp/Fpoiecdoan(noal2te1Fouphcieanfeetgptf1orCiouhkisitdd.gluitcaacpuofskani(D.3il2onG.ntocgeafiiDdos3nnyln.y)1tecntuar,eeDitsye1)(cto-obmwodeBm.Fh1tucyn)fsesfF.[iAh2ehtmtfogsm1roitaioNiCnL.hondobc7lnaaraalphfeisoa3,oDnttrdct1lt.ihosCeewrdarwci9ed1nruiasobel8l)li]yp1iensi.tkGdnpla.f5n,elrcfoooiisgshWee%ncdecr2,enActra-eaGicldy)1esnaeigttlttrentalrnhoeh+wtroeeesa-feCadefc,bsaacts7f1emoleeustatisrDdlftsonetomcdoerlsiaAoAo/amtteo1awtB2aatn.m.ldimi1yiiclen5jeAftCoonisfcouashidruB0nenh.LrmlDeVpmme%liAta)yTnw.hi,iil1oci(gtoLtego.neihepiaae1Dbanruttoo(dtlcerhecstFlulck+ufoceenohsriontsAorubiptgebuehtayccvuu..kxyaoocsmayhnpeamefonpsh3trnutdednaearnbtcteoCamsbvidilldyseeos5nlliVeei,elnturug(%fnewgBerdnsrDairtnaGsstoAteeatrgfDnhugo)omsrooooater.Lsa)yeeer---tffff.,,
PLOSONE | www.plosone.org 4 June2014 | Volume 9 | Issue 6 | e99805
RoleofVitDonHostResistancetoA.fumigatus
Figure1.EnhancedsusceptibilityofVitD-miceafterchallengewithA.fumigatus.VitD+andVitD-micewerechallengedintratracheallywith
56107(A),or76107(B)A.fumigatusconidiainavolumeof50ml.Survivalwasmonitoredfor28days(n=7pergroup)andexpressedaspercentage
ofsurvival.Thisfigureillustratescumulativeresultsfromthreeindependentstudiesperinoculumdose.Thedifferencebetweensurvivalrateswas
significantatp,0.05(Log-ranktestandWilcoxontest).C.FungalgrowthinthelungsofVitD+andVitD-micechallengedintratracheallywith26107A.
fumigatusconidia.Miceweresacrificedattheindicatedtimesp.i.,andserialdilutionsoftissuehomogenateswereplatedforCFUdetermination.
Resultsareexpressedaslog ofCFUaveragevaluesdeterminedintriplicateofthreeindependentexperiments(threeforeachdilutionusingtwo
10
animalsofeachgroup).Eachdatapointrepresentsonemouse.DifferencesinCFUbetweenVitD+andVitD-miceweresignificantatp,0.05(*)and
p,0.01(**)(Mann–WhitneyUtest).D.Infection-inducedweightchangesinmiceweredetermineddailyafterintratrachealchallengewith26107A.
fumigatusconidia.Valuesrepersentmeans6SE(n=12).p,0.001(***),p,0.01(**),p,0.05(*)(pairedStudent’sttest).
doi:10.1371/journal.pone.0099805.g001
Enhanced pro-inflammatory cytokine and chemokine peaked at 48hr, returned to baseline at 7 d p.i.; while in VitD-
productioninthelungsofA.fumigatus-challengedVitD- micetheypeakedat72hr,andwerestillsignificantlyhigherthan
mice. We next investigated the kinetic production of pro- baseline at 7 dp.i.
