The Fate of Bacteriophages in Recirculating Aquaculture Systems (RAS) : Towards Developing Phage Therapy for RAS

This is a sel -a chi ed e sion o an o iginal a icle. This e sion may di e om he o iginal in pagina ion and ypog aphic de ails. Au ho (s): Ti le: Yea : Ve sion: Copy igh : Righ s: Righ s u l: Please ci e he o iginal e sion: CC BY 4.0 h ps://c ea i ecommons.o g/licenses/by/4.0/ The Fa e o Bac e iophages in Reci cula ing Aquacul u e Sys ems (RAS) : Towa ds De eloping Phage The apy o RAS © 2019 by he au ho s. Licensee MDPI, Basel, Swi ze land Published e sion Almeida, Gab iel M. F..; Mäkelä, Ka i; Laan o, Elina; Pulkkinen, Jani; Vielma, Jouni; Sundbe g, Lo a-Riina Almeida, G. M. F., Mäkelä, K., Laan o, E., Pulkkinen, J., Vielma, J., & Sundbe g, L.-R. (2019). The Fa e o Bac e iophages in Reci cula ing Aquacul u e Sys ems (RAS) : Towa ds De eloping Phage The apy o RAS. An ibio ics, 8(4), A icle 192. h ps://doi.o g/10.3390/an ibio ics8040192 2019 an ibio ics A icle The Fa e o Bac e iophages in Reci cula ing Aquacul u e Sys ems (RAS)—Towa ds De eloping Phage The apy o RAS Gab iel M.F. Almeida 1, Ka i Mäkelä 1, Elina Laan o 1,2 , Jani Pulkkinen 3, Jouni Vielma 3and Lo a-Riina Sundbe g 1,* 1Depa men o Biological and En i onmen al Science, Nanoscience Cen e , Uni e si y o Jy äskylä, 40014 Jy äskylä, Finland; [email p o ec ed] (G.M.F.A.); [email p o ec ed] (K.M.); [email p o ec ed] (E.L.) 2Facul y o Biological and En i onmen al Sciences, Molecula and In eg a i e Biosciences Resea ch P og amme, Uni e si y o Helsinki, 00014 Helsinki, Finland 3Na u al Resou ces Ins i u e Finland, P oduc ion Sys ems, 40500 Jy äskylä, Finland; [email p o ec ed] (J.P.); Jouni.V[email p o ec ed] (J.V.) *Co espondence: lo a- iina.sundbe [email p o ec ed] Recei ed: 8 Oc obe 2019; Accep ed: 21 Oc obe 2019; Published: 24 Oc obe 2019   Abs ac : Aquacul u e p oduc ion has inc eased emendously du ing he las decades, and new echniques ha e been de eloped, e.g., eci cula ing aquacul u e sys ems (RAS). In RAS, he majo i y o wa e olume is ci cula ed ia mechanical and biological il e s and eused in he anks. Howe e , he p e en ion and ea men o diseases in hese sys ems a e challenging, as he pa hogens sp ead h oughou he sys em, and he addi ion o chemicals and an ibio ics dis up s he mic obiome o he bio il e s. The inc easing an ibio ic esis ance has made phage he apy a ele an al e na i e o an ibio ics in ood p oduc ion. Indeed, as hos -speci ic and sel - eplica ing agen hey migh be op imal o a ge ed pa hogen e adica ion in RAS. We es ed he su i al and sp ead o Fla obac e ium columna e -in ec ing phage FCL-2 in eci cula ing aquacul u e ish a m wi h ainbow ou (Onco hynchus mykiss) in a ully con olled s udy. A e a single addi ion, phage pe sis ed in wa e samples collec ed om ank, ixed bed, mo ing bed, and ae a ion uni up o 14 days, and in he wa e o ea ing anks, ainbow ou mucus, and bio eac o ca ie media om he ixed and mo ing bed bio il e s o 21 days. Fu he mo e, phage adso bed p e e en ially o mo ing bed ca ie media, which con ained bio ilm a ached and om which highe phage numbe s we e eco e ed. This s udy shows phages as a po en s a egy o main aining biosecu i y in RAS sys ems. Keywo ds: aquacul u e; bac e iophage; bio il e ; disease; phage he apy; RAS; eci cula ing aquacul u e sys ems 1. In oduc ion Aquacul u e p oduc ion is an impo an sou ce o a p o ein des ined o human consump ion. I is an expanding indus y wi h many s a egies employed o imp o e e iciency and en i onmen al impac . Du ing ecen yea s, a g owing in e es in eci cula ing aquacul u e sys ems (RAS) has appea ed [ 1 , 2 ]. Wa e in RAS is eused a e being ea ed, d ama ically educing he amoun o wa e needed o ish a ming in aquacul u e si es. Con ining la ge popula ions o aqua ic animals a e a isk ac o when conside ing in ec ious diseases, and aquacul u e ea ing uni s ha e been shown o a o pa hogens and i ulence inc ease [ 3 – 5 ]. The euse o wa e in RAS may