Does the freshwater mussel Anodonta anatina remove the fish pathogen Flavobacterium columnare from water?

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/ Does he eshwa e mussel Anodon a ana ina emo e he ish pa hogen Fla obac e ium columna e om wa e ? © The Au ho (s) 2021 Published e sion Hajisa a ali, Mahsa; Aal onen, Sa i; Pulkkinen, Ka ja; Taskinen, Jouni Hajisa a ali, M., Aal onen, S., Pulkkinen, K., & Taskinen, J. (2022). Does he eshwa e mussel Anodon a ana ina emo e he ish pa hogen Fla obac e ium columna e om wa e ?. Hyd obiologia, 849(4), 1067-1081. h ps://doi.o g/10.1007/s10750-021-04769-6 2022 PRIMARY RESEARCH PAPER Does he eshwa e mussel Anodon a ana ina emo e he ish pa hogen Fla obac e ium columna e om wa e ? Mahsa Hajisa a ali .Sa i Aal onen .Ka ja Pulkkinen .Jouni Taskinen Recei ed: 3 Feb ua y 2021 / Re ised: 19 No embe 2021 / Accep ed: 22 No embe 2021 ÓThe Au ho (s) 2021 Abs ac Global decline o eshwa e mussels (Unionoida) is h ea ening biodi e si y and he essen- ial ecosys em se ices ha mussels p o ide. As il e - eeding o ganisms, eshwa e mussels emo e phy- oplank on and suspended pa icles om he wa e . By il e ing bac e ia, eshwa e mussels also dec ease pa hogen loads in he wa e . The objec i e o his s udy was o e alua e whe he he common eshwa e bi al e Anodon a ana ina (duck mussel) could emo e he bac e ial ish pa hogen Fla obac e ium columna e om he wa e . Mussels educed bac e ia in bo h o he wo expe imen s pe o med, so ha he bac e ial concen a ion a he end o he 96-h moni- o ing in mussel ea men s was only 0.3–0.5 imes ha o he con ols. Su p isingly, mussels did no educe algal cell concen a ion s a is ically signi i- can ly. Mussel beha io (shell openness, oo posi ion, and mo emen ) was no a ec ed by he p esence o bac e ia o algae, excep o biodeposi ion o ma ion, which was g ea es in algal- ed ea men s, ollowed by bac e ial- ed ea men s and con ols, espec i ely. The in es ines o bac e ia-incuba ed A. ana ina ha - bo ed F. columna e, sugges ing ha mussels inges ed he bac e ia. P esen esul s sugges ha eshwa e mussels may also ha e a po en ial o mi iga e aqua- cul u e pa hogen p oblems, as well as play a ole in wa e quali y managemen . Keywo ds Bio il a ion Columna is disease Duck mussel Unionidae Bac e ia Ecosys em se ices In oduc ion F eshwa e mussels (Unionoida) a e one o he mos h ea ened animal g oup in he wo ld (Lydea d e al., 2004; Lopes-Lima e al., 2017). In he ecen decades, a numbe o eshwa e mussel species ha e gone ex inc (Haag & Williams, 2014). Mussel popula ions a e declining due o pollu ion, habi a deg ada ion and agmen a ion, in oduced species, and he loss o obliga o y ish hos s (Bogan, 2008; Geis , 2010; Fe ei a-Rod ı ´guez e al., 2019), educing he biolog- ical di e si y in lakes and i e s. F eshwa e mussels ca y ou impo an ecosys em se ices ia hei il a ion and bu owing ac i i y, including nu ien ecycling and he deposi ion o o ganic ma e om he wa e column o he bo om in he o m o eces and pseudo eces (e.g., Vaughn, 2018). The Handling edi o : E ic R. La son Supplemen a y In o ma ion The online e sion con ains supplemen a y ma e ial a ailable a h ps://doi.o g/10.1007/ s10750-021-04769-6. M. Hajisa a ali (&)S. Aal onen  K. Pulkkinen J. Taskinen Depa men o Biological and En i onmen al Science, Uni e si y o Jy a ¨skyla ¨, P.O. Box 35, 40014 Jy a ¨skyla ¨n, Finland e-mail: [email p o ec ed] 123 Hyd obiologia h ps://doi.o g/10.1007/s10750-021-04769-6(0123456789().,- olV)(0123456789().,- olV) biodeposi ion o eces and pseudo eces ( he non- inges ed pa icles loosely w apped in mucus and ejec ed wi hou unde going he diges i e p ocess; Be g e al., 1996) by eshwa e mussels p o ides a nu ien - ich and easily assimila ed ood sou ce o ben hic mic obes and in e eb a es, which suppo s ish popula ions and links mul iple ophic le els in aqua ic ecosys ems (Howa d & Cu ey, 2006; Vaughn e al., 2008; Vaughn, 2018). The e o e, in addi ion o biodi e si y educ ion pe