inflammatory cytokine and chemokine in lung to better under- Because AMs are a critical source for pro-inflammatory
stand the inflammatory responses occurring after exposure. cytokine and chemokine production in response to A. fumigatus
Vitamin D deficiency significantly enhanced the basal expression [20], we isolated AMs from na¨ıve VitD+ and VitD-mice and
of IL-1b, IL-6, and TNF-a in the lungs of mice, while did not assessed the kinetic pro-inflammatory cytokine and chemokine
influencethebasalexpressionofCXCL1andCCL3(Fig.4A).A. production in response to live A. fumigatus conidia in vitro. Results
fumigatusinfectionelicitedarobustinflammatoryresponseinlung, showed that compared with VitD+ AMs, VitD- AMs had
characterized by the increased production of IL-1b, IL-6, and significantly higher basal levels of IL-1b, IL-6, and TNF-a
TNF-a. This response was more profound in the lungs of VitD- (Fig. 4B). In response to A. fumigatus conidia, VitD- AMs still had
mice at all the indicated times p.i. (Fig. 4A). Upon infection, the higherlevelsofthemversusVitD+AMsatalltheindicatedtimes.
levelsofthesecytokineswereelevatedfrom6 hron;inVitD+mice NodifferencesinthebasallevelsofCXCL1andCCL3expression
they peaked at 12hr, returned to baseline at 7 d; in contrast in were observed between VitD+ and VitD- AMs (Fig. 4B). In
VitD- mice, they peaked at 48hr, and were still significantly response to A. fumigatus conidia, VitD- AMs had significantly
elevated over baseline levels at 7 d p.i. VitD- mice had higher amounts of CXCL1 and CCL3 versus VitD+ AMs at
significantly higher amounts of CXCL1 and CCL3 versus VitD+ 10hr, 14hr and 18hr. Furthermore, the up-regulation of
mice at 12 to72hr, and 7 d p.i. (Fig. 4A). In VitD+ mice they cytokines and chemokines in VitD+ AMs peaked at 14hr,
PLOSONE | www.plosone.org 5 June2014 | Volume 9 | Issue 6 | e99805
RoleofVitDonHostResistancetoA.fumigatus
Figure2.HistologicalevidenceofaggravatedandsustainedinflammationinVitD-micechallengedwithA.fumigatusconidia.A.
Representativelung sections from VitD+ and VitD- mice threedays post intratracheal challenge with 56107 A. fumigatus conidia. Sections were
stainedwithHE(upper,magnification100x)foranalysisofinflammation,andwithGMS(lower,magnification400x)forthedetectionofconidiaand
hyphae. B. Representative HE-stained lung sections (magnification 100x) from VitD- and VitD+ mice are shown at the indicated times after
intratrachealchallengewith26107A.fumigatusconidia.C.HPSofserialHE-stainedlungsectionsbetween6hand7dafterintratrachealchallenge
with26107A.fumigatusconidia.HPSispresentedasmean6SE.Dataarerepresentativeofthreeindependentexperiments(n=3/group).Statistical
analysiswasperformedbyaMann-Whitneyranksumtest.p,0.001(***),p,0.01(**),p,0.05(*).
doi:10.1371/journal.pone.0099805.g002
decreasedrapidlythereafter;whileinVitD-AMstheirexpression and protein levels of these PRRs than VitD+ mice at all the
showeda continuously risingtrend for 18hr (Fig.4B). indicated times. In VitD+ mice, these PRRs peaked at 12hr,
TheseresultssuggestedthatvitaminDdeficiencyenhancedthe returned to baseline levels at 7 d p.i. While, in VitD- mice, they
basal levels of pro-inflammatory cytokines in the lungs of mice. peaked at 48hr, declined slowly thereafter, and were still much
UponA.fumigatusinfection,vitaminDdeficiencyledtoenhanced higherthan basal levels at 7 dp.i.
