The abundance of health-associated bacteria is altered in PAH polluted soils - Implications for health in urban areas

RESEARCH ARTICLE The abundance o heal h-associa ed bac e ia is al e ed in PAH pollu ed soils—Implica ions o heal h in u ban a eas? Ani ud a Pa ajuli 1 , Mi a G o ¨n oos 1 , Sa i Kauppi 2 , Tomasz Płociniczak 3 , Ma ja I. Roslund 1 , Polina Gali skaya 4 , Olli H. Lai inen 5 , Heikki Hyo ¨ y 5,6 , A i Jumpponen 7 , Rauni S o ¨mme 1 , Ma in Roman schuk 1,4 , Nan Hui 1 *, Aki Sinkkonen 1,4 * 1Depa men o En i onmen al Sciences, Sec ion o En i onmen al Ecology, Uni e si y o Helsinki, Lah i, Finland, 2Finnish En i onmen Ins i u e, SYKE, Cen e o Sus ainable Consump ion and P oduc ion, Con aminan s, Helsinki, Finland, 3Depa men o Mic obiology, Uni e si y o Silesia, Ka owice, Poland, 4Kazan Fede al Uni e si y, Kazan, Russia, 5Depa men o Vi ology, School o Medicine, Uni e si y o Tampe e, Tampe e, Finland, 6Fimlab Labo a o ies, Pi kanmaa Hospi al Dis ic , Tampe e, Finland, 7Di ision o Biology, Kansas S a e Uni e si y, Manha an, Kansas, Uni ed S a es o Ame ica *aki.sinkkonen@helsinki. i (AS); [email p o ec ed] (NH) Abs ac Long- e m exposu e o polya oma ic hyd oca bons (PAHs) has been connec ed o ch onic human heal h diso de s. I is also well-known ha i) PAH con amina ion al e s soil bac e ial communi ies, ii) human mic obiome is associa ed wi h en i onmen al mic obiome, and iii) al e a ion in he abundance o membe s in se e al bac e ial phyla is associa ed wi h ad e se o bene icial human heal h e ec s. We hypo hesized ha soil pollu ion by PAHs al e ed soil bac e ial communi ies ha had known associa ions wi h human heal h. The a ionale behind ou s udy was o inc ease unde s anding and po en ially acili a e econside ing ac o s ha lead o heal h diso de s in a eas cha ac e ized by PAH con amina ion. La ge con aine s illed wi h ei he sp uce o es soil, pine o es soil, pea , o glacial sand we e le o incuba e o con amina ed wi h c eoso e. Biological deg ada ion o PAHs was moni o ed using GC- MS, and he bac e ial communi y composi ion was analyzed using 454 py osequencing. P o eobac e ia had highe and Ac inobac e ia and Bac e oide es had lowe ela i e abun- dance in c eoso e con amina ed soils han in non-con amina ed soils. Ea lie s udies ha e demons a ed ha an inc ease in he abundance o P o eobac e ia and dec eased abun- dance o he phyla Ac inobac e ia and Bac e oide es a e pa icula ly associa ed wi h ad e se heal h ou comes and immunological diso de s. The e o e, we p opose ha pollu- ion-induced shi s in na u al soil bac e ial communi y, like in PAH-pollu ed a eas, can con- ibu e o he p e alence o ch onic diseases. We encou age s udies ha simul aneously add ess he classic “ad e se oxin e ec ” pa adigm and ou no el “al e ed en i onmen al mic obiome” hypo hesis. PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187852 No embe 16, 2017 1 / 18 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Ci a ion: Pa ajuli A, G o¨n oos M, Kauppi S, Płociniczak T, Roslund MI, Gali skaya P, e al. (2017) The abundance o heal h-associa ed bac e ia is al e ed in PAH pollu ed soils— Implica ions o heal h in u ban a eas? PLoS ONE 12(11): e0187852. h ps://doi.o g/10.1371/jou nal. pone.0187852 Edi o : Ma ie-Joelle Vi olle, Uni e si e Pa is-Sud, FRANCE Recei ed: Augus 12, 2017 Accep ed: Oc obe 29, 2017 Published: No embe 16, 2017 Copy igh : ©2017 Pa ajuli e al. This is an open access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal au ho and sou ce a e c edi ed. Da a A ailabili y S a emen : The bac e ial sequence da a a e a ailable in he Sequence Read A chi e a NCBI unde accession numbe SRR5229978. All o he ele an da a a e wi hin he pape and i s Suppo ing In o ma ion iles. Funding: The esea ch was unded by Finnish Funding Agency o Technology and Inno a ionTekes (decision numbe 40333/14), h ps://www. ekes. i/en/. The unde s had no ole in In oduc ion Pollu an s, such as polya oma ic hyd oca bons (PAHs), can lead o shi s in mic obial commu- ni ies [1–4]. PAH deg ada ion ypically las s o decades in pollu ed en i onmen s including soil, wa e , ai , and sedimen s [5–7]. Di ec PAH oxici y, abio ic ans o ma ion, and mic o- bial deg ada ion a o some mic obial axa, whe eas o he s become less p e alen in PAH-pol- lu ed soils [8]. The bac e ial axa ha h i e unde PAH exposu e a y depending on he ypes o soil and en i onmen al condi ions [9–12]. A sea ch o gene al ends in mic obial shi s in PAH-pollu ed soils necessi a es concu en s udies o mic obial communi ies in di e en soil ypes. Exposu e o PAHs is associa ed wi h se e e human heal h de ici s and ecological impac s, such as p ocess o PAH-DNA adduc o ma ion and ca cinogenesis [13], and he e o e hey ha e been classi ied as p io i y en i onmen al pollu an s by he Uni ed S a es En i onmen al P o ec ion Agency (US-EPA) and he En i onmen al Eu opean Agency (EEA) [14,15]. The en i onmen al concen a ion o PAHs and he connec ion be ween di ec PAH exposu e and human heal h ha e been well in es iga ed [16–23]. Almos 90% o he PAHs eleased in o he a mosphe e accumula e in su ace soil laye s [14] whe e hey a e p ima ily deg aded by soil bac e ia o bind o soil pa icles [22]. The heal h ou comes o PAH-exposu e ha e been a ib- u ed o di ec