pose a biosecu i y challenge o in ec ious diseases, as he emo al o pa hogens o chemical esidues om he anks is di icul , inc easing he unnecessa y exposu e o aqua ic animals o bo h [ 6 ]. Many o he pa hogens ha An ibio ics 2019,8, 192; doi:10.3390/an ibio ics8040192 www.mdpi.com/jou nal/an ibio ics An ibio ics 2019,8, 192 2 o 9 h ea en aquacul u e a e bac e ia, and he diseases a e ea ed wi h an ibio ics. An ibio ic leakage o he en i onmen can lead o en i onmen al and heal h issues (i.e., al e a ions in mic obial communi ies and selec ion o esis ance agains an ibio ics) [ 7 ]. Fo example, an ibio ic esis ance genes a e en iched in sedimen s associa ed wi h aquacul u e si es [ 8 , 9 ]. Thus, means o implemen he One Heal h app oach (h ps://www.who.in / ea u es/qa/one-heal h/en/) in aquacul u e p oduc ion need o be de eloped. One possible al e na i e o an ibio ics and a solu ion o he an imic obial esis ance c isis and One Heal h app oach is he use o (bac e ia) phage he apy. Phages a e i uses ha speci ically in ec only bac e ia, wi hou causing ha m o su ound mic obio a o euka yo ic cells. Phage he apy is no a no el concep , as i has been used o almos a cen u y o ea bac e ial in ec ions in humans [10,11]. I has also been applied o aquacul u e ela ed bac e ial pa hogens, expe imen ally, and in p ac ice [ 12 , 13 ]. So a , he e is no in o ma ion abou phage use in RAS, bu his ype o sys em can be conside ed an ideal en i onmen o phage he apy applica ions [ 14 ]. The wa e eci cula ion p ocess is an ad an age o phage deli e y since, in heo y, phage would emain in he sys em o long ime pe iods o being small enough o pass h ough il e s and o he ba ie s, eci cula ing eely wi h he wa e . This would enable p olonged p o ec ion ia ex ending he ime animals a e exposed o phage, while also a o ing phage e olu ion in esponse o phage- esis an bac e ia, which has been documen ed in open aquacul u e sys ems [ 15 ]. In addi ion, whe eas chemicals and an ibio ics migh in luence he mic obiome composi ion o biological il e s [ 16 ], hos -speci ic phages in ec only hei a ge bac e ia. Howe e , phage s abili y o e ime on RAS a ms, phage dis ibu ion in he anks, and how he di e en ea men uni s a ec phage su i al ha e no been es ed. In his wo k, we in es iga ed phage dynamics in RAS esea ch ish a m wi h ainbow ou (Onco hynchus mykiss). As a model, we used Fla obac e ium columna e -in ec ing phage FCL-2, which has been isola ed om he same ish a m yea s ea lie and shown o be sui able o phage he apy in ou ea lie s udies [ 12 , 17 ]. Phage solu ion was added o he wa e o h ee ea ing anks, and phage numbe s we e measu ed o e ime om di e en loca ions o each sys em, on he ish mucus, and on plas ic ca ie media used o collec bio ilm in he mo ing and ixed-bed bio eac o s. We demons a e phage pe sis ence in a RAS o h ee weeks a e a single phage exposu e. Phage concen a ion was highes in he plas ic ca ie media, indica ing ha bac e ial bio ilm migh be a subs a e whe e phage p e e en ially en ich. Addi ionally, phage p esence did no al e wa e quali y pa ame e s no esul ed in ish mo ali y. The use o phage he apy on RAS can be e ec i e since only one phage dose is su icien o pe sis ence and sp ead o e he whole sys em wi hou being des oyed by he wa e ea men p ocesses. 2. Resul s 2.1. Phages Pe sis in RAS o Up o Th ee Weeks Phages we e eco e ed om all sampled componen s o he RAS sys em up o 14 days, and om he wa e om ea ing anks, ish mucus and ca ie media om he ixed and mo ing bed bio il e s o 21 days a e a single phage exposu e (Figu e 1A–G). In all samples, a simila end was obse ed wi h he phage numbe s decaying as e in he i s days hen becoming mo e s able a he end. Phage i e s we e highe in il e ca ie media samples, sugges ing ha bac e ial bio ilm may ha e a ole in phage en ichmen . On day se en, he phage p esence on ish gills was es ed. Phage i e s in gills we e simila o he ones ound in mucus and sligh ly lowe han in wa e (Figu