and sustained expression of pro-inflammatory cytokines and Because AMs are of the main cells that express PRRs and
chemokines bothinvivoandin vitro. recognize microorganism [20]. We investigated whether vitamin
Enhanced up-regulation of pattern recognition recep- DdeficiencyaffectstheexpressionofTLR2,TLR4,anddectin-1
tors (PRRs) both at baseline and post A. fumigatus on AMs at baseline by flow cytometry. We found that the
challeng in the lungs of VitD- mice. The initial sensing of expressionofthesePRRsonAMsisolatedfromna¨ıveVitD-mice
fungal infection is mediated by PRRs, such as TLRs and lectin wasup-regulatedsignificantlycomparedtoAMsfromna¨ıveVitD+
receptors,whichtriggerintracellularsignalingcascades,leadingto mice (Fig. 6A). We further investigated the expression of these
transcriptional expression of inflammatory mediators [21,22]. PRRsonAMsinresponsetoA.fumigatusinvitro.AMswereisolated
TLR2, TLR4 and _b-glucan receptor dectin-1 are key PRRs and co-cultured with viable A. fumigatus conidia. Gene expression
known to be involved in A. fumigatus recognition [21]. Hence, we of TLR2, TLR4, and dectin-1 was analyzed by q-PCR. Results
investigated whether vitamin D deficiency modulates the expres- showed that A. fumigatus challenge significantly increased the
sion of those PRRs in the lungs of mice (Fig. 5). Results showed expression of these PRRs in vitro (Fig. 6B). Similar to what was
thatvitaminDdeficiencyincreasedpulmonaryTLR2,TLR4and found in vivo, increased expression of these PRRs was more
dectin-1expressionatbaselinebothatmRNAandproteinlevels. pronounced in VitD- AMs compared with VitD+ AMs (Fig.6B).
Upon A. fumigatus challenge, these PRRs expression were up- The up-regulation of PRR genes in VitD+ AMs peaked at 14hr
regulated significantly, while VitD- mice still had higher mRNA p.i. and decreased rapidly thereafter, while PRR gene expression
PLOSONE | www.plosone.org 6 June2014 | Volume 9 | Issue 6 | e99805
RoleofVitDonHostResistancetoA.fumigatus
Figure3.IncreasedinfiltrationofinflammatorycellsinthelungsofA.fumigatus-challengedVitD-mice.A.BALcellswerestainedwith
Wright-Giemsastainandexaminedunderalightmicroscopeafterintratrachealchallengewith26107A.fumigatusconidia(magnification400x).B.
Total cell counts in BALF were enumerated on a hemacytometer of individual mice after intratracheal challenge with 26107 conidia. Data are
representative of three independent experiments with n=3 per group (mean6SE) and analyzed by paired Student’s t test. p,0.001(###), for
comparisonofA.fumigatus-challengedVitD+/VitD-miceversusbaselinelevels;p,0.001(***),p,0.01(**),p,0.05(*),forcomparisonofVitD+with
VitD-mice.C.BALcellswereanalyzedbyflowcytometryafterintratrachealchallengewith26107conidia.LavagecellsweregatedbyCD45+;the
resultingpopulationswerefurtheridentifiedonthebasisofCD11candGr-1expressionandsortedasshownbythedensityplots.WithintheCD45+
population,themajorcelltypesweremacrophages(CD11c+Gr-1+/2)atbaseline,neutrophils(CD11c2Gr-1+)becamethemajorcelltypes6–72hrpost
infection.FlowcytometrywasperformedtoquantifythepercentageofcelltypesinCD45+cells.D.Flowcytometrywasperformedtoquantifythe
percentageofneutrophilsinCD45+cellsinBALFafterintratrachealchallengewith26107conidia.Neutrophils(CD11c2Gr-1+)weregatedfromthe
CD45+cellsbasedonGr-1andCD11cexpressionprofiles.Dataarerepresentativeofthreeindependentexperimentswithn=3pergroup(mean6SE),
andanalyzedbypairedStudent’sttest.
doi:10.1371/journal.pone.0099805.g003
showed a continuously rising trend for 18hr in VitD- AMs presence of immunosuppressive drugs (cyclophosphamide, corti-
(Fig.6B). sone acetate, etc.) or during transient neutropenia (i.e., Ab-
Thus, vitamin D deficiency caused enhanced up-regulation of mediated depletion of neutrophils). Our position was that
TLR2, TLR4 and dectin-1 in the lungs of mice both at baseline induction of immunosuppression before infection might exagger-
and postA. fumigatus challenge. ate or underestimate the role of vitamin D whose importance in
lung clearance of A. fumigatus in immunocompetent mice was
Discussion clearly supportedby thedataof our experiments.