and indi ec oxin e ec s on humans [23,24]. Se e al s udies ha e in es iga ed he ela ionship be ween human heal h and he mos common bac e ial axa ound in he human mic obiome. A ich o dominan P o eobac e ial and diminished Bac e oide es communi ies in he human gu can con ibu e o a isk o immune sys em diso de s including ch onic obs uc i e pulmona y disease and as hma [25], and such e ec s can be ansgene a ional. Fo example women whose in an s de eloped IgE- associa ed eczema had lowe di e si y o Bac e oide es in hei gu du ing he p egnancy [26]. Al e a ion in he P o eobac e ial abundance in a ious egions o human body is associ- a ed wi h se e al heal h diso de s, bu i s e ec s a e wo old. On he posi i e side, di e se P o eobac e ial communi y on skin is ela ed o a educed isk o a opy [27,28]. On he neg- a i e side, o e g ow h o P o eobac e ia is connec ed o as hma and ch onic obs uc i e pul- mona y disease [29]. Wi hin P o eobac e ia, Be ap o eobac e ia a e en iched in he gu mic obiome o indi iduals wi h ype 2 diabe es [30]. Impo an ly, ecen indings indica e ha P o eobac e ial communi y o indi idual’s skin depends on land use in hei li ing en i- onmen and ha he en i onmen unes esponse o alle gens [27–31]. Soil is he majo es- e oi o Ac inobac e ia, Bac e oide es, and P o eobac e ia [32,33], and hus he li ing en i onmen likely unes indi idual’s esponses o alle gens, suga me abolism, and heal h diso de s. Fo hese easons, i is possible ha ad e se heal h e ec s connec ed o li ing in con amina ed en i onmen may pa ly be a ibu able o undamen al changes in he en i onmen al mic obiome people a e exposed o, in addi ion o he di ec physiological esponses caused by he oxins. Su p isingly, o he bes o ou knowledge, no s udies ha e ocused on he po en ial connec ion be ween PAH pollu ion and soil communi y changes in he abundance o heal h-associa ed bac e ia. In he p esen s udy, we sampled ou di e en su ace soils, con amina ed hem wi h PAHs and ollowed popula ion shi s in bac e ial communi y in con amina ed and non-con- amina ed con aine s unde con olled condi ions. Finally, based on p e ious empi ical e i- dence as well as ou new indings and no el ideas, we p opose a new “al e ed en i onmen al mic obiome” hypo hesis ha should be in es iga ed in pa allel wi h he adi ional “di ec oxin e ec ” pa adigm as a po en ial explana ion o he complex ela ionship be ween human heal h and en i onmen al pollu ion. PAH pollu ion al e s he abundance o heal h-associa ed bac e ia PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187852 No embe 16, 2017 2 / 18 s udy design, da a collec ion and analysis, decision o publish, o p epa a ion o he manusc ip . Compe ing in e es s: The au ho s ha e decla ed ha no compe ing in e es s exis . Ma e ials and me hods PAH sou ce We selec ed c eoso e as he con aminan because i is among he mos widely used wood p e- se a i es wi h a his o y o mo e han a cen u y in wood imp egna ion indus y [5]. C eoso e consis s o 85% PAHs which a e chemical compounds consis ing o a leas wo a oma ic ings used oge he [34,35]. The emaining is less han 10% phenolic compounds and 5–10% he - e ocyclic a oma ic compounds consis ing o oxygen, sulphu , and ni ogen. The e o e, c eo- so e is a ele an compound o es en i onmen al changes caused by PAH-pollu ion since i consis s mainly o PAHs and is a common p oblem in u ban a eas wi h a his o y o sawmills. Soil collec ion Su ace soil was collec ed a ou sepa a e si es in Finland. The i s si e (Pea in Table 1) was a Haapasuo pea p oduc ion a ea in Lei onma¨ki, Finland [36] (61˚54’N 26˚4’E). The cu en su ace laye was o med hund eds o yea s ago, and i ep esen s he ansi ion laye be ween mine o ophic Ca ex and omb o ophic Sphagnum domina ed pea bogs. The o iginal bog was d ained, and he wa e able has been kep 30–40 cm below he pea su ace o mo e han h ee decades. The second si e, e e ed o as pine o es soil, is a bo eal pine (Pinus syl es is L.) o es in Hollola, loca ed in sou he n Finland (61˚0’N 25˚29’E) mo aine idge ( he undis- u bed ecosys em has been desc ibed be o e [36]). The si e is cha ac e ized by a hin o ganic soil laye on op o mine al soil [37]. The hi d si e (Sp uce o es soil in Table 1) is a sp uce o es in Vie uma¨ki, sou he n Finland (60˚52’7N 25˚41’E). Sp uce (Picea abies) is he domi- nan ee species, and he ield laye consis ed o Sphagnum and Pleu ozium sch ebe i mosses as well as dwa sh ubs, mainly Vaccinium species. The ou h si e (sand) was nex o (dis ance 50 m) he si e 3 (Sp uce o es soil), bu i consis ed o ba e glacial sand and sca e ed ude al he bs as o iginal su ace soil was emo ed yea s ea lie when he si e became a s o age ield o lumbe . A all ou si es, soil was collec ed om h ee sepa a e (dis ance >5m) ca. 1 m 2 plo s as desc ibed in ea lie wo k [38,39]. In sho , a each plo , li e ege a ion and plan deb is we e emo ed and 15–20 L o su ace soil (dep h 2–15 cm) was collec ed and mixed ho oughly. The soil was hen andomly di ided in o wo 10 L polye hylene con aine s, and he p ocedu e was epea ed a each 1 m 2 plo . Filled con aine s we e co e ed wi h polye hylene lids wi h wo 5 mm diame e holes sealed loosely wi h co on wool o acili a e passi e ae a ion. Soil weigh in con aine s a ied be ween 3–12 kg depending on soil