se, he decline o eshwa e mussels can ha e a ma ked e ec on he ecosys em. Bi al es’ po en ial ole in wa e quali y manage- men , hei ‘‘bi al e en i onmen al se ices’’, has al eady been ecognized in ma ine sys ems (e.g., nu ien apping connec ed wi h mussel p oduc ion) (Nielsen e al., 2016; Clemen s & Comeau, 2019; Taylo e al., 2019; Ko a e al., 2020). Mussels’ e ec i eness in nu ien and pa icle emo al and in was ewa e ea men has also been shown in esh- wa e sys ems (S aye , 2014; Lumme e al., 2016; Mezzano e e al., 2016; Hoellein e al., 2017; K eege e al., 2018). Thus, eshwa e mussels can po en ially be u ilized in wa e quali y managemen o educe nu ien e luen s and eu ophica ion (K eege e al., 2018). Depending on hei abili y o emo e bac e ia, eshwa e mussels could also be used o educe disease ou b eaks. Se e al esea che s ha e ound ha he d eissenid mussel D eissena polymo pha (Pallas, 1771) can clea he pa hogenic bac e ium Esche ichia coli (Migula, 1895) Cas ellani & Chalme s, 1919 om wa e in labo a o y expe imen s (Sil e man e al., 1995), in ield condi ions (Co ne e al., 1995), and om ea ed e luen s o municipal was e wa e (Mezzano e e al., 2016). Unionidae eshwa e mussels a e also capable o il e ing bac e ia (Jø gen- sen e al., 1984; K yge & Riisga ˚ d, 1988; Nichols & Ga ling, 2002; Ch is ian e al., 2004Vande ploeg e al., 2011b), including E. coli (Sil e man e al., 1995; Ismail e al., 2015). Howe e , he abili y o eshwa e mussels o elimina e bac e ial ish pa ho- gens has no been in es iga ed. The aim o he p esen s udy was o in es iga e he capabili y o he duck mussel, Anodon a ana ina (Linnaeus, 1758), o emo e he bac e ium Fla obac- e ium columna e Be na de & G imon 1989 om he wa e column in labo a o y condi ions. Anodon a ana ina is a unionid mussel ha commonly occu s in he lakes and i e s o Eu ope (Lopes-Lima e al., 2016). This mussel has been shown o il e Diplos omum pseudospa haceum Niewiadomska, 1984 la ae om wa e , he eby educing he ans- mission o his ema ode pa asi e om aqua ic snails o i s ish hos (Gopko e al., 2017). Mo e impo an ly, A. ana ina has he po en ial o il e and inges colonial and ilamen ous cyanobac e ia (Bon es e al., 2007). Fla obac e ium columna e is a eshwa e bac e ium and an oppo unis ic ish pa hogen ha causes colum- na is disease (‘‘wa m wa e disease’’) in ish. Fla obac e ium columna e can in ec a ange o ish species h oughou he wo ld, such as salmon, ou , ainbow ou , channel ca ish, ca p, pe ch, pike, eel, and ilapia (Decle cq e al., 2013). Columna is disease is one o he mos ha m ul diseases in channel ca ish, Ic alu us punc a us (Ra inesque, 1818) a ming in he USA, leading o app oxima ely $30 million yea ly losses (Wagne e al., 2006). I is also a e y di icul p oblem in young salmonid cul u e, causing high mo ali y and he use o an ibio ics (Pulkkinen e al., 2010). This pa hogen can also be ound in na u al lakes and i e s (Kun u e al., 2012) and has been isola ed om issues o he unionid mussel Villosa i is (I.Lea, 1829) (Clinch Ri e , USA; S a lipe e al., 2008). Ou hypo hesis was ha A. ana ina would be able o emo e F. columna e om he wa e , mani es ed as a lowe a e age bac e ial concen a ion in he wa e wi h mussel ea men as compa ed o he con ol wa e wi hou mussel. As a me hodological con ol o he eeding o mussels, we es ablished a ea men g oup ed wi h a comme cial mic oalgae p oduc commonly used o mussel eeding. As he bac e ial suspension may a ec he beha io o mussels, especially he il a ion ac i i y, moni o ing o mussel beha io was included o he s udy. In addi ion, as he biodeposi o ma ion can indica e il a ion o ma e om wa e , biodeposi ion o ma ion was also moni- o ed du ing he expe imen . The p esen s udy is he i s e o o e alua e he e iciency o eshwa e mussels o emo e he ha m ul bac e ial ish pa hogen om he wa e . Ma e ials and me hods Mussel collec ion and p epa a ion In o al, 30 A. ana ina indi iduals (Table 1) we e collec ed om Lake Koija ¨ i, Finland (60.97°N, 123 Hyd obiologia 23.73°E) du ing Sep embe 2016. The mussels we e anspo ed o Konne esi Resea ch S a ion, Uni e si y o Jy a ¨skyla ¨and we e kep in a 163 L low- h ough ank wi h sand on he bo om un il he expe imen s began. The mussels we e no ed, bu hey ecei ed ood om incoming Lake Konne esi wa e du ing he main enance pe iod. In Janua y 2017, he mussels we e anspo ed o he labo a o y o he Depa men o Biological and En i onmen al Science, Uni e si y o Jy a ¨skyla ¨. The mussels we e placed in a la ge bucke o well wa e , ae a ed wi h an aqua ium ai pump, and kep on a 12 h ligh –da k cycle a *17 ° C. In he labo a o y du ing he p e-expe imen main- enance, he mussels we e ed ad libi um wi h a comme cial phy oplank on p oduc (Shell ish Die 1800Ò, SFD, Reed Ma icul u e, USA) con aining six ma ine mic oalgae: Isoch ysis sp., Pa lo a sp., Te- aselmis sp., Chae oce os calci ans (Paulsen) H.Takano, Thalassiosi a weiss logii (G unow) G.A.F yxell & Hasle, and Thalassiosi a pseudonana Hasle & Heimdal.The mussels we e kep wi hou ood o 24 h be o e s a ing he expe imen s un il hey no longe eleased exc emen . The wa e empe a u e a Konne esi Resea ch S a ion du ing he main enance pe iod was ini ially 17 °C in Sep embe ; i g adually declined o 3 °C in Decembe 2016, bu i was a i icially g adually inc eased o 17 °C in he las wo weeks be o e he expe imen s, which we e conduc ed a 17 °C. P epa a ion o bac e ial suspension, Shieh medium and aga pla es The F. columna e s ain (B613) used in he expe i- men s was isola ed om a wa e sample aken om he sho e o Lake Kynsi esi (62°2500 N, 26°1500 E) in 2014; i was hen cul i a ed in a modi ied Shieh medium (Song e al. 1988) and s o ed a -80 °C wi h 10% glyce ol and 10% e al cal se um. In o de o p epa e he bac e ial solu ion o he expe imen s, 20 ll om he ozen s ock was added o 5 ml o modi ied Shieh medium and incuba ed wi h con inuous s i ing a 120 pm o 24 h a oom empe a u e. One ml o he o e nigh -g own bac e ial suspension was u he added o 9 ml o modi ied Shieh medium, wi h ano he incuba ion a oom empe a u e o e nigh . The op ical densi y (OD) o he bac e ial suspension was measu ed wi h a spec opho ome e a 595 nm and con e ed o a bac e ial concen a ion using a p e- es ablished ela ionship be ween OD and colony- o ming uni s (CFUs); his was done o allow he esea che s o adjus he ini ial bac e ia concen a ion le els du ing he expe imen s. Modi ied Shieh medium was p epa ed acco ding o Song e al. (1988) o se e as a selec i e g ow h medium o isola e F. columna e. Aga pla es we e used o pla e-coun he iable bac e ial concen a ions; he e, 10 g o aga powde we e added pe li e and supplemen ed wi h ob amycin in o de o inhibi he g ow h o bac e ia o he han F. columna e (Decos e e e al., 1997). Expe imen 1 The i s expe imen o e alua e A. ana ina’s abili y o emo e F. columna e om wa e began in Feb ua y 2017 and included 5 ea men s wi h 5 eplica es each ( eplica e = 1 sedimen /sand- ee aqua ium illed wi h 5 L o well wa e wi h o wi hou a mussel indi idual; see below). A schema ic p esen a ion o he expe i- men al design is gi en in he Supplemen a y Fig. 1. The aqua ia we e ae a ed and kep on a ligh –da k cycle o 12 h:12 h a *17 °C. The expe imen al ea men s we e as ollows: Table 1 The numbe o mussels (Anodon a ana ina), a e age leng h (mm, mean ±SD) and age (y, mean ±SD), sex (F= emale, M= male), in ensi y o ema ode in ec ion (Rhipidoco yle ennica) (ze o, ligh , mode a e, hea y; see ex o ca ego iza ion), and occu ence o glochidia in mussels used in he 1s and 2nd expe imen Numbe o mussels Leng h Age Sex T ema ode Glochidia Mean Mean F M Unknown Ze o Ligh Mode a e Hea y Numbe 1s expe imen 15 83 ±4 5.6 ±16 4 5 1 4 4 6 0 2nd expe imen 9 87 ±4 6.1 ±11 4 4 2 2 2 3 0 123 Hyd obiologia (1) Bac e ia and mussel: A known concen a ion (5 910 5 CFU ml -1 )o F. columna e wi h an indi idual mussel o e alua e he po en ial o A. ana ina o il e ou F. columna e. (2) Bac e ia con ol: A