Roseobacter-Clade Bacteria As Probiotics in Marine Larvae- Culture

Home , Roseobacter, Roseobacter litoralis

Roseobacter-clade bacteria as probiotics in marine larvae- culture

______

Torben Grotkjær

PhD Thesis

Technical University of Denmark

Department of Bioengineering

July 2016

Preface

This PhD Study was carried out at the Department of Systems Biology at the Technical University of Denmark from April 2013 to June 2016 under the supervision of Professor Lone Gram. Co-supervisors were post doc Paul D’Alvise and post doc Mikkel Bentzon-Tilia.

The PhD study was funded by the Danish Council for Strategic Research | Programme Commission on Health, Food and Welfare (12-132390) to Lone Gram, and by grant VKR023285 from the Villum Kann Rasmussen foundation to Lone Gram

A part of the study was carried out at The Artemia Reference Center at the University of Ghent, under the supervision of Professor Peter Bossier and Kristof Dierckens. This part was separately funded by European Community’s Seventh Framework Programme (FP7/2007- 2013) under the grant agreement n° FP7-262336-AQUAEXCEL

The results of the PhD study are described in three published articles included in this thesis

Torben Grotkjær

July 2016

Summary

Disease caused by fish pathogenic bacteria can cause large scale crashes in marine fish larval rearing units. One of the biggest challenges for aquaculture is the management of these bacterial outbreaks. Vaccines can be admitted to fish but only the juvenile and the adult fish because they need to have a mature immune system. This means that the larvae of the fish, until they are 2-3 weeks old are more prone to bacterial infections. A short term solution is antibiotics but this leaves way for the selection for antibiotic resistance among the pathogenic bacteria, which again can be transferred to human pathogens. Alternatives are therefore needed and one could be the use of probiotic bacteria. Marine bacteria from the Roseobacter clade (Phaeobacter inhibens) have shown great potential as probiotic bacteria, and we have hypothesized that they could be used to antagonize pathogenic fish and crustacean bacteria in the environment of the larvae.

The purpose of the present PhD study was to determine if antagonistic Roseobacter clade bacteria occurred in marine aquaculture units. The study would determine their clonal relationship and elucidate the mechanisms by which these potential probiotic bacteria affect the fish pathogens. The efficiency as probionts was studied in cultures of microalgae and live feed organisms in both axenic and none axenic cultures. Chemical, genetic and bioinformatic tools were used to identify and quantify the production of the antibiotic compound and then compare to reference strains from the Roseobacter clade. Long-

ii term exposure of the fish pathogen to the antimicrobial compound produced by Phaeobacter was done in order to determine risk of resistance and possible genetic and phenotypic effects on the pathogen.

The study showed that some of the isolates from the Greek rearing unit could only be identified via 16S rRNA gene sequencing as Phaeobacter sp. following guidelines from a previously published study. Using whole genome sequencing and in silico DNA-DNA hybridization gave an indication of a new Phaeobacter species. Two such strains together with Phaeobacter inhibens DSM 17395 were chosen for further study. The study demonstrated that in axenic live feed cultures, the potentially new species of Phaeobacter were able to antagonize both the fish pathogen Vibrio anguillarum with up to four logarithmic units and the crustacean pathogen Vibrio harveyi with up to two logarithmic units. This corresponded well with results from reference strains and previous studies. To confirm the in vivo mechanism of action of the antibacterial compound tropodithietic acid (TDA) a defective mutant was included in the study. The mutant was significantly less efficient at antagonizing the pathogen indicating that TDA production is the major contribution to the probiotic action.

To further elucidate the probiotic potential of Phaeobacter inhibens, the probiont were added to cultures of Artemia and Duniella tertiolectra with four added bacterial strains representing aquaculture background microbiota. The Phaeobacter inhibens were able to colonize both cultures and still antagonize the pathogen with up to four logarithmic

iii units. Lastly, the efficiency of the probiont was tested in completely non-axenic cultures of either Artemia or Tetraselmis suecica received from an aquaculture unit. The cultures were inoculated with the pathogen Vibrio anguillarum. The pathogen was reduced with up to three logarithmic units compared to the control.