e 1H). No in ec i e phage pa icles appea ed in samples aken om he con ol anks o e he cou se o he expe imen . An ibio ics 2019,8, 192 3 o 9 An ibio ics 2019, 8, x 3 o 9 3 Table 1. Mean (±SD) wa e quali y pa ame e s o eci cula ing aquacul u e sys em (RAS) con ol uni s and phage- ea ed uni s. TAN = o al ammonia ni ogen, NO2-N = Ni i e, NO3-N = Ni a e, TOC = o al o ganic ca bon. Pa ame e Con ol uni s Phage- ea ed uni s TAN (mg·L−1) 1.31 ± 0.11 1.20 ± 0.23 NO2-N (mg·L−1) 0.53 ± 0.01 0.57 ± 0.05 NO3-N (mg·L−1) 42.4 ± 0.2 43.1 ± 1.2 pH 7.2 ± 0.0 7.2 ± 0.0 T (°C) 15.5 ± 0.3 15.2 ± 0.3 O2 (mg·L−1) 7.3 ± 1.0 8.7 ± 0.9 TOC (mg·L−1) 14.4 ± 4.6 17.6 ± 4.0 Figu e 1. Phage pe sis ence o e ime in he eci cula ing aquacul u e sys em (RAS) anks and ish. Phage FCL-2 i e s (p u/mL) on (A) he ish ank wa e , (B) he mo ing bed wa e , (C) he ixed-bed wa e , (D) he ae a ion uni wa e , (E) ish mucus, (F) he mo ing bed plas ic ca ie media, (G) he Figu e 1. Phage pe sis ence o e ime in he eci cula ing aquacul u e sys em (RAS) anks and ish. Phage FCL-2 i e s (p u/mL) on ( A ) he ish ank wa e , ( B ) he mo ing bed wa e , ( C ) he ixed-bed wa e , ( D ) he ae a ion uni wa e , ( E ) ish mucus, ( F ) he mo ing bed plas ic ca ie media, ( G ) he ixed-bed plas ic ca ie media, ( H ) on ish gills (se en days a e phage exposu e). In A , B , C, and D, each ank was sampled once un il day se en and in iplica es om day nine onwa ds. In E, h ee ish pe ank we e sampled in he i s and i h day, hen h ee ish pe con ol ank, and en ish pe phage- ea ed ank we e measu ed om day se en onwa ds. In F and G, i e ca ie media pieces pe ank we e sampled in all ime poin s. In H, h ee ish om each con ol ank and en ish om each phage ea ed ank we e measu ed. No phages capable o in ec ing Fla obac e ium columna e s ain B185 we e de ec ed om wa e o mucus samples collec ed a ime ze o. Also, no F. columna e was isola ed om ish gills sampled a ime ze o. These nega i e con ols, allied o he da a om con ol anks, suppo he idea ha phages ound An ibio ics 2019,8, 192 4 o 9 o e ime we e de i ed om he o iginal lysa e and no om he en i onmen o na u al in ec ions du ing he cou se o he expe imen . Wa e pa ame e s we e measu ed o e he cou se o he expe imen , and no signi ican changes we e de ec ed be ween con ol and phage ea ed RAS (Table 1). No ish mo ali ies no dec eased eed in ake a e we e obse ed in he RAS uni s ea ed wi h phage. Table 1. Mean ( ± SD) wa e quali y pa ame e s o eci cula ing aquacul u e sys em (RAS) con ol uni s and phage- ea ed uni s. TAN = o al ammonia ni ogen, NO2-N =Ni i e, NO3-N =Ni a e, TOC = o al o ganic ca bon. Pa ame e Con ol Uni s Phage-T ea ed Uni s TAN (mg·L−1)1.31 ±0.11 1.20 ±0.23 NO2-N (mg·L−1)0.53 ±0.01 0.57 ±0.05 NO3-N (mg·L−1)42.4 ±0.2 43.1 ±1.2 pH 7.2 ±0.0 7.2 ±0.0 T (◦C) 15.5 ±0.3 15.2 ±0.3 O2 (mg·L−1)7.3 ±1.0 8.7 ±0.9 TOC (mg·L−1)14.4 ±4.6 17.6 ±4.0 2.2. A achmen o Phages o Fil e Pelle s Is Dependen on Bio ilm As he da a collec ed om he RAS expe imen indica ed, phages we e p e e en ially en iched in he plas ic ca ie media o ixed and mo ing bed bio il e s. The e o e we decided o es phage adso p ion in his ma e ial in i o di ec ly. S e ile ca ie media and media collec ed om con ol (no phage) anks we e exposed o phage, and he numbe o ee phages was de e mined in he supe na an . I should be no ed ha he media collec ed om ish anks we e colonized by bio ilms, while au ocla ed media we e clean and s e ile. The pe cen age o a ached phage was calcula ed om he i e (p u/mL) o he ee phage pa icles in he wa e a each ime poin (which in u n was used o calcula e he co esponding numbe o phage pa icles a ached o he pelle s). (Figu e 2). Al hough phage inac i a ion was no es ed, i is unlikely o ha e a ec ed he esul s since phage FCL-2 is conside ably s able o e ime. I is clea ha phage adso bed p e e en ially o mo ing bed bio eac o media. A e six hou s, mo e phages we e adso bed o mo ing bed han o ixed bed media and s e ile media (p=0.011 