known concen a ion (5 910 5 CFU ml -1 )o F. columna e wi hou a mussel was used o s udy changes in he bac e ial concen a ion o he wa e in he absence o he mussel. The null hypo hesis was ha he e was no di e ence in bac e ial concen a ion be ween ea men s (1) and (2). (3) Algae and mussel: A known concen a ion o algae wi h a mussel; his se ed as a e e ence ea men o s udy A. ana ina’s algal il a ion. (4) Algae con ol: A known concen a ion o algae wi hou a mussel was used o ack changes in algae concen a ion in he wa e in he absence o he mussel. In ea men s (3) and (4), 35 ml o Shell ish Die 1800Òwas added o 1.1 L wa e ; hen, 100 ml o his suspension was added o each aqua ium. The null hypo hesis was ha he e was no di e ence in algae concen a ion be ween ea - men s (3) and (4). (5) Mussel con ol: A mussel wi hou F. columna e and wi hou algae was used o s udy mussel beha io wi hou he p esence o bac e ia and algae. The null hypo hesis was ha he e was no di e ence in beha io s be ween ea men s (1), (3), and (5). Expe imen 2 The second expe imen began in Ma ch 2017 o e i y he esul s o Expe imen 1 in ega d o bac e ial il a ion; i used a bac e ial inocula ion ha was an o de o magni ude lowe han in Expe imen 1 (see he Resul s sec ion and Fig. 1). A schema ic p esen- a ion o he expe imen al design is gi en in he Supplemen a y Fig. 2. The aqua ia we e ae a ed and kep on a ligh –da k cycle o 12 h:12 h a *17 °C. The expe imen al ea men s we e as ollows: (1) Bac e ia and mussel: A known concen a ion o F. columna e wi h an indi idual mussel o e alua e A. ana ina’s po en ial o il e F. columna e; 9 eplica e aqua ia wi h a single mussel in each we e used. (2) Bac e ia con ol: A known concen a ion o F. columna e wi hou a mussel was used o s udy changes in bac e ial concen a ion in he absence o he mussel; 5 eplica e aqua ia we e used. The null hypo hesis was ha he e was no di e - ence in bac e ial concen a ion be ween ea men s (1) and (2). Bac e ial sampling om he wa e Wa e samples we e aken om all he aqua ia p io o s a ing each expe imen o exclude he possibili y o p io con amina ion by F. columna e. The concen a- ion o bac e ia was e alua ed om 100 ll wa e samples om ea men s wi h added bac e ia (bac e ia and mussel, bac e ia con ol) a ime poin s 0 h, 6 h, and 24 h; 1 sample pe aqua ium was e alua ed. In o de o dec ease he obse a ional e o associa ed wi h educing bac e ial le els a ime poin s 48 h and 96 h, he numbe o epea ed wa e samples aken pe aqua ium was inc eased o h ee. In Expe imen 2, 100 ll o wa e was aken a ime poin s 0 h, 48 h, and 96 h; 2 epea ed samples we e aken pe aqua ia. A dilu ion se ies (10 -1 –10 -7 ) o each wa e sample was p epa ed and cul u ed on Shieh aga pla es; colony o ma ion was checked a e 48 h. A he end o bo h expe imen s, ea men s wi hou bac e ia we e checked as desc ibed abo e (2 9100 ll wa e sam- ples pe aqua ium) o exclude po en ial con amina ion om added bac e ia du ing he expe imen . Bac e ial sampling om in es ines and gonads To examine i mussels can inges bac e ia and become in ec ed wi h F. columna e, bac e ial samples we e aken om in es ine and gonad o he mussels om all ea men s a he end o bo h expe imen s. To a oid bac e ial con amina ion om he aqua ium wa e , mussel shell su aces we e sp ayed wi h 70% e hanol be o e dissec ion. A e opening he al es, he mussel body was cu in 2 so ha a c oss-sec ion h ough he gonads—and he in es ine winding h ough he gonads—could be accessed. Then, he bac e ial sam- ples we e collec ed wi h a s e ile loop and cul u ed on Shieh aga pla es supplemen ed wi h ob amycin. In Expe imen 1, bac e ial sampling was done i s om he in es ine and hen om he gonads. In Expe imen 2, he sampling o de was e e sed o minimize 123 Hyd obiologia possible con amina ion o he gonads om he in es ine. Algal cell sampling To in es iga e A. ana ina’s il a ion capabili y, algal cell concen a ion was moni o ed om 3 eplica e wa e samples pe aqua ium a di e en