An important finding of this study was that the expression of
Our results showed that VitD- mice demonstrated higher pro-inflammatory cytokines (TNF-a, IL-1b and IL-6) and key
mortality after lung challenge with A. fumigatus coupled with PRRs involved in A. fumigatus recognition (TLR2, TLR4, and
impaired early clearance of the organism from the lungs. VitD-
dectin-1)wassignificantlyelevatedinthelungsandAMsofVitD-
mice demonstrated aggravated and prolonged histological evi- mice versus VitD+ mice already at baseline. Microbial infection
denceoflunginflammationaswellasenhancedBALcellcounts,
causes cytokine-associated inflammation to remove pathogens.
dominatedbyneutrophilsafterA.fumigatusinoculation.Increased
The initial sensing of infection is mediated by innate PRRs. The
basal levels of pro-inflammatory cytokines in the lungs and AMs
intracellularsignalingcascadestriggeredbyPRRsrecognitionlead
fromna¨ıveVitD-micewereobserved.UponA.fumigatusexposure,
to transcriptional expression of inflammatory mediators that
vitaminDdeficiencyledtoenhancedandsustainedexpressionof
coordinate the elimination of pathogens and infected cells [22].
pro-inflammatory cytokines and chemokines both in vivo and in Studieshaveconfirmedthatna¨ıvemicelackingdectin-1orTLRs
vitro. Up-regulation of key PRRs involved in A. fumigatus signaling were more sensitive to intratracheal challenge with A.
recognitionwasobservedinthelungsandAMsfromVitD-mice fumigatus than control mice, and demonstrated impaired pro-
both at baselineandafter A.fumigatus exposure. inflammatory cytokines production and insufficient lung neutro-
Our survival experiments were performed with inocula of 5– philrecruitment[24,25].MicelackingTNF-a,orCCR1(receptor
76107conidia, based on published reports using inocula at this for CCL3), were more susceptible to A. fumigatus lung infection,
levelwhenassessingmortalityandfungalclearanceinanimalsthat also through impaired neutrophil recruitment [26,27]. We thus
were not dually immunosuppressed with cyclophosphamide and expected that the increased basal levels of these PRRs and pro-
cortisone [23,24]. Our experiments were not conducted in the inflammatory cytokines in the lungs of VitD- mice, which led to
PLOSONE | www.plosone.org 7 June2014 | Volume 9 | Issue 6 | e99805
RoleofVitDonHostResistancetoA.fumigatus
PLOSONE | www.plosone.org 8 June2014 | Volume 9 | Issue 6 | e99805
RoleofVitDonHostResistancetoA.fumigatus
Figure4.Enhancedpro-inflammatorycytokineandchemokineproductioninthelungsofA.fumigatus-challengedVitD-mice.A.
Micewereinfectedintratracheallywith26107A.fumigatusconidia.Atspecifictimesp.i.BALFcollectedfrommicewasanalyzedforcytokineand
chemokine content by ELISA. Data are representative of three independent experiments with n=5 per group (mean6SE). p,0.001(###), p,
0.01(##),p,0.05(#),forcomparisonofA.fumigatus-challengedVitD+/VitD-miceversusbaselinelevels;p,0.001(***),p,0.01(**),p,0.05(*),for
comparisonofVitD+withVitD-miceusingpairedStudent’sT-test.B.AMsisolatedfromna¨ıvemouselungsbyBAL,wereincubatedinmediumalone
orinthepresenceofviableA.fumigatusconidia(MOI=0.1)invitro.TotalRNAfromAMswasextractedatindicatedtimes.Transcriptabundancewas
measured by real-time RT-PCR. Representative data from three independent experiments with n=3 per group are shown. P,0.001(###), P,
0.01(##), P,0.05(#), for comparison of A. fumigatus-challenged VitD+/VitD- AMs versus basal levels; P,0.001(***), P,0.01(**), P,0.05(*), for
comparisonofVitD+andVitD-AMsusingpairedStudent’sT-test.