and p=0.000033, espec i ely). This pe sis ed up o 24 hou s a e exposu e ( p=0.000013 when compa ed o ixed-bed il e s and p<0.000001 when compa ed o s e ile ca ie media). An ibio ics 2019, 8, x 4 o 9 4 ixed-bed plas ic ca ie media, (H) on ish gills (se en days a e phage exposu e). In A, B, C, and D, each ank was sampled once un il day se en and in iplica es om day nine onwa ds. In E, h ee ish pe ank we e sampled in he i s and i h day, hen h ee ish pe con ol ank, and en ish pe phage- ea ed ank we e measu ed om day se en onwa ds. In F and G, i e ca ie media pieces pe ank we e sampled in all ime poin s. In H, h ee ish om each con ol ank and en ish om each phage ea ed ank we e measu ed. 2.2. A achmen o Phages o Fil e Pelle s is Dependen on Bio ilm As he da a collec ed om he RAS expe imen indica ed, phages we e p e e en ially en iched in he plas ic ca ie media o ixed and mo ing bed bio il e s. The e o e we decided o es phage adso p ion in his ma e ial in i o di ec ly. S e ile ca ie media and media collec ed om con ol (no phage) anks we e exposed o phage, and he numbe o ee phages was de e mined in he supe na an . I should be no ed ha he media collec ed om ish anks we e colonized by bio ilms, while au ocla ed media we e clean and s e ile. The pe cen age o a ached phage was calcula ed om he i e (p u/mL) o he ee phage pa icles in he wa e a each ime poin (which in u n was used o calcula e he co esponding numbe o phage pa icles a ached o he pelle s). (Figu e 2). Al hough phage inac i a ion was no es ed, i is unlikely o ha e a ec ed he esul s since phage FCL-2 is conside ably s able o e ime. I is clea ha phage adso bed p e e en ially o mo ing bed bio eac o media. A e six hou s, mo e phages we e adso bed o mo ing bed han o ixed bed media and s e ile media (p = 0.011 and p = 0.000033, espec i ely). This pe sis ed up o 24 hou s a e exposu e (p = 0.000013 when compa ed o ixed-bed il e s and p < 0.000001 when compa ed o s e ile ca ie media). Figu e 2. Adso p ion o phage o plas ic ca ie media. Each condi ion was es ed in iplica es. Unpai ed - es s we e used o compa ing he con ols and es ed condi ions. * S a is ical di e ence ound (exac p alues a e men ioned in he ex ). 3. Discussion Bac e ial disease epidemics need an ibio ic ea men also in ish p oduc ion, bu he legisla ion ela ed o an ibio ic use in aquacul u e di e s be ween coun ies. Ne e heless, an ibio ic use in luences mic obial ecology and esis ance e olu ion also ou side ish a ming [7,18,19], a ec ing en i onmen al and ood sa e y. A he ime o he ‘pos -an ibio ic e a’ decla ed by he WHO, in e es owa ds sus ainable ools ha educe he need o an ibio ics is inc easing. Phages in ec ing aquacul u e pa hogens such as Fla obac e ium and Vib io species ha e al eady been isola ed [14,20,21]. The use o RAS is ad an ageous o o he aquacul u e me hods when conside ing wa e use and nu ien discha ges. Howe e , i is also a ulne able en i onmen o ou b eaks o in ec ious diseases. RAS has been conside ed o be an op imal en i onmen o he use o phage he apy agains bac e ial diseases [14], bu so a , phage dis ibu ion and s abili y ha e no been es ed on hese sys ems. He e Figu e 2. Adso p ion o phage o plas ic ca ie media. Each condi ion was es ed in iplica es. Unpai ed - es s we e used o compa ing he con ols and es ed condi ions. * S a is ical di e ence ound (exac p alues a e men ioned in he ex ). An ibio ics 2019,8, 192 5 o 9 3. Discussion Bac e ial disease epidemics need an ibio ic ea men also in ish p oduc ion, bu he legisla ion ela ed o an ibio ic use in aquacul u e di e s be ween coun ies. Ne e heless, an ibio ic use in luences mic obial ecology and esis ance e olu ion also ou side ish a ming [ 7 , 18 , 19 ], a ec ing en i onmen al and ood sa e y. A he ime o he ‘pos -an ibio ic e a’ decla ed by he WHO, in e es owa ds sus ainable ools ha educe he need o an ibio ics is inc easing. Phages in ec ing aquacul u e pa hogens such as Fla obac e ium and Vib io species