ype, he hea ies being he mine al soil- ypes. Basic physicochemical analyses. Mois u e con en was measu ed by d ying samples in an o en (+ 90˚C) o 24 hou s. O ganic ma e (OM) was de e mined as loss o igni ion a 550˚C o 4h. Nu ien con en s o he di e en soil ypes we e de e mined wi h QuikChem 8000 low injec ion analysis sys em (LACHAT Ins umen s Inc., USA) [38]. To measu e he pH, 10g ( esh weigh ) o each soil ype was mixed in 50 mL o 1M CaCl 2 , shaken o 5 min and allowed o se le o 2–24 hou s [38]. Table 1. The pH and o ganic ma e con en o each soil ype. Pea Pine o es Sp uce o es Sand pH 3.52 ±0.03 b 3.6 ±0.26 b 3.25 ±0.04 c 4.87 ±0.05 a O ganic ma e (%) 97.75±0.41 a 63.42 ±8.15 b 35.19 ±12.41 c 1.21 ±0.17 d Values a e mean ±1 SD. Le e s in he supe sc ip deno e s a is ical di e ences in Tukey’s es s among di e en soil ypes h ps://doi.o g/10.1371/jou nal.pone.0187852. 001 PAH pollu ion al e s he abundance o heal h-associa ed bac e ia PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187852 No embe 16, 2017 3 / 18 PAH deg ada ion expe imen . Fo each soil ype, h ee pai s o 10L polye hene con ain- e s we e included. In he beginning o he expe imen , soil was mixed ho oughly wi hin pai s in la ge polye hene con aine s and he ea e di ided again in o wo 10L polye hene con ain- e s. O each pai , one con aine was andomly chosen o be spiked wi h c eoso e, whe eas ano he ecei ed 100g glacial sand (Lohja Rudus Oy, Lah i, Finland). Ini ial (day 0) samples o basic chemical analyses we e acqui ed immedia ely a e spiking. C eoso e spiking was done as desc ibed ea lie [38]. In sho , c eoso e (6–24 g) was mixed ho oughly wi h 100g glacial sand, and he mix u e was inco po a ed in o soil in he con aine s selec ed o con amina ion. This esul ed in he concen a ion o ~ 1% PAHs simila o hose in c eoso e con amina ed si es. A e c eoso e addi ion, con aine s we e le o incuba e o ou weeks (28 days) a 16 ±1˚C and sampled a days 28, 91, and 189 o chemical and bac e ial communi y analyses. Each sam- ple ( o al weigh 10 g) consis ed o i e subsamples om ou dep hs (2, 5, 10, and 15 cm) col- lec ed om each con aine . Communi y sequencing analyses we e pe o med on day 0 and when he o al PAH concen a ion had dec eased by mo e han 20% om week ou (day 28) alue. The e o e, week 13 (day 91) samples we e analyzed in mos cases. Howe e , week 27 (day 189) samples we e u ilized o each o con ol and con amina ed mine al soil con aine s because hei PAH concen a ions a week 13 we e mo e han 95% o week 4 concen a ions. To a oid he ans e o mic oo ganisms be ween con aine s, he sampling equipmen was ca e ully lame s e ilized wi h 70% e hanol igh be o e aking each sample. PAH analyses PAH concen a ions we e de e mined using oluene ex ac ion as desc ibed in No dic Guide- lines o Chemical Analysis o Con amina ed Soil Samples [40] and analyzed as desc ibed ea - lie [41] wi h he excep ion ha aqueous sodium py ophospha e decahyd a e solu ion (0.05 M) ins ead o hexane was used. The ex ac s we e analyzed wi h Shimadzu GC–MS-QP5000 sys em equipped wi h AOC-20i au oinjec o and 30-m ZB-5MS column (0.25 mm i.d., 0.25 μm ilm hickness). The o en p og am was se as ollows: 80˚C o 1 min, 10˚C/min o 250˚C, 7˚C/min o 280˚C, 20˚C/min o 320˚C wi h a hold o 10 min o a o al un ime o 34.29 min. PAH-mix 9 (16 PAHs included) was used o GC-MS and PAH-Mix 31 o i e deu- e a ed PAHs o soil samples (D . Eh ens o e , GmbH Ge many) as s anda ds and An h a- cene-D10 (D . Eh ens o e , GmbH Ge many) as a eco e y s anda d. DNA ex ac ion, ampli ica ion, and sequencing To al DNA was ex ac ed om each soil sample using Fas DNA SPIN Ki o Soil (MP Biome- dicals, Illki ch, F ance) acco ding o he manu ac u e ’s s anda d p o ocol. The highly hype - a iable egion V3 o bac e ial 16S RNA gene was ampli ied using p ime cons uc s ha inco po a ed he py osequencing adap e s (A), sample-speci ic DNA ags, and MF341 5’ CTA CGG GAG GCA GCA G 3’ o R518 50ATT ACC GCG GCT GCT GG 30[42]. The PCR we e conduc ed unde he ollowing condi ions: 200 nM o each o wa d and e e se p ime s, 5 ng empla e DNA, 200 μM o each dNTP, 2.5 mM MgCl 2 , 1 U GoTaq Ho S a DNA polyme ase (P omega, Madison, WI), and 2.5 μl PCR bu e . The PCR cycle pa ame e s consis ed o an ini ial dena u a ion a 94˚C o 3 min, hen 25 cycles o dena u a ion a 94˚C o 1 min, annealing a 54˚C o 1 min, and ex ension a 72˚C o 2 min, ollowed by a inal ex ension s ep a 72˚C o 10 min. The PCR p oduc s we e pu i ied using Ampu e XP Mag- ne ic Clean-up (Agencou Bioscience Co po a ion, Be e ly, MA, USA), and quan i ied using Nanod op (The mo Scien i ic, Rock o d, IL, USA) and Bioanalyze 2100 wi h DNA 1000 chips (Agilen Technologies Inc., San a Cla a, CA, USA). The sequencing was pe o med using he 454 GS FLX p o ocol and he GS FLX Ti anium Rapid Lib a y P epa a ion Ki (454 PAH pollu ion al e s he abundance o heal h-associa ed bac e ia PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187852 No embe 16, 2017 4 / 18 Li e Sciences, Roche Diagnos ics, CT, USA). The bac e ial sequence da a a e a ailable in he Sequence Read A chi e a NCBI unde accession numbe SRR5229978. Sequence analysis. The sequence da a we e analyzed using MOTHUR ( 1.35.0, 64-bi o Linux) acco ding o he s anda d ope a ing p o ocol as desc ibed