doi:10.1371/journal.pone.0099805.g004
enhanced release of pro-inflammatory cytokines and chemokines immune reconstitution inflammatory syndrome [33]. The active
after lung challenge with A. fumigatus, accompanied by more formofvitaminDhasbeenprovedtoinhibitM.tuberculosis,virus
neutrophil infiltration in the lungs in our model, might increase and bacteria through the induction of the cathelicidin antimicro-
hostdefensetoA.fumigatus.However,thedatainourstudyshowed bial peptide and b-defensins[34,35], or autophagy[36], while
that up-regulation of them did not improve host resistance to A. decreasing the inflammatory response to microbial infec-
fumigatus. tions[37,38,39,40,41]. Upon in vitro LPS, C. albicans, and M.
Excessiveoruncontrolledreleaseofpro-inflammatorycytokines tuberculosis stimulation, the active form of vitamin D has been
can be fatal, which causes immunodeficiency, septic shock, or observedtoinhibittheproductionofTNF-a,IL-1b,andIL-6by
induction of autoimmunity, and may eventually impair disease monocytesdirectly,orthroughdown-regulating theexpressionof
eradication[22,28,29].InfluenzaAvirusandsomeGram-negative TLR2, TLR4 and dectin-1 on monocytes [39,40,41,42,43]. An
bacteria trigger life-threatening ‘‘cytokine storms’’ in the host increasedserumvitaminD3levelinsummerwasassociatedwitha
resulting in significant pathology and ultimately death [28,30]. It significant drop in pro-inflammatory cytokines production by
has been proposed that in invasive fungal infections, disease humanmonocytesinvitro,whencomparedtocytokineproduction
pathology may also be attributable to an aggravated or in winter [44]. In addition, the reduced expression of TLR2 and
dysregulated host inflammatory response that results in extensive TLR4onhumanmonocyteswasassociatedwithelevatedvitamin
tissue damage [14,29,31]. In mice with chronic granulomatous D levels during summer, and the reduction of TLR2 expression
disease, the intrinsic, genetically determined failure to control during summer was accompanied by a corresponding drop in
inflammation after exposure to sterile fungal components deter- TNF-a production by human monocytes in vitro [44]. The active
mines the animals’ inability to resolve an infection with A. vitamin D can be produced locally within the lungs. Vitamin D
fumigatus[32].Chronicdisseminatedcandidiasis(CDC)istypically deficiency resulted in decreased synthesis of active vitamin D in
observed during neutrophil recovery in patients with acute lung.Therefore,itispossiblethatlackofvitaminDincreasedthe
leukemia, and the efficacy of corticosteroid therapy supports the basallevelsofpro-inflammatorycytokinesandPRRsinthelungs
pathophysiological hypothesis that CDC is a fungus-related
Figure 5. Enhanced and prolonged up-regulation of PRRs in the lungs ofA.fumigatuschallenged VitD- mice. Mice were infected
intratracheallywith26107A.fumigatusconidia.Atspecifictimep.i.,miceweresacrificedandlungswereexcised,andtotalRNAandproteinwas
extracted.Expressionofdectin-1,TLR2andTLR4atthemRNAlevelwasquantifiedbyreal-timeRT-PCR,andproteinlevelswerequantifiedbyWestern
blotting (shown in middle panel). Representative data from three independent experiments are shown with n=3 per group (mean6SE). p,
0.001(###),p,0.01(##),p,0.05(#),forcomparisonofA.fumigatus-challengedVitD+/VitD-versusbaselinelevels;p,0.001(***),p,0.01(**),p,
0.05(*),forcomparisonofVitD+andVitD-miceusingpairedStudent’sT-test.