ha e al eady been isola ed [14,20,21]. The use o RAS is ad an ageous o o he aquacul u e me hods when conside ing wa e use and nu ien discha ges. Howe e , i is also a ulne able en i onmen o ou b eaks o in ec ious diseases. RAS has been conside ed o be an op imal en i onmen o he use o phage he apy agains bac e ial diseases [14], bu so a , phage dis ibu ion and s abili y ha e no been es ed on hese sys ems. He e we show ha a e a single phage addi ion, he phage can be e-isola ed om he RAS uni s a a ish a m o up o h ee weeks. Phages we e e-isola ed om wa e collec ed om di e en poin s in he sys em, om ish mucus and gills, and also om plas ic ca ie media used in bio il e uni s. Phage pe sis ence was longes in he ea ing ank wa e , ish mucus, and bio ilm media. This indica es ha in e ac ions wi h he ank mic obiome o wi h il a ion and wa e ea men sys ems a e no dele e ious o phage p esence. Fu he mo e, and pe haps mo e impo an ly, phage addi ion did no cause al e a ions in wa e quali y no in ish heal h, demons a ing i is a sa e ea men in he ish a ming sys em. Howe e , de ailed knowledge o he applied phage needs o be a ailable o conside i s sui abili y as a ea men [22]. Phage e en ion in animal mucus has been shown p e iously and associa ed wi h he p esence o Ig-like domains in he phage s uc u al p o eins [ 23 ]. We ha e al eady shown ha phages FCL-2 and T4 a e held o up o a week in ainbow ou mucus in open low- h ough sys ems, and ha FCL-2 e en ion has a p o ec i e e ec agains Fla obac e ium columna e in ec ions [ 17 ]. Since in his s udy, phage FCL-2 was p esen in he wa e a all ime poin s due o wa e eci cula ion in he sys em, phage en ichmen in mucus was no e iden al hough i las ed longe han in he low- h ough sys ems [ 17 ]. Longe pe sis ence o phages in mucus and wa e sugges ha p o ec ion agains in ec ions will also be main ained o longe pe iods, which could educe o delay he need o an ibio ic ea men s. The ea ing wa e quali y in RAS sys ems elies on he capaci y o bio il e s o emo e ammonia, ye a as po ion o bac e ia species in he bio il e s a e o he , mainly he e o ophic bac e ia [ 24 , 25 ]. Su p isingly, he plas ic media colonized by bio ilms in he mo ing and ixed bed bio il e s con ained mo e phages han he es o he RAS sys em. Indeed, in i o es ing e ealed ha bio ilm was needed o phage e en ion in he pelle s. While we did no explo e his adhesion in mo e de ail, i may be caused by unspeci ic in e ac ions wi h he ex acellula ma ix (mainly exopolysaccha ides) p o ec ing he bio ilm [ 26 ], simila ly o wha was obse ed o occu be ween phages and mucins. Ye , he mechanism by which he phages a ach o he mic obial bio ilm emains unclea . Ano he possibili y could be he p esence o he Fla obac e ium hos in he bio il e mic obiome. P e ious s udies on he RAS bio il e mic obiome ha e shown he p esence o Fla obac e ium spp in he bio il e s [ 24 ], bu , al hough his possibili y canno be excluded, we did no de ec F. columna e du ing he expe imen . Howe e , ou in i o esul s sugges a speci ic mic obiome plays a ole in phage adhesion in bio il e s, as adhesion was signi ican ly highe in mo ing bed ca ie media compa ed o ixed bed media. This may ha e also been caused by quan i a i e di e ences in he bio ilm o hese wo il e ma e ials. In conclusion, we ha e shown ha a phage can pe sis in RAS uni s o ex ended pe iods o ime, and he wa e ea men p ocesses o sys em componen s a e no a h ea o i al pa icles o ice e sa. Componen s o he sys em such as bio il e ca ie media and ish mucus may ha e a posi i e impac in phage pe sis ence, by selec i ely en iching phages compa ed o he wa e , and slowly eleasing hem o e ime. This migh be enough o delay he onse o a bac e ial disease ou b eak by keeping he in ec i e bac e ial doses low due o phage in ec ions. Fu he mo e, by selec ing phages a ge ing unwan ed con aminan s (pa hogens o o he bac e ia), his app oach could be used in de eloping phage-based echniques in managing he RAS