ime poin s (0 h, 6 h, 24 h, 48 h) using a mic oscope coun ing chambe (Bu ¨ ke , Ma ien eld, Ge many, dep h = 0.1 mm, squa e wid h = 0.25 mm). A e 24 h, algal cell concen a ion declined sha ply in bo h ea men s, and moni o ing was discon inued. Mussel beha io In o de o check he possible e ec o he bac e ial and algal die s on he mussels’ beha io ( o example, a comple e inac i a ion o mussels when incuba ed wi h bac e ia), in bo h expe imen s, A. ana ina beha io was eco ded 3 imes pe day a 9:00, 12:00, and 15:00 h in a 2 min isual inspec ion o 99 h. The ai s ha we e moni o ed included shell openness, oo posi ion (ex ension o con ac ion), mo emen , and biodeposi ion ( eces and pseudo eces) o ma ion since he p e ious moni o ing. Shell open- ness was sco ed as ollows: ully closed (siphons comple ely in isible) = 0, sligh ly opened (siphons a e s ill in isible) = 1, hal open (siphons isible, pa ly p o uding) = 2, ully open (siphons ully isible, ully p o uding) = 3. Foo posi ion was g aded as ollows: oo in = 0, oo pa ly ou = 1, oo ou = 2. Mo emen (posi ion changes in he mussel) was moni o ed by compa ing mussel posi ion o a ma k on he lid o each aqua ium deno ing he posi ion a he p e ious inspec ion. I was sco ed as ollows: mo emen = 1, no mo emen = 0. Biodepo- si ion p esence was classi ied in o 2 ca ego ies: Fig. 1 Bac e ial concen a ion (CFU ml -1 ) (A) in aqua ia wi h mussels and wi hou mussels a di e en ime poin s (N = 5) (mean ±SD) in he 1s expe imen , in he 2nd expe imen (B), and algal cell concen a ion (cell ml -1 )(C) in aqua ia wi h mussels and wi hou mussels a di e en ime poin s (N = 5) (mean ±SD) 123 Hyd obiologia wi hou biodeposi ion = 0, and wi h biodeposi ion = 1. T ema ode pa asi ism, sex, and ep oduc ion o mussels Possible ema ode pa asi ism (pa asi e species, in en- si y o in ec ion) in he mussels was examined a he end o each expe imen when he bac e ial sampling om he in es ine and gonads was pe o med; his was done because ema odes can ha e a ma ked in luence on mussel physiology (Taskinen, 1998; Jokela e al., 2005). The gonads a e he main si e o ema ode in ec ion in A. ana ina (Taskinen e al., 1997). The e o e, a 2–3 mm slice ac oss he gonads was dissec ed a e he bac e ial sampling was conduc ed, p essed be ween la ge glass pla es, and examined mic oscopically o ema ode spo ocys s using ans- mi ed ligh . This was accompanied wi h he sex de e mina ion o he mussel and an es ima ion o he in ensi y o he in ec ion (Taskinen e al., 1994; Taskinen & Val onen, 1995). The in ensi y o he ema ode in ec ion was sco ed as ollows: unin- ec ed = 0, ligh in ec ion (one spo ocys ubule o one g oup o spo ocys ubules) = 1, mode a e in ec ion om [1 g oup o spo ocys ubules o almos com- ple e eplacemen o mussel gonad issue by spo o- cys s = 2, and hea y in ec ion (gonad issue comple ely eplaced by ema ode spo ocys s) = 3 (Taskinen e al., 1994). S a is ical analyses The s a is ical analyses o bac e ial and algal concen- a ions we e pe o med using IBM SPSS so wa e ( e sion 24.0.0). Cell concen a ions ( he CFUs o F. columna e and he coun s o algal cells) we e used as esponse a iables; aqua ia we e used as he s a is ical uni s. Gene al Linea Model (GLM) epea ed mea- su es analyses we e pe o med o compa e di e ences in bac e ial and algal concen a ion in ime ( he wi hin-subjec ac o ) be ween ea men s wi h and wi hou a mussel ( he be ween-subjec ac o ). The di e ences be ween di e en ime poin s we e u he examined using wi hin-subjec con as (‘‘simple’’). Le ene’s es was used o assess he homogenei y o a iance, and he G eenhouse–Geisse co ec ion was applied o co ec he numbe o deg ees o eedom when he sphe ici y assump ion was no alid. Di e ences in mussel beha io (shell openness, oo posi ion, mo emen , biodeposi ion o ma ion) a di - e en ime poin s be ween he ea men s (mussel only, mussel ?bac e ia, mussel ?algae) we e ana- lyzed using mixed e ec s logis ic eg ession (GLMM) in R ( e sion 4.0.2.) s