doi:10.1371/journal.pone.0099805.g005
PLOSONE | www.plosone.org 9 June2014 | Volume 9 | Issue 6 | e99805
RoleofVitDonHostResistancetoA.fumigatus
Figure6.Enhancedandprolongedup-regulationofPRRsonA.fumigatus-stimulatedVitD-AMs.A.ExpressionofsurfaceTLR2,TLR4and
dectin-1onAMsisolatedfromVitD-orVitD+mice.AMswereisolatedfrommouselungsbyBAL,andstainedforCD11c,dectin-1,TLR2orTLR4and
thenanalyzedusingflowcytometry.Thedatashowtheexpressionofdectin-1,TLR2andTLR4onCD11c-positivecells(redline,isotypecontrol;blue
line,VitD+AMs;greenline,VitD-AMs).Thevaluesshownintheflowcytometryprofilesarethemeanfluorescenceintensity(MFI)indices.Dataare
representativeofthreeindependentexperimentswithn=3/group.p,0.01(**),p,0.05(*)usingpairedStudent’sT-test.B.mRNAexpressionofTLR2,
TLR4anddectin-1inAMsafterco-culturedwithliveA.fumigatusconidiainvitro.AMswereisolatedfromna¨ıvemouselungsbyBAL,incubatedin
medium alone or in the presence of viable A. fumigatus conidia (MOI=0.1). Total RNA from AMs was extracted at indicated times. Transcript
abundance was measured by real-time RT-PCR. Representative data from three independent experiments with n=3 per group are shown. P,
0.001(###),P,0.01(##),P,0.05(#),forcomparisonofA.f.-challengedVitD+/VitD-versusbaselinelevels;P,0.001(***),P,0.01(**),P,0.05(*),for
comparisonofVitD+andVitD-usingpairedStudent’sT-test.
doi:10.1371/journal.pone.0099805.g006
ofmice,whichsubsequentlyledtoahyper-inflammatoryresponse Moreover,ourexperimentswereconductedinimmunocompe-
toA.fumigatusinlungandimpairedhostresistancetoA.fumigatus. tentmice.Inthiscase,theincreasedfungalburdeninthelungsof
Neutrophilsareessentialininitiationandexecutionoftheacute VitD- mice might further enhance the PRRs signal, and
inflammatory response in fungal infection [45]. 1 alpha, 25- exacerbated the values of inflammatory mediators, which aggra-
Dihydroxyvitamin D3 was demonstrated to inhibit neutrophil vatedtheinflammationinthelungs,andeventuallyledtoapoor
motivation by decreasing ICAM-1 and ELAM-1 expressions on prognosis.
pulmonary microvascular endothelial cells [46], and inhibited Human diseases caused by A. fumigatus have been considered
neutrophil recruitment in lung in animal models of acute lung underthreecategories:invasiveinfections;infectionscausedbythe
injury[47,48]. In addition, the active form of vitamin D was colonizationofmucosalsurfaceswithoutinvasionintotissue;and
observed tosuppressIL-8 productioninPseudomonas-stimulated hypersensitivitydiseases.Thekeydeterminantofthepathogenicity
neutrophils and have benefit for resisting persistent P. aeruginosa of Aspergillus species, and the reason for the diversity of host
lower respiratory tract infection[49].Thus, vitamin Ddeficiency outcomes, is hypothesized to be the nature of the immune
mightleadtomoreneutrophilsinfiltrationinlunginresponsetoA. responseofthehost[50],andthediseasesmaybeconceptualized
fumigatus challenge, which aggravated pulmonary inflammatory aspointsalongaspectrumofabnormalimmuneresponsesofthe
response. host. Vitamin D may have a role in inhibiting A. fumigatus
infections, while modulating and limiting any unnecessary
PLOSONE | www.plosone.org 10 June2014 | Volume 9 | Issue 6 | e99805