bio il e mic obiome. Fo example, geosmin An ibio ics 2019,8, 192 6 o 9 compounds accumula e in ish issue, causing unwan ed ea hy as e and odo [ 27 ]. Phages in ec ing geosmin-p oducing S ep omyces species ha e al eady been shown o be capable o educing geosmin p oduc ion [28] and could be used in RAS sys ems o con ol p oduc quali y. 4. Ma e ial and Me hods 4.1. Expe imen al Si e This s udy was pe o med a he Laukaa ish a m o he Na u al Resou ces Ins i u e Finland. Six indi idual RAS uni s we e used, each con aining o a 500 L ish ank, a swi l sepa a o , a d um il e , a 150 L ixed and a mo ing bed bio il e (each illed wi h 70 L o plas ic ca ie media, RK Bioelemen s), and a ickling il e o ca bon dioxide emo al. In ixed-bed bio eac o s, he ca ie media lies s a ic in he bo om o he eac o , whe eas in mo ing bed bio eac o s, he ca ie media is agi a ed con inuously wi h ai . Each RAS uni had a o al wa e olume o 890 li e s, wi h a wa e low o 720 li e s pe hou and wa e enewal o 150 li e s pe day ( esul ing in he u no e a e o he en i e wa e olume in six days). Each ank was s ocked wi h 270–280 ainbow ou (Onco hynchus mykiss) (mean 50 g), and ish we e ed 300 g · d −1 . Fish heal h and eed in ake a e we e moni o ed daily. The loca ion o he ish a m and an example o he RAS sys em ( ish ank, solids emo al uni s, and mo ing and ixed bed bio eac o s) can be seen in Figu e 3. An ibio ics 2019, 8, x 6 o 9 6 media lies s a ic in he bo om o he eac o , whe eas in mo ing bed bio eac o s, he ca ie media is agi a ed con inuously wi h ai . Each RAS uni had a o al wa e olume o 890 li e s, wi h a wa e low o 720 li e s pe hou and wa e enewal o 150 li e s pe day ( esul ing in he u no e a e o he en i e wa e olume in six days). Each ank was s ocked wi h 270–280 ainbow ou (Onco hynchus mykiss) (mean 50 g), and ish we e ed 300 g·d −1 . Fish heal h and eed in ake a e we e moni o ed daily. The loca ion o he ish a m and an example o he RAS sys em ( ish ank, solids emo al uni s, and mo ing and ixed bed bio eac o s) can be seen in Figu e 3. Figu e 3. RAS a m loca ion and an example o he ish anks used. (A) Fish a m loca ion in Laukaa, Finland. (B) Fish ank (in g een) and solids emo al uni s. (C) Fixed and mo ing bed bio il e s. 4.2. FCL-2 Phage FCL-2, a myophage in ec ing Fla obac e ium columna e, was used as he model. I has been isola ed om he same ish a m whe e his expe imen was pe o med and shown o be e icien as a phage he apy agen in labo a o y condi ions [12,17,29]. The phage s ock used was p epa ed by in ec ing hos cul u es supplemen ed wi h mucin, as desc ibed p e iously [17]. B ie ly, he supe na an o o e nigh F. columna e s ain B185 cul u es made in 0.5× Shieh media supplemen ed wi h 0.1% pu i ied po cine mucin (Sigma, USA) we e ans e ed o s e ile lasks and in ec ed wi h FCL-2. Twen y- ou hou s la e , he whole cul u es we e cen i uged, and he supe na an il e ed o ob ain a s e ile high- i e s phage lysa e. Phages we e quan i ied by he double-aga o e lay me hod. Th ee hund ed mic oli e s o o e nigh cul u es o F. columna e s ain B185 we e added o h ee millili e s o so -Shieh aga supplemen ed wi h 0.1% mucin, and he mix u e was added o he op o Shieh-aga pla es. Then, dilu ions o he samples we e added as d ops o he op o he mix u e and phage plaques we e coun ed wo days la e . 4.3. Phage T ea men and Sampling Be o e s a ing he expe imen , wa e and ish mucus samples we e collec ed and es ed o gua an ee ha no p e-exis ing F. columna e o phage capable o in ec ing F. columna e we e al eady p esen a he sys ems. Fish gills we e also sampled o es o he p esence o F. columna e. Phage Figu e 3. RAS a m loca ion and an example o he ish anks used. ( A ) Fish a m loca ion in Laukaa, Finland. (B) Fish ank (in g een) and solids emo al uni s. (C) Fixed and mo ing bed bio il e s. 4.2. FCL-2 Phage FCL-2, a myophage in ec ing Fla obac e ium columna e, was used as he model. I has been isola ed om he same ish a m whe e his expe imen was