a is ics so wa e and Lme4 package (R De elopmen Co e Team, 2015). The in e dependence o epea ed obse a ions om he same indi idual mussel du ing he expe imen was accoun ed o by ea ing mussel iden i y as a andom a iable. To be able o i he models using binomial (logi ) dis ibu ion, all beha io al da a we e ecoded as 0 o 1 be o e he analysis. Thus, shell openness was eca ego ized as 0 ( al es ully closed) and 1 ( al es sligh ly, hal o ully open), and oo posi ion as 0 ( oo in) and 1 ( oo pa ly o comple ely ou ). The di e ences in he in ensi y o ema ode in ec ion be ween ea men g oups in Expe imen 1 sco ed om ze o (unin ec ed) o 3 (hea ily in ec ed) could no be es ed wi h a Chi-squa e es because expec ed alues we e less han 5. Ins ead, we used Fishe s exac es ex ended o 3 93 con ingency able (McDonald , 2014). In addi ion, he di e ence in he mean in ensi y o he ema ode in ec ion was es ed be ween he ‘‘bac e ia and mussel’’ ea men s in bo h expe imen s wi h a - es . Di e ences be ween ea men s we e conside ed signi ican i p 0.05. Resul s E ec o mussel il a ion on bac e ial concen a ion (Expe imen 1) A he beginning o Expe imen 1, he mean concen- a ion o bac e ia was simila in bo h ea men g oups wi h F. columna e (Fig. 1A). Du ing he nex 48 h, he bac e ial concen a ion inc eased in bo h ea men s, bu he inc ease was mo e p onounced in he ea men wi hou a mussel han in he ea men wi h a mussel. Be ween 48 and 96 h, he bac e ial concen a ion dec eased in bo h ea men s (Fig. 1A). The bac e ial concen a ion was lowe in he ea men wi h a mussel han in he ea men wi hou a mussel (Table 2, be ween-subjec e ec s, F 1,8 = 16.31, p= 0.004). Howe e , his di e ence became signi ican only a e 96 h o he expe imen , as indica ed by he signi ican in e ac ion be ween ime and ea men and he con as s be ween he le els o he wi hin-subjec 123 Hyd obiologia ac o and ea men ( ime x ea men ) (Table 2). Thus, i ook 48 h om he beginning o he expe - imen be o e he emo al o he bac e ia by he mussels was appa en (Fig. 1A). The ea e , he mus- sels educed he bac e ia so ha he bac e ial concen- a ion in mussel ea men was 0.7–0.5 imes ha o he con ols a 48 and 96 h, espec i ely (Fig. 1A). Thus, he mussels emo ed hal he bac e ia wi hin 96 h as compa ed o he con ols. E ec o Mussel Fil a ion on Bac e ial Concen a ion (Expe imen 2) In Expe imen 2, a simila pa e n in bac e ial concen a ions o hose in Expe imen 1 was obse ed. The bac e ial concen a ions we e simila in he beginning o he expe imen in he ea men s wi h and wi hou a mussel, espec i ely (Fig. 1B). A 48 h, he mean concen a ion o F. columna e had inc eased in bo h ea men s, bu wi h a mussel he inc ease was app oxima ely hal o ha in he ea men wi hou a mussel. Be ween 48 and 96 h, he bac e ial Table 2 Resul s o GLM epea ed measu es ANOVA o he mean concen a ion o Fla obac e ium columna e (CFU ml -1 ) in he wo ea men s (wi h and wi hou mussel) a di e en ime poin s in he 1s and 2nd expe imen Sou ce Fi s expe imen Second expe imen d MS F pd MS F p Be ween subjec s In e cep 1 2.95E?12 654.47 0.001 1 4.75E?11 180.30 0.001 T ea men 1 7.36E?10 16.31 0.004 1 489E?10 18.55 0.001 E o 8 3.61E?10 12 2.63E?09 Wi hin subjec s Time 2.335 1.78E?12 61.63 0.001 2 2.83E?11 61.17 0.001 Time 9 ea men 2.335 1.52E?11 a 5.26 0.012 2 2.74E?10 5.91 0.008 E o 18.676 2.88E?10 a 24 4.64E?09 Wi hin-subjec con as s b Time Time 6 h s ime 0 h 1 8.26E?09 3.79 0.087 Time 24 h s. ime 0 h 1 1.55E?12 47.93 0.001 Time 48 h s. ime 0 h 1 5.95E?12 139.97 0.001 1 9.09E?11 107.46 0.001 Time 96 h s. ime 0 h 1 2.11E?12 61.64 0.001 1 4.22E?09 0.46 0.511 Time 9 ea men Time 6 h s ime 0 h 1 6.76E?07 0.03 0.865 Time 24 h s. ime 0 h 1 1.37E?09 0.04 0.842 Time 48 h s. ime 0 h 1 1.72E?11 4.06 0.079 1 1.09E?11 12.92 0.004 Time 96 h s. ime 0 h 1 4.19E?11 12.21 0.008 1 3.36E?10 3.64 0.081 E o Time 6 h s ime 0 h 8 2.18 E?09 Time 24 h s. ime 0 h 8 3.23 E?10 Time 48 h s. ime 0 h 8 4.25 E?10 12 8.46 E?09 Time 96 h s. ime 0 h 8 