ea lie [43]. B ie ly, he aw sequence da a we e quali y con olled, and eads wi h ambiguous bases o homopolyme s lon- ge han 8 bp (722 sequences) we e emo ed. The UCHIME algo i hm [44] iden i ied 943 sequences as chime ic, and hese we e subsequen ly omi ed. Sequences we e assigned in o Ope a ional Taxonomic Uni s (OTU) a 97% simila i y and OTUs assigned o axon a ini ies using Naï e Bayesian Classi ie [45] agains he RDP aining se ( e sion 10). Ra e OTUs occu ing h ee o ewe imes ac oss all samples we e omi ed o a oid p oblems caused by unce ain y in o igin. S a is ical analyses. The ela i e abundance was calcula ed as he numbe o sequences in a axon di ided by o al numbe o sequences in a sample. To compa e con amina ed and p is- ine ea men s, - es on ela i e abundances o bac e ial phyla was conduc ed. Signi ican esul s a e indica ed by as e isks (p<0.05). False De ec ion Ra e (FDR) was used o he p- al- ues co ec ion in - es s. The di e ence in he pH and he o ganic ma e con en ac oss he ou soil ypes was calcula ed using ANOVA in JMP ( .11.0 64-bi ; SAS Ins i u e, Ca y, No h Ca olina). To isualize bac e ial communi y composi ions o whole bac e ial communi ies as well as he majo bac e ial phyla, non-me ic mul idimensional scaling (NMDS) analyses was pe o med based on ela i e abundance o OTUs using egan package in R ( 3.2.2, R De elop- men Co e Team 2015). The B ay-Cu is dis ance was chosen in he NMDS analysis since he e we e null alues be ween samples in he da a [46]. We pe o med he NMDS analyses a 99, 97, 95, 93, 91% OTU simila i y le els bu only p esen he da a a 97% because OTU h esh- old had no impac on he o e all conclusions. Di e ences in he bac e ial communi y composi ion be ween c eoso e con amina ed and con ol g oups we e es ed using pe mu a i e analysis o a ia ion (PERMANOVA, unc ion adonis in R-package egan and B ay-Cu is me ic). The di e ence in he ela i e abundances o bac e ial phyla be ween c eoso e con amina ed and con ol soils was calcula ed using he T- es in JMP. Resul s Soil chemical cha ac e is ics The ou soil ypes used in his s udy di e ed in pH (F = 164.4, d = 3, 20, p<0.001) and o ganic ma e con en (F = 183.2, d = 3, 20, p<0.001). Sandy soil had he highes and sp uce o es soil he lowes pH (Table 1). O ganic ma e con en dec eased in he ollowing o de : pea >pine o es soil >sp uce o es soil >sandy soil (Table 1). Thus, he ou soil ypes ep- esen a ange o habi a s wi h highly a iable bac e ial communi ies. Cha ac e is ics o bac e ial ope a ional uni s (OTUs) and ela i e abundances o bac e ial phyla The py osequenced bac e ial 16S RNA gene agmen da ase s we e analysed om 24 soil samples aken on days 0, 91, and 189. Day 0 samples o mic obiological analyses we e aken be o e con amina ion. Then he ime poin a which 20% o mo e o PAH concen a ion on day 28 had disappea ed was de e mined. In sand, 20% deg ada ion was eached on day 189 and in o he soils on day 91 (S1 Table). Day 28 was used as a e e ence day because he educ- ion in PAH concen a ion un il day 28 is ypically caused by he e apo a ion o naph halene and o he low molecula weigh compounds [47,48]. PAH pollu ion al e s he abundance o heal h-associa ed bac e ia PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187852 No embe 16, 2017 5 / 18 In o al 3626 OTUs we e ob ained, which ep esen ed mo e han 20 bac e ial phyla om di e en soil ypes wi h and wi hou PAH con amina ion. Bac e ial OTUs ep esen ing P o- eobac e ia we e he mos abundan g oup accoun ing o 50.9%, 31.5%, and 35.8% o he o al sequences in con amina ed, p is ine, and day 0 samples, espec i ely (Table 2). Likewise, P o- eobac e ia was he p edominan phylum when compa ed ac oss he di e en soil ypes wi h 46.20% (pea soil), 37.7% (pine o es soil), 36.0% (sand), and 37.8% (sp uce o es soil) o all sequences. O he dominan phyla in all soils we e Acidobac e ia, Ac inobac e ia, Ve ucomi- c obia, and Bac e oide es (Tables 2and 3). The ela i e abundances o di e en phyla be ween con amina ed and p is ine ea men s we e compa ed. P o eobac e ia (d = 1, p= 0.046) we e mo e abundan in con amina ed soils, whe eas he ela i e abundances o Ac inobac e ia (d = 1, p= 0.043) and Bac e oide es (d = 1, p= 0.005) declined in con amina ed soils (Table 2). In addi ion, i was obse ed ha he ela- i e abundances o Pa cubac e ia and Candidus Saccha ibac e ia we e lowe in he con ami- na ed soils, al hough he ela i e abundances o hese wo we e e y low (Table 2). As axa belonging o P o eobac e ia ep esen ed mo e han 35% o mic obial communi y in all soil ypes, we compa ed he ela i e abundances o he classes unde he phylum P o eobac e ia Table 2. Rela i e abundances o bac e ia phyla (classes) on day 0 and when 20% o PAH con amina ion had disappea ed, i.e. days 91 and 189 in sand and o he soils (bo h p is ine and con amina ed), espec i ely. Week 0 No c eoso e C eoso e added Phylum Mean 1SD Mean 1SD Mean 1SD Acidobac e ia 0.240 0.110 0.269 0.089 0.268 0.170 Ac inobac e ia 0.150 0.060 0.092*0.041 0.049*0.016 A ma imonade es 0.001 0.001 0.002 0.002 0.000 0.000 Bac e oide es 0.048 0.026 0.069*0.034 0.010*0.013 Candida e_di ision_WPS-1 0.000 0.000 0.001 0.002 0.000 0.000 Candida us_Saccha ibac e ia 0.014 0.010 0.013*0.006 0.002*0.003 Chlamydiae 0.001 0.001 0.004 0.007 