pe o med and shown o be e icien as a phage he apy agen in labo a o y condi ions [ 12 , 17 , 29 ]. The phage s ock used was p epa ed by in ec ing hos cul u es supplemen ed wi h mucin, as desc ibed p e iously [ 17 ]. B ie ly, he supe na an o o e nigh F. columna e s ain B185 cul u es made in 0.5 × Shieh media supplemen ed wi h 0.1% pu i ied po cine mucin (Sigma, USA) we e ans e ed o s e ile lasks and in ec ed wi h FCL-2. Twen y- ou hou s An ibio ics 2019,8, 192 7 o 9 la e , he whole cul u es we e cen i uged, and he supe na an il e ed o ob ain a s e ile high- i e s phage lysa e. Phages we e quan i ied by he double-aga o e lay me hod. Th ee hund ed mic oli e s o o e nigh cul u es o F. columna e s ain B185 we e added o h ee millili e s o so -Shieh aga supplemen ed wi h 0.1% mucin, and he mix u e was added o he op o Shieh-aga pla es. Then, dilu ions o he samples we e added as d ops o he op o he mix u e and phage plaques we e coun ed wo days la e . 4.3. Phage T ea men and Sampling Be o e s a ing he expe imen , wa e and ish mucus samples we e collec ed and es ed o gua an ee ha no p e-exis ing F. columna e o phage capable o in ec ing F. columna e we e al eady p esen a he sys ems. Fish gills we e also sampled o es o he p esence o F. columna e. Phage p esence was e i ied by pla ing he samples on esh lawns o F. columna e s ain B185 made in so -aga -Shieh while F. columna e p esence was e i ied by pla ing he wa e samples in Shieh-aga con aining ob amycin. Fo phage ea men , wa e low was s opped in six RAS uni s. Then 470 millili e s o phage FCL-2 lysa e (2 × 10 10 p u/mL) was added o he ish anks (n=3). The inal phage concen a ion in each ank was 1 × 10 7 p u/mL a ime ze o. An equal olume o s e ile media was added o con ol anks (n=3). A e one hou , wa e samples we e collec ed, and wa e low was es a ed. Tank wa e , ish mucus, and plas ic ca ie media om mo ing and ixed beds we e sampled pe iodically. Wa e samples we e aken om he ish ank and om he mo ing and ixed beds and om he ickling il e (ae a ion uni ). Fish we e emo ed om he anks, eu hana ized wi h an o e dose o benzocaine, and skin mucus was sc apped wi h a glass slide. S e ile wa e was added o he mucus samples, o a inal olume o one millili e . Plas ic media om he mo ing and ixed beds we e collec ed and kep in alcon ubes imme sed in wa e om he espec i e bed. All samples we e p ese ed by he addi ion o chlo o o m un il being p ocessed. 4.4. E hics Fish handling was conduc ed acco ding o he Finnish Ac on he Use o Animals o Expe imen al Pu poses, unde pe mission ESAVI/8187/2018 g an ed o Lo a-Riina Sundbe g by he Na ional Animal Expe imen Boa d a he Regional S a e Adminis a i e Agency o Sou he n Finland. 4.5. Wa e Quali y Moni o ing Wa e quali y was moni o ed in he ish anks wi h an online moni o ing sys em consis ing o a spec ome e p obe measu ing o al o ganic ca bon, a pH p obe, and an op ical oxygen p obe. To al ammonia ni ogen (TAN), ni i e, and ni a e we e analyzed spec opho ome ically once a week om he ank ou le wa e . 4.6. Adso p ion Tes s on Plas ic Ca ie Media Clean au ocla ed plas ic media and media collec ed om he mo ing and ixed beds o con ol (no phage) ish anks we e used o es ing phage adso p ion and e en ion in i o . The media we e added o 50 mL alcon ubes and co e ed wi h nine ml o ank wa e . A e one hou gen ly shaking, 1 ml o a phage solu ion (2 × 10 5 p u/mL) was added, and he ubes e u ned o he shake . 10 µ L samples we e collec ed om he ubes pe iodically, and added in o 990 µ L o Shieh medium and a d op o chlo o o m, and main ained on ice un il he phage i e s we e de e mined. 