The antibacterial compound tropodithietic acid (TDA), an antiporter that disrupts the proton motive force, is key in the antibacterial activity of several roseobacters. Introducing probiotics at a larger scale requires understanding of any potential side effects of long term exposure of the pathogen to the probionts or any compounds they produce. We here exposed the fish pathogen, Vibrio anguillarum, to TDA for several hundred generations in an adaptive evolution experiment. No tolerance or resistance arose during the 90 days of exposure and whole genome sequencing of TDA-exposed lineages and clones revealed few mutational changes as compared to lineages grown without presence of TDA. Amino acid changing mutations were found in two to six different genes per clone, however, no mutations appeared unique to the TDA- exposed lineages or clones. None of the virulence genes of V. anguillarum were affected and infectivity assays using fish cell lines indicated that the TDA- exposed lineages and clones were less invasive than the wild type. Thus, long term TDA exposure does not appear to result in TDA resistance and the physiology of V. anguillarum appears unaffected, supporting the application of TDA-producing roseobacters as probiotics in aquaculture.

In summary, this study demonstrates that Phaeobacter inhibens and newly isolated Phaeobacter sp. strains can be used as probiotics against fish and crustacean pathogens both in a short term effect to reduce the concentrations, but more importantly also as in a long term effect as no resistance is seen in the pathogen after continuous exposure of the antibiotic compound TDA. All strains of Phaeobacter sp. were able to colonize cultures of Artemia and microalgae whether it be axenic, defined non-axenic or completely non-axenic.

Resumé

Sygedomme forårsaget af fiskepatogene bakterier kan medføre store populations kollaps hos fiskeyngel virksomheder. En af de største udfordringer for akvakultur er forvaltningen af disse bakterielle udbrud. Vacciner kan gives til fisk, men kun til juvenile og voksne fisk, da det kræver at de har et modent immunsystem. Dette betyder, at fiske larver, indtil de er 2-3 uger gamle, er mere udsatte i forhold til bakterielle infektioner. En kortsigtet løsning er antibiotika men dette kan medføre selektion for antibiotikaresistens blandt de sygdomsfremkaldende bakterier, som igen kan overføres til human- patogener. Derfor er der et behov for alternativer, og et kunne være brugen af probiotiske bakterier. Marine bakterier fra Roseobacter claden (Phaeobacter inhibens) har vist stort potentiale som probiotiske bakterier, og vores hypotese er, at de kan anvendes til at antagonisere patogene fisk- og krebsdyrsbakterier i larvernes miljø.

Formålet med nærværende ph.d. studie var at bestemme, om antagonistiske Roseobacter clade bakterier fandtes i marine akvakultur enheder. Studiet vil fastslå deres klonale slægtskab og belyse de mekanismer, hvormed disse potentielle probiotiske bakterier påvirker fiskepatogener. Effektiviteten som probionter blev undersøgt i kulturer af mikroalger og levende foderorganismer i både axeniske og ikke- axeniske kulturer. Kemiske, genetiske og bioinformatiske værktøjer blev brugt til at identificere og kvantificere produktionen af det antibiotiske stof, og derefter sammenligne det med referencestammer fra

Roseobacter claden. Langvarig eksponering af fiskepatogenen til det antimikrobielle stof produceret af Phaeobacter blev udført for at bestemme risikoen for resistens og mulige genetiske og fænotypiske virkninger på patogenen.