0.001 0.002 Chlo o lexi 0.002 0.003 0.002 0.004 0.001 0.001 Fib obac e es 0.001 0.003 0.002 0.003 0.000 0.000 Fi micu es 0.008 0.010 0.015 0.015 0.031 0.062 Gemma imonade es 0.030 0.060 0.007 0.013 0.001 0.003 Ni ospi ae 0.001 0.004 0.002 0.003 0.001 0.002 Pa cubac e ia 0.001 0.001 0.003*0.003 0.001*0.000 Planc omyce es 0.004 0.005 0.005 0.005 0.004 0.006 P o eobac e ia 0.358 0.077 0.315*0.071 0.509*0.166 Alphap o eobac e ia 0.100 0.026 0.120 0.033 0.127 0.075 Be ap o eobac e ia 0.165 0.035 0.082*0.032 0.234*0.057 Gammap o eobac e ia 0.043 0.021 0.054 0.019 0.076 0.018 Del ap o eobac e ia 0.047 0.015 0.057 0.041 0.061 0.022 Unclassi ied p o eobac e ia 0.004 0.002 0.003 0.001 0.010 0.003 Spi ochae es 0.003 0.008 0.009 0.022 0.000 0.001 Unclassi ied bac e ia 0.081 0.070 0.130 0.107 0.049 0.061 Ve ucomic obia 0.040 0.030 0.060 0.035 0.074 0.054 To compa e con amina ed and p is ine ea men s, we conduc ed - es be ween c eoso e added and no c eoso e. Signi ican di e ences indica ed by as e isk (p<0.05). P- alues co ec ed by FDR. h ps://doi.o g/10.1371/jou nal.pone.0187852. 002 PAH pollu ion al e s he abundance o heal h-associa ed bac e ia PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187852 No embe 16, 2017 6 / 18 and ound ha he ela i e abundance o Be ap o eobac e ia was highe (d = 1, p= 0.013) in con amina ed soils. Bac e ial communi y composi ion The o al bac e ial communi y composi ion did no di e be ween c eoso e con amina ed and p is ine soils a he OTU le el as e ealed by he nonme ic mul idimensional scaling (NMDS) o dina ion (Fig 1a) and pe mu a ional mul i a ia e analysis o a iance PERMANOVA (Table 3). The NMDS a he genus le el, howe e , e ealed dis inc ly di e en bac e ial com- muni ies in he c eoso e con amina ed and p is ine soils (Fig 1b). PERMANOVA indica ed ha con amina ion and soil ype had dis inc main e ec s on mic obial communi y a he genus le el and wi hou a s ong in e ac ion e ec (Table 3). These indings unde line ha Table 3. The e ec s o soil ype and c eoso e con amina ion on bac e ial communi y composi ion a genus and OTU le els in PERMANOVA. Genus le el D Sum o squa es Mean squa e F Model R 2 P alue Soil ype (1) 3 0.09 0.03 4.81 0.38 0.001 C eoso e addi ion (2) 1 0.06 0.06 9.08 0.24 0.001 1:2 3 0.04 0.01 2.16 0.17 0.014 Residuals 8 0.05 0.01 0.21 To al 15 0.24 1.00 OTU le el Soil ype (1) 3 2.02 0.67 2.74 0.37 0.001 C eoso e addi ion (2) 1 0.35 0.35 1.41 0.06 0.078 1:2 3 1.07 0.36 1.45 0.20 0.028 Residuals 8 1.98 0.25 0.36 To al 15 5.41 1.00 h ps://doi.o g/10.1371/jou nal.pone.0187852. 003 Fig 1. NMDS o dina ion (B ay-Cu is dis ance) o soil bac e ial communi ies in he c eoso e con amina ed and con ol soil samples. (a) The bac e ial communi y composi ion in he wo ea men g oups o e lap a he OTU le el. (b) The communi ies a e dis inc i ely di e en a he genus le el. Fo pea (Pea in he igu e), pine o es soil (Pine.F) and sp uce o es soil (Sp uce.F), samples aken on week 31 (day91) we e u ilized, whe eas week 27 (day 189) samples we e used in he case o mine al soil (Sand). h ps://doi.o g/10.1371/jou nal.pone.0187852.g001 PAH pollu ion al e s he abundance o heal h-associa ed bac e ia PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187852 No embe 16, 2017 7 / 18 c eoso e con amina ion changed mic obial communi y a he genus le el ega dless o he soil ype. Communi y composi ion analysis o majo bac e ial phyla a OTU le el The majo bac e ial phyla de ec ed om he soil samples we e examined o hei di e ences in communi y composi ion a he OTU le el be ween he c eoso e con amina ed and p is ine soil samples. Since OTUs ep esen ing Bac e oide es we e no de ec ed in wo o he c eoso e con amina ed samples, communi y composi ion analysis was no pe o med o Bac e oide es. Consequen ly, he analyses we e pe o med only o P o eobac e ia and Ac inobac e ia. P o eobac e ia. P o eobac e ial communi ies we e dis inc ly di e en in con amina ed han in con ol soils in NMDS (Fig 2). PERMANOVA e ealed ha he soil ype and c eoso e con amina ion had s ong indi idual e ec s on he P o eobac e ial communi y composi ion, while he signi icance o he in e ac ion e m was en imes lowe (Table 4). Fig 2. NMDS o dina ion o P o eobac e ial OTUs in c eoso e con amina ed and p is ine (con ol) soil samples. The communi y composi ion is no iceably di e en in he wo ea men g oups. The o dina ion is based on B ay-Cu is dissimila i y me ic. h ps://doi.o g/10.1371/jou nal.pone.0187852.g002 PAH pollu ion al e s he abundance o heal h-associa ed bac e ia PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187852 No embe 16, 2017 8 / 18 Ac inobac e ia. Al hough he ela i e abundance o Ac inobac e ia dec eased in c eoso e con amina ed soil samples, i was obse ed ha i s communi y composi ion did no di e be ween he c eoso e con amina ed and p is ine soil samples (Fig 3). Ins ead, he soil ype a ec ed he Ac inobac e ial communi y composi ion ound in ha soil ype (Table 5). Tempo al a ia ion. The o al bac e ial, P o eobac e ial and Ac inobac e ial communi ies sampled on day 0 we e compa ed wi h hose sampled la e . Day 0 samples we e compa ed wi h bo h c eoso e con amina ed and p is ine samples o de e mine he empo al communi y dynamics. Only P o eobac e ial communi y on day 0 dis inc ly di e