4.7. S a is ical Analysis Da a analysis was made using he G aphPad so wa e e sion 8.0.1 (G aphPad So wa e, San Diego, USA). Unpai ed - es s we e employed o compa ing es ed condi ions in he app op ia e da ase s. An ibio ics 2019,8, 192 8 o 9 5. Conclusions Al hough o he phage species mus be es ed and he e ec o phage held in he sys em on bac e ial ou b eaks e alua ed, ou esul s highligh he ele ance o phage-based echniques in de eloping heal hy RAS sys ems. Phages could be applied as a ea men o ongoing in ec ions in a con en ional phage he apy app oach, bu he po en ial o p e en i e phage he apy could be mo e in e es ing. By en iching he RAS anks wi h he igh ype o phage, aqua ic animal a me s could achie e long- e m p o ec ion om common aquacul u e bac e ial diseases, p o ec ing hei p oduc ion while a he same ime a oiding he cos s and isks o an ibio ic use. Au ho Con ibu ions: All au ho s ook pa in he concep ualiza ion o he s udy. K.M., E.L., G.M.F.A. and J.P. pe o med he expe imen s. Resou ces we e p o ided by J.V. and L.-R.S. Da a was cu a ed by K.M., J.P., G.M.F.A. O iginal d a was p epa ed by G.M.F.A. All au ho s ook pa in w i ing and edi ing he manusc ip , and app o e he inal e sion. Funding: We acknowledge unding om he ishe ies inno a ion p og am o he Eu opean Ma i ime and Fishe ies Fund (EMFF), Academy o Finland (g an s #314939 and #321985), and Jane and Aa os E kko Founda ion. This wo k esul ed om he BONUS FLAVOPHAGE p ojec suppo ed by BONUS (A 185), unded join ly by he EU and Academy o Finland. Acknowledgmen s: The au ho s would like o hank he s a a he Na u al Resou ces Ins i u e Finland in Laukaa o help wi h he RAS sys em. Con lic s o In e es : Au ho s decla e no con lic o in e es . Re e ences 1. Ma ins, C.I.M.; Eding, E.H.; Ve degem, M.C.J.; Heinsb oek, L.T.N.; Schneide , O.; Blanche on, J.P.; d’O bcas el, E.R.; Ve e h, J.A.J. New de elopmen s in eci cula ing aquacul u e sys ems in Eu ope: A pe spec i e on en i onmen al sus ainabili y. Aquac. Eng. 2010. [C ossRe ] 2. Badiola, M.; Mendiola, D.; Bos ock, J. Reci cula ing aquacul u e sys ems (RAS) analysis: Main issues on managemen and u u e challenges. Aquac. Eng. 2012. [C ossRe ] 3. Pulkkinen, K.; Suomalainen, L.R.; Read, A.F.; Ebe , D.; Rin amäki, P.; Val onen, E.T. In ensi e ish a ming and he e olu ion o pa hogen i ulence: The case o columna is disease in Finland. P oc. R. Soc. B. Biol. Sci. 2010. [C ossRe ] 4. Kennedy, D.A.; Ku a h, G.; B i o, I.L.; Pu cell, M.K.; Read, A.F.; Win on, J.R.; Wa go, A.R. Po en ial d i e s o i ulence e olu ion in aquacul u e. E ol. Appl. 2016. [C ossRe ] [PubMed] 5. Sundbe g, L.R.; Ke ola, T.; Laan o, E.; Kinnula, H.; Bam o d, J.K.H.; Pen inen, R.; Mappes, J. In ensi e aquacul u e selec s o inc eased i ulence and in e e ence compe i ion in bac e ia. P oc. R. Soc. B. Biol. Sci. 2016. [C ossRe ] [PubMed] 6. Delabbio, J.; Mu phy, B.R.; Johnson, G.R.; McMullin, S.L. An assessmen o biosecu i y u iliza ion in he eci cula ion sec o o in ish aquacul u e in he Uni ed S a es and Canada. Aquacul u e 2004 ,242, 165–179. [C ossRe ] 7. Cabello, F.C.; God ey, H.P.; Tomo a, A.; I ano a, L.; Dölz, H.; Millanao, A.; Buschmann, A.H. An imic obial use in aquacul u e e-examined: I s ele ance o an imic obial esis ance and o animal and human heal h. En i on. Mic obiol. 2013,15, 1917–1942. [C ossRe ] [PubMed] 8. Shah, S.Q.A.; Cabello, F.C.; L’Ab é e-Lund, T.M.; Tomo a, A.; God ey, H.P.; Buschmann, A.H.; Sø um, H. An imic obial esis ance and an imic obial esis ance genes in ma ine bac e ia om salmon aquacul u e and non-aquacul u e si es. En i on. Mic obiol. 2014. [C ossRe ] [PubMed] 9. Muziasa i, W.I.; Pi känen, L.K.; Sø um, H.; S ed eld, R.D.; Tiedje, J.M.; Vi a, M. The esis ome o a med ish eces con ibu es o he en ichmen o an ibio ic esis ance genes in sedimen s below bal ic sea ish a ms. F on . Mic obiol. 2017. [C ossRe ] 10. Ko igh , K.E.; Chan, B.K.; Ko , J.L.; Tu ne , P.E. Phage he apy: A enewed app oach o comba an ibio ic- esis an bac e ia. Cell Hos Mic obe. 2019. [C ossRe ] 11. Wa s, G. Phage he apy: Re i al o he bygone an imic obial. Lance 2017. [C ossRe ] 12. Laan o, E.; Bam o d, J.K.H.; Ra an i, J.J.; Sundbe g, L.R. The use o phage FCL-2 as an al e na i e o chemo he apy agains columna is disease in aquacul u e. F on . Mic obiol. 2015. [C ossRe ] [PubMed]