3.43 E?10 12 9.22 E?09 D = deg ees o eedom, MS = mean squa es a G eenhouse–Geisse co ec ed alues b Wi hin-subjec con as o he in e ac ion be ween ime (hou ) and ea men s. The con as is ‘‘simple’’, which compa es each ime poin o he i s ime poin ( ime 0 h). No e ha o he 2 nd expe imen measu emen s we e made a ime poin s 0, 48 and 96 h 123 Hyd obiologia concen a ions dec eased in bo h ea men s (Fig. 1B). The bac e ial concen a ion was lowe in he ea men wi h a mussel han in he ea men wi hou a mussel (Table 2, be ween-subjec e ec s, F 1,12 = 18.55, p= 0.001). Con as s be ween he le els o he wi hin-subjec ac o and ea men ( ime x ea men ) indica ed ha his di e ence was signi ican a e 48 h bu no a e 96 h o beginning he expe imen (Table 2). Mussels educed he bac e ial concen a ion in mussel ea men o 0.5 and 0.3 imes ha o he con ols a 48 and 96 h, espec i ely (Fig. 1B). Thus, he mussels emo ed wo- hi ds o he bac e ia wi hin 96 h as compa ed o he con ols. Gonad and In es ine Analysis (Expe imen s 1 and 2) In Expe imen 1, F. columna e was isola ed bo h om he mussels’ in es ines and gonads in he ea men wi h added bac e ia, bu no in he o he ea men g oups (i.e., he mussels wi h algae and he mussels kep in clean wa e ). A he indi idual (mussel) aqua ium le el, among he 5 eplica es o mus- sel ?bac e ium ea men , 3 mussels had F. colum- na e in bo h he in es ine and he gonads, 1mussel had F. columna e in he in es ine bu no in he gonads, and 1 es ed nega i e (no bac e ium in ei he he in es ine o in he gonads). In Expe imen 2, among he 9 eplica es o mussel and bac e ia, F. columna e was only ound in he in es ines o 8 indi iduals; i was no ound in he gonads. E ec o Mussel Fil a ion on Algal Cell Concen a ion (Expe imen 1) The mean concen a ion o algal cells (Shell ish Die ) dec eased be ween he s a o he expe imen and 48 h in bo h ea men s wi h and wi hou a mussel (Fig. 1C, Table 3, wi hin-subjec e ec [ ime], F 3,24 = 83.20, p= 0.001). Howe e , he p esence o A. ana ina did no a ec algal cell concen a ion (Table 3, F1,8 = 1.872, p= 0.208). Mussel Beha io Shell openness (Expe imen s 1 and 2) In bo h expe imen s, he mussels in all ea men s had open o pa ly open al es a he beginning o he expe imen . The mussels g adually closed hei al es du ing he i s 24 h bu s a ed opening hei shells again owa d he end o he expe imen (Fig. 2A and E). In Expe imen 1, he p esence o bac e ia did no a ec he mussels’ shell openness as compa ed o he mussels wi h algal cells o hose in clean wa e (GLMM, p [0.05, Table 4). Foo posi ion (Expe imen s 1 and 2) Foo posi ion sco es e ealed ha he mussels kep hei ee subs an ially con ac ed (Fig. 2B and F). In Expe imen 1, he bac e ial p esence did no a ec he oo posi ion as compa ed o he mussels exposed o he algae o hose kep in clean wa e (GLMM p[0.05, Table 4). Mo emen (Expe imen s 1 and 2) Anodon a ana ina in all ea men g oups in bo h expe imen s exhibi ed a con inuous and cohe en pa e n in mo emen beha io h oughou he expe - imen (Fig. 2C and G). The p esence o bac e ia had a sligh e ec in dec easing mussels’ mo emen ac i i y (GLMM, he ela i e change in compa ison o mussel only ea men 0.69 ±0.36, p = 0.053, Table 4), while he mo emen ac i i y o mussels wi h algae was lowe han mussels in clean wa e (GLMM, he ela i e change in compa ison o mussel only ea - men 0.75 ±0.36, p = 0.036, Table 4). Table 3 Resul s o GLM epea ed measu es ANOVA o he mean o algal cell concen a ion (cell ml -1 ) in he wo ea - men s wi h and wi hou mussel a di e en ime poin s. D = deg ees o eedom, MS = mean squa es Sou ce d MS Fp Be ween subjec s In e cep 1 3.68E?13 1681.82 0.001 T ea men 1 4.10E?10 1.87 0.208 E o 8 2.19E?10 Wi hin subjec s Time 3 2.46E?12 83.16 0.001 Time 9 ea men 3 8.33E?10 2.82 0.061 E o 24 2.96E?10 123 Hyd obiologia subadul unionidae. Jou nal o Shell ish Resea ch 21: 875–881. Nielsen, P., P. C an o d, M. Maa & J. Pe e sen, 2016. Magni- ude, spa ial scale and op imiza ion o ecosys em se ices om a nu ien ex ac ion mussel a m in he eu ophic Ski e Fjo d, Denma k. Aquacul u e En i onmen In e ac- ions 8: 311–329. Pulkkinen, K., L.