ed om he communi ies sampled a he la e ime poin s (Fig 4a). Nei he o al bac e ial communi y no Ac inobac e - ial communi y changed du ing he expe imen signi ican ly (Fig 4b and 4c). Discussion In he p esen s udy, we compa ed he esponses o bac e ial communi y o c eoso e con ami- na ion in ou di e en ypes o soil. Ou da a indica es shi s in bac e ial communi y composi- ion and al e a ions in he ela i e abundances o h ee majo bac e ial phyla as a esponse o c eoso e con amina ion. The pollu ion-induced shi was a ibu able o an inc ease in he ela- i e abundance o P o eobac e ia, pa icula ly Be ap o eobac e ia, and a dec ease in he ela- i e abundance o Ac inobac e ia and Bac e oide es. These phyla a e he majo bac e ial phyla p esen in soil as well as on human skin and o he biological samples [49–51]. Because hey ha e been ecen ly linked o human immune unc ion and ch onic diseases [25–27,52], a majo change in hei communi y composi ion in he en i onmen can po en ially a ec human exposu e o hem and, e en ually, lead o modula ed human immune sys em unc ions [27,31]. As a as we know, his idea ha we call he al e ed en i onmen al mic obiome hypo hesis has ne e been p oposed no es ed. The i s s ep o e alua e he hypo hesis is he e o e o pe o m a li e a u e e iew on he po en ial connec ions be ween heal h-associ- a ed and pollu ion-induced changes. While se e al s udies ha e in es iga ed and poin ed ou he al e a ion o he en i e bac e ial communi y as a esul o PAH pollu ion, only a hand ul o s udies ha e ocused on pollu ion- induced changes in P o eobac e ial, Bac e oide es, and Ac inobac e ial communi ies in su ace soil [1,12,53–58]. Those s udies ha e e ealed ha P o eobac e ia we e mo e abundan in PAH pollu ed soils compa ed o p is ine soils. Phylum Bac e oide es, in con as , ei he dec eased in abundance [55] o was no s udied. The ela i e abundance o Ac inobac e ia was usually lowe in con amina ed soil han in non-con amina ed soil [55], bu i s abundance is also dependen on he PAHs in ol ed[57] (Table 6). The inc ease in he abundance o P o eo- bac e ia and he dec ease in Ac inobac e ia and Bac e oide es a e consis en wi h ou indings. As he s udies lis ed in Table 6 co e a wide ange o geog aphic a eas and land use his o ies, i seems plausible o assume ha shi s in he abundance o P o eobac e ia, Ac inobac e ia, and Bac e oide es a e a ypical consequence o PAH pollu ion in soil. Table 4. The e ec s o soil ype and c eoso e con amina ion on P o eobac e ial communi y composi ion a he OTU le el in PERMANOVA. D Sum o squa es Mean squa e F Model R 2 P alue Soil ype (1) 3 1.50 0.50 2.13 0.28 0.001 C eoso e addi ion (2) 1 0.82 0.82 3.50 0.15 0.001 1:2 3 1.06 0.35 1.51 0.20 0.010 Residuals 8 1.87 0.23 0.36 To al 15 5.26 1.00 h ps://doi.o g/10.1371/jou nal.pone.0187852. 004 PAH pollu ion al e s he abundance o heal h-associa ed bac e ia PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187852 No embe 16, 2017 9 / 18 34. Kulik N, Goi A, T apido M, Tuhkanen T. Deg ada ion o polycyclic a oma ic hyd oca bons by combined chemical p e-oxida ion and bio emedia ion in c eoso e con amina ed soil. Jou nal o En i onmen al Managemen . 2006; 78(4):382–91. h ps://doi.o g/10.1016/j.jen man.2005.05.005 PMID: 16154683 35. Valde ama C, Gamisans X, De las He as X, Fa an A, Co ina J. So p ion kine ics o polycyclic a o- ma ic hyd oca bons emo al using g anula ac i a ed ca bon: in apa icle di usion coe icien s. Jou nal o Haza dous Ma e ials. 2008; 157(2):386–96. 36. Ran alainen M-L, To kkeli M, S o ¨mme R, Se a ¨la ¨H. Lead con amina ion o an old shoo ing ange a ec ing he local ecosys em—a case s udy wi h a holis ic app oach. Science o he o al en i onmen . 2006; 369(1):99–108. 37. Sinkkonen A, Kauppi S, Simpanen S, Ran alainen A-L, S o ¨mme R, Roman schuk M. Laye o o ganic pine o es soil on op o chlo ophenol-con amina ed mine al soil enhances con aminan deg ada ion. En i onmen al Science and Pollu ion Resea ch. 2013; 20(3):1737–45. h ps://doi.o g/10.1007/s11356- 012-1047-1 PMID: 22752813 38. Kauppi S, Roman schuk M, S o ¨mme R, Sinkkonen A. Na u al a enua ion is enhanced in p e iously con amina ed and coni e ous o es soils. En i onmen al Science and Pollu ion Resea ch. 2012; 19 (1):53–63. h ps://doi.o g/10.1007/s11356-011-0528-y PMID: 21660637 39. Sinkkonen A, Ollila S, Roman schuk M. Changes in TcpA gene equency explain 2, 4, 6- ichlo ophenol deg ada ion in mesocosms. Jou nal o En i onmen al Science and Heal h, Pa B. 2014; 49(10):756–9. 40. Ka s ensen KH. No dic guidelines o chemical analysis o con amina ed soil samples: No d es ; 1996. 41. Honkonen O, Ran alainen A-L. Impac o u baniza ion on he concen a ions and dis ibu ion o o ganic con aminan s in bo eal lake sedimen s. En i onmen al moni o ing and assessmen . 2013; 185 (2):1437–49. h ps://doi.o g/10.1007/s10661-012-2643-8 PMID: 22527470 42. Muyze G, De Waal EC, Ui e linden AG. P o iling o complex mic obial popula ions by dena u ing g adi- en gel elec opho esis analysis o polyme ase chain eac ion-ampli ied genes coding o 16S RNA. Applied and en i onmen al mic obiology. 1993; 59(3):695–700. PMID: 7683183 43. Schloss PD, Ge e s D, Wes co SL. Reducing he e ec s o PCR ampli ica ion and sequencing a i- ac s on 16S RNA-based s udies. PloS one. 2011; 6(12):e27310. h ps://doi.o g/10.1371/jou nal.pone. 0027310 PMID: 22194782 44. Edga RC, Haas BJ, Clemen e JC, Quince C, Knigh R. UCHIME imp o es sensi i i y and speed o chi- me a de ec ion. Bioin o ma ics. 2011; 27(16):2194–200. h ps://doi.o g/10.1093/bioin o ma ics/b 381 PMID: 21700674 45. Wang Q, Ga i y GM, Tiedje JM, Cole JR. Nai e Bayesian classi ie o apid assignmen o RNA sequences in o he new bac e ial axonomy. Applied and en i onmen al mic obiology. 2007; 73 (16):5261–7. h ps://doi.o g/10.1128/AEM.00062-07 PMID: 17586664 46. Bo ca d D, Gille F, Legend e P. Nume ical ecology wi h R: Sp inge Science & Business Media; 2011. 47. Cousins IT, Beck AJ, Jones KC. A e iew o he p ocesses in ol ed in he exchange o semi- ola ile o ganic compounds (SVOC) ac oss he ai –soil in e ace. Science o he To al En i onmen . 1999; 228 (1):5–24. 48. Jacobson M, Hansson HC, Noone K, Cha lson R. O ganic a mosphe ic ae osols: Re iew and s a e o he science. Re iews o Geophysics. 2000; 38(2):267–94. 49. Cho I, Blase MJ. The human mic obiome: a he in e ace o heal h and disease. Na u e e iews Gene - ics. 2012; 13(4):260. h ps://doi.o g/10.1038/n g3182 PMID: 22411464 50. Zoe endal E, Rajilić-S ojano ićM, De Vos W. High- h oughpu di e si y and unc ionali y analysis o he gas oin es inal ac mic obio a. Gu . 2008; 57(11):1605–15. h ps://doi.o g/10.1136/gu .2007.133603 PMID: 18941009 51. Ley RE, Hamady M, Lozupone C, Tu nbaugh PJ, Ramey RR, Bi che JS, e al. E olu ion o mammals and hei gu mic obes. Science. 2008; 320(5883):1647–51. h ps://doi.o g/10.1126/science.1155725 PMID: 18497261 52. Hanski I. Biodi e si y, mic obes and human well-being. E hics in Science and En i onmen al Poli ics. 2014; 14(1):19–25. 53. Mukhe jee S, Juo onen H, Sii onen P, Quesada CL, Tuomi P, Pulkkinen P, e al. Spa ial pa e ns o mic obial di e si y and ac i i y in an aged c eoso e-con amina ed si e. The ISME jou nal. 2014; 8 (10):2131. h ps://doi.o g/10.1038/ismej.2014.151 PMID: 25105905 54. Peng J, Zhang Y, Su J, Qiu Q, Jia Z, Zhu Y-G. Bac e ial communi ies p edominan in he deg ada ion o 13 C 4–4, 5, 9, 10-py ene du ing compos ing. Bio esou ce echnology. 2013; 143:608–14. h ps://doi. o g/10.1016/j.bio ech.2013.06.039 PMID: 23845709 55. Ma in F, To elli S, Le Paslie D, Ba bance A, Ma in-Lau en F, B u D, e al. Be ap o eobac e ia domi- nance and di e si y shi s in he bac e ial communi y o a PAH-con amina ed soil exposed o PAH pollu ion al e s he abundance o heal h-associa ed bac e ia PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187852 No embe 16, 2017 16 / 18 phenan h ene. En i onmen al Pollu ion. 2012; 162:345–53. h ps://doi.o g/10.1016/j.en pol.2011.11. 032 PMID: 22243884 56. Sawulski P, Clipson N, Doyle E. E ec s o polycyclic a oma ic hyd oca bons on mic obial communi y s uc u e and PAH ing hyd oxyla ing dioxygenase gene abundance in soil. Biodeg ada ion. 2014; 25 (6):835–47. h ps://doi.o g/10.1007/s10532-014-9703-4 PMID: 25095739 57. Llado ´S, Co ino S, Solanas A, Pe uccioli M, D’annibale A, Viñas M. Py osequencing e eals he e ec o mobilizing agen s and lignocellulosic subs a e amendmen on mic obial communi y composi ion in a eal indus ial PAH-pollu ed soil. Jou nal o haza dous ma e ials. 2015; 283:35–43. h ps://doi.o g/10. 1016/j.jhazma .2014.08.065 PMID: 25261758 58. Tejeda-Ag edano M, Gallego S, Vila J, G i oll M, O ega-Cal o J, Can os M. In luence o he sun lowe hizosphe e on he biodeg ada ion o PAHs in soil. Soil Biology and Biochemis y. 2013; 57:830–40. 59. Hansi M, Weidenhame JD, Sinkkonen A. Plan g ow h esponses o ino ganic en i onmen al con ami- nan s a e densi y-dependen : Expe imen s wi h coppe sul a e, ba ley and le uce. En i onmen al Pollu- ion. 2014; 184:443–8. h ps://doi.o g/10.1016/j.en pol.2013.09.027 PMID: 24121419 60. Belz RG, Sinkkonen A. Selec i e oxin e ec s on as e and slowe g owing indi iduals in he o ma ion o ho mesis a he popula ion le el—A case s udy wi h Lac uca sa i a and PCIB. Science o he To al En i onmen . 2016; 566:1205–14. h ps://doi.o g/10.1016/j.sci o en .2016.05.176 PMID: 27267716 61. Płociniczak T, Sinkkonen A, Roman schuk M, Sułowicz S, Pio owska-Sege Z. Rhizosphe ic bac e ial s ain B e ibac e ium casei MH8a colonizes plan issues and enhances Cd, Zn, Cu phy oex ac ion by whi e mus a d. F on ie s in plan science. 2016; 7. 62. La sen N, Vogensen FK, an den Be g FW, Nielsen DS, And easen AS, Pede sen BK, e al. Gu mic o- bio a in human adul s wi h ype 2 diabe es di e s om non-diabe ic adul s. PloS one. 2010; 5(2):e9085. h ps://doi.o g/10.1371/jou nal.pone.0009085 PMID: 20140211 63. Ciaccio CE, Ba nes C, Kennedy K, Chan M, Po noy J, Rosenwasse L. Home dus mic obio a is diso - de ed in homes o low-income as hma ic child en. Jou nal o As hma. 2015; 52(9):873–80. h ps://doi. o g/10.3109/02770903.2015.1028076 PMID: 26512904 64. Lepage P, Ha ¨sle R, Spehlmann ME, Rehman A, Z i bliene A, Begun A, e al. Twin s udy indica es loss o in e ac ion be ween mic obio a and mucosa o pa ien s wi h ulce a i e coli is. Gas oen e ology. 2011; 141(1):227–36. h ps://doi.o g/10.1053/j.gas o.2011.04.011 PMID: 21621540 65. Mondo S, Kang S, Fu e J-P, Agui e de Ca ce D, McSweeney C, Mo ison M, e al. Highligh ing new phylogene ic speci ici ies o C ohn’s disease mic obio a. In lamma o y bowel diseases. 