-R. Suomalainen, A. F. Read, D. Ebe , P. Rin ama ¨ki & E. T. Val onen, 2010. In ensi e ish a ming and he e olu ion o pa hogen i ulence: he case o columna is disease in Finland. P oceedings o he Royal Socie y B: Biological Sciences Royal Socie y 277: 593–600. Rubiolo, J. A., L. M. Bo ana, & P. Ma inez, 2019. Insigh s in o Mussel Mic obiome. Mic obial Communi ies in Aquacul- u e Ecosys ems 95–120. R De elopmen Co e Team, 2015. R: A Language and En i- onmen o S a is ical Compu ing, R Founda ion o S a- is ical Compu ing, Vienna: Seide e , L., C. Da is, F. Robb & R. Newell, 1984. U ilisa ion o bac e ia as ni ogen esou ce by kelp-bed mussel Cho - omy ilus me idionalis. Ma ine Ecology P og ess Se ies 15: 109–116. Sil e man, H., E. C. Achbe ge , J. W. Lynn & T. H. Die z, 1995. Fil a ion and u iliza ion o labo a o y-cul u ed bac e ia by D eissena polymo pha,Co bicula luminea, and Ca un- culina exasensis. The Biological Bulle in 189: 308–319. Sil e man, H., S. J. Nichols, J. S. Che y, E. Achbe ge , J. W. Lynn & T. H. Die z, 1997. Clea ance o labo a o y- cul u ed bac e ia by eshwa e bi al es: di e ences be ween len ic and lo ic unionids. Canadian Jou nal o Zoology 75: 1857–1866. S a lipe , C. E., R. J. Ne es, S. D. Hanlon & P. Whi ing on, 2008. A su ey o he indigenous mic obio a (bac e ia) in h ee species o mussels om he Clinch and Hols on Ri e s, Vi ginia. Jou nal o Shell ish Resea ch 27: 1311–1317. S aye , D. L., 2008. F eshwa e mussel ecology: a mul i ac o app oach o dis ibu ion and abundance, Uni e si y o Cali o nia P ess, Be keley: S aye , D. L., 2014. Unde s anding how nu ien cycles and eshwa e mussels (Unionoida) a ec one ano he . Hyd obiologia 735: 277–292. Taskinen, J., 1998. In luence o ema ode pa asi ism on he g ow h o a bi al e hos in he ield. In e na ional Jou nal o Pa asi ology 28: 599–602. Taskinen, J., T. Ma ¨kela ¨& E. T. Val onen, 1997. Exploi a ion o Anodon a piscinalis (Bi al ia) by ema odes: pa asi e ac ics and hos longe i y. Annales Zoologici Fennici 34: 37–46. Taskinen, J. & E. T. Val onen, 1995. Age-, size-, and sex- speci ic in ec ion o Anodon a piscinalis (Bi al ia: Unionidae) wi h Rhipidoco yle ennica (Digenea: Bucephalidae) and i s in luence on hos ep oduc ion. Canadian Jou nal o Zoology 73: 887–897. Taskinen, J., E. T. Val onen & D. I. Gibson, 1991. S udies on bucephalid digeneans pa asi ising molluscs and ishes in Finland I. Ecological da a and expe imen al s udies. Sys- ema ic Pa asi ology 19: 81–94. Taskinen, J., E. T. Val onen & T. Ma ¨kela ¨, 1994. Quan i y o spo ocys s and seasonali y o wo Rhipidoco yle species (Digenea: Bucephalidae) in Anodon a piscinalis (Mol- lusca: bi al ia). In e na ional Jou nal o Pa asi ology 24: 877–886. Taylo , D., C. Sau el, P. Nielsen & J. K. Pe e sen, 2019. P o- duc ion cha ac e is ics and op imiza ion o mi iga ion mussel cul u e. F on ie s in Ma ine Science F on ie s 6: 698. Tu le-Rayc a , S. & J. Acke man, 2018. Does size ma e ? Pa icle size s. quali y in bi al e suspension eeding. F eshwa e Biology 63: 1560–1568. Vande ploeg, H., J. R. Liebig & T. F. Nalepa, 1995. F om picoplank on o mic oplank on: empe a u e-d i en il a- ion by he unionid bi al e Lampsilis adia a siliquoidea in Lake S . Clai . Canadian Jou nal o Fishe ies and Aqua ic Sciences 52: 63–74. Vaughn, C. C., 2018. Ecosys em se ices p o ided by esh- wa e mussels. Hyd obiologia 810: 15–27. Vaughn, C., S. Nichols & D. Spoone , 2008. Communi y and oodweb ecology o eshwa e mussels. Jou nal o he No h Ame ican Ben hological Socie y 27: 409–423. Wagne , B. A., D. J. Wise, L. H. Khoo & J. S. Te hune, 2006. The Epidemiology o bac e ial diseases in ood-size channel ca ish. Jou nal o Aqua ic Animal Heal h 18: 263–272. Publishe ’s No e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps and ins i u ional a ilia ions. 123 Hyd obiologia