2011; 17 (1):185–92. h ps://doi.o g/10.1002/ibd.21436 PMID: 20722058 66. Chen Y, Yang F, Lu H, Wang B, Chen Y, Lei D, e al. Cha ac e iza ion o ecal mic obial communi ies in pa ien s wi h li e ci hosis. Hepa ology. 2011; 54(2):562–72. h ps://doi.o g/10.1002/hep.24423 PMID: 21574172 67. Panze AR, Lynch SV. In luence and e ec o he human mic obiome in alle gy and as hma. Cu en opinion in heuma ology. 2015; 27(4):373–80. h ps://doi.o g/10.1097/BOR.0000000000000191 PMID: 26002029 68. Thomas F, Hehemann J-H, Rebu e E, Czjzek M, Michel G. En i onmen al and gu bac e oide es: he ood connec ion. F on ie s in mic obiology. 2011; 2. 69. Yu D, Sinkkonen A, Hui N, Ku ola J, Kukkonen S, Pa ikka P, e al. Molecula p o ile o mic obio a o Finnish comme cial compos supp essi e agains Py hium disease on cucumbe plan s. Applied Soil Ecology. 2015; 92:47–53. 70. Zhang H, DiBaise JK, Zuccolo A, Kud na D, B aido i M, Yu Y, e al. Human gu mic obio a in obesi y and a e gas ic bypass. P oceedings o he Na ional Academy o Sciences. 2009; 106(7):2365–70. 71. Timmis KN, McGeni y T, Van De Mee JR, de Lo enzo V. Handbook o hyd oca bon and lipid mic obiol- ogy: Sp inge Be lin; 2010. 72. Mazmanian SK, Round JL, Kaspe DL. A mic obial symbiosis ac o p e en s in es inal in lamma o y disease. Na u e. 2008; 453(7195):620. h ps://doi.o g/10.1038/na u e07008 PMID: 18509436 73. Round JL, Mazmanian SK. Inducible Foxp3+ egula o y T-cell de elopmen by a commensal bac e ium o he in es inal mic obio a. P oceedings o he Na ional Academy o Sciences. 2010; 107(27):12204–9. 74. Kos ic AD, Ge e s D, Pedamallu CS, Michaud M, Duke F, Ea l AM, e al. Genomic analysis iden i ies associa ion o Fusobac e ium wi h colo ec al ca cinoma. Genome esea ch. 2012; 22(2):292–8. h ps:// doi.o g/10.1101/g .126573.111 PMID: 22009990 75. Jensen A, Fago ¨-Olsen H, Sø ensen CH, Kilian M. Molecula mapping o species le el o he onsilla c yp mic obio a associa ed wi h heal h and ecu en onsilli is. PLoS One. 2013; 8(2):e56418. h ps:// doi.o g/10.1371/jou nal.pone.0056418 PMID: 23437130 PAH pollu ion al e s he abundance o heal h-associa ed bac e ia PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187852 No embe 16, 2017 17 / 18 76. de S eenhuijsen Pi e s WA, Huijskens EG, Wyllie AL, Biesb oek G, Van Den Be gh MR, Veenho en RH, e al. Dysbiosis o uppe espi a o y ac mic obio a in elde ly pneumonia pa ien s. The ISME jou - nal. 2016; 10(1):97–108. h ps://doi.o g/10.1038/ismej.2015.99 PMID: 26151645 77. An iza -Ladislao B, Spano a K, Beck AJ, Russell NJ. Mic obial communi y s uc u e changes du ing bio emedia ion o PAHs in an aged coal- a con amina ed soil by in- essel compos ing. In e na ional Biode e io a ion Biodeg ada ion. 2008; 61(4 SRC—GoogleSchola ):357–64. 78. Johnsen AR, Schmid S, Hybhol TK, Hen iksen S, Jacobsen CS, Ande sen O. S ong impac on he polycyclic a oma ic hyd oca bon (PAH)-deg ading communi y o a PAH-pollu ed soil bu ma ginal e ec on PAH deg ada ion when p iming wi h bio emedia ed soil domina ed by mycobac e ia. Applied and en i onmen al mic obiology. 2007; 73(5 SRC—GoogleSchola ):1474–80. 79. To ¨ neman N, Yang X, Båå h E, Beng sson G. Spa ial co a ia ion o mic obial communi y composi ion and polycyclic a oma ic hyd oca bon concen a ion in a c eoso e-pollu ed soil. En i onmen al oxicology and chemis y. 2008; 27(5):1039–46. h ps://doi.o g/10.1897/07-440.1 PMID: 18419193 80. Ma i PR, S e n DA, W igh AL, Billheime D, Ma inez FD. As hma-associa ed di e ences in mic obial composi ion o induced spu um. Jou nal o Alle gy and Clinical Immunology. 2013; 131(2):346–52. e3. h ps://doi.o g/10.1016/j.jaci.2012.11.013 PMID: 23265859 81. Schno SL, Candela M, Rampelli S, Cen anni M, Consolandi C, Basaglia G, e al. Gu mic obiome o he Hadza hun e -ga he e s. Na u e communica ions. 2014; 5:3654. h ps://doi.o g/10.1038/ ncomms4654 PMID: 24736369 82. Ga els W, Tallu i TR, Ap elbaum R, Ca a ala ´YP, Bosch P, Po ¨ zsch K, e al. One-s ep mul iplex ans- genesis ia sleeping beau y ansposi ion in ca le. Scien i ic epo s. 2016; 6:21953. h ps://doi.o g/10. 1038/s ep21953 PMID: 26905416 83. Mah KW, Sangsupawanich P, Tunyapani W, an Be e H, Shek LP, Chua KY, e al. Gu mic obio a o child en li ing in u al sou h Thailand and u ban Singapo e. Alle gology In e na ional. 2008; 57(1):65– 71. h ps://doi.o g/10.2332/alle golin .O-07-501 PMID: 18089942 84. De Filippo C, Ca alie i D, Di Paola M, Ramazzo i M, Poulle JB, Massa S, e al. Impac o die in shap- ing gu mic obio a e ealed by a compa a i e s udy in child en om Eu ope and u al A ica. P oceed- ings o he Na ional Academy o Sciences. 2010; 107(33):14691–6. 85. Ya sunenko T, Rey FE, Mana y MJ, T ehan I, Dominguez-Bello MG, Con e as M, e al. Human gu mic obiome iewed ac oss age and geog aphy. na u e. 2012; 486(7402):222. h ps://doi.o g/10.1038/ na u e11053 PMID: 22699611 86. Ying S, Zeng D-N, Chi L, Tan Y, Galzo e C, Ca dona C, e al. The in luence o age and gende on skin- associa ed mic obial communi ies in u ban and u al human popula ions. PLoS One. 2015; 10(10): e0141842. h ps://doi.o g/10.1371/jou nal.pone.0141842 PMID: 26510185 PAH pollu ion al e s he abundance o heal h-associa ed bac e ia PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0187852 No embe 16, 2017 18 / 18