Studiet viste, at nogle af isolaterne fra den græske akvakulturvirksomhed kun kunne identificeres via 16S rRNA-gen sekventering som Phaeobacter sp. ifølge retningslinjer fra et tidligere offentliggjort studie. Brugen af hel-genomsekventering og in silico DNA- DNA-hybridisering gav en indikation af en ny Phaeobacter art. To sådanne stammer sammen med Phaeobacter inhibens DSM 17395 blev valgt til yderligere undersøgelse. Studiet viste, at i axeniske kulturer med levende foder, kunne de potentielt nye arter af Phaeobacter antagonisere både fiskpatogenen Vibrio anguillarum med op til fire logaritmiske enheder og fiske- og krebsdyrpatogenen Vibrio harveyi med op til to logaritmiske enheder. Dette stemte overens med resultater fra referencestammer og tidligere undersøgelser. For at bekræfte in vivo virkningsmekanismen af det antibakterielle stof tropodithietic acid (TDA), blev en defekt mutant inkluderet i undersøgelsen. Mutanten var betydeligt mindre effektiv til at antagonisere patogenen, hvilket indikerede at TDA produktionen var det største bidrag til den probiotiske handling.

For yderligere at belyse det probiotiske potentiale af Phaeobacter inhibens, blev probionten tilføjet kulturer af Artemia og Duniella tertiolectra med fire tilføjede bakteriestammer, der repræsenterede

vii mikrobiotaen fra akvakultur. Phaeobacter inhibens kunne kolonisere begge kulturer og stadig reducere patogenen med op til fire logaritmiske enheder. Til sidst blev effektiviteten af probiont afprøvet i fuldstændigt ikke-axeniske kulturer af enten Artemia eller Tetraselmis suecica modtaget fra en akvakultur virksomhed. Kulturerne blev inokuleret med Vibrio anguillarum. Patogenen blev reduceret med op til tre logaritmiske enheder i sammenligning med kontrollen.

Det antibakterielle stof tropodithietic acid (TDA), et antiporter der forstyrrer den protondrivende kraft, er nøglen i den antibakterielle aktivitet af flere roseobacterer. Indførelse af probiotika på en større skala kræver forståelse af de potentielle bivirkninger af langvarig eksponering af patogenen til probionter eller eventuelle stoffer de producerer. Her udsatte vi fiskepatogenen, Vibrio anguillarum, for TDA i flere hundrede generationer i et adaptivt evolutionseksperiment. Ingen tolerance eller resistens opstod under de 90 dages eksponering og sekventering af hele genomet af de TDA-eksponerede slægter og kloner afslørede kun få mutationsændringer sammenlignet med slægter dyrket uden tilstedeværelse af TDA. Aminosyre-skiftende mutationer blev fundet i to til seks forskellige gener pr klon, dog var ingen mutationer unikke for de TDA-eksponerede slægter eller kloner. Ingen af virulensgenerne fra V. anguillarum blev berørt og infektions assays af fiskcellelinjer viste, at TDA-udsatte slægter og kloner var mindre invasive end vildtypen. Således resulter en langvarig TDA-eksponering ikke i TDA resistens og fysiologien hos V. anguillarum synes upåvirket, hvilket

viii understøtter brugen af TDA-producerende roseobactere som probiotika i akvakultur.

Sammenfattende viser dette studie at Phaeobacter inhibens og nyligt isolerede Phaeobacter sp. stammer kan anvendes som probiotika mod fiske- og krebsdyrpatogener, både i en kortvarig virkning til at reducere koncentrationerne, men endnu vigtigere også som i en langvarig virkning da ingen resistens ses i patogenen efter kontinuerlig eksponering af det antibiotiske stof TDA. Alle stammer af Phaeobacter sp. var i stand til at kolonisere kulturer af Artemia og mikroalger, hvad enten det var axenisk, defineret ikke-axenisk eller helt ikke-axenisk.

Table of contents

Preface ...... i

Summary ...... ii

Resumé ...... vi

Table of contents ...... x

List of articles ...... xii

1. Objectives and hypothesis ...... 1

2. Aquaculture ...... 3

2.2 Larviculture ...... 4

2.3 Live feed ...... 5

2.3.1 Micro algae ...... 6

2.3.2 Artemia ...... 7

2.3.3 Copepods ...... 10

2.3.4 Rotifers ...... 12

2.4 Disease in aquaculture ...... 13

2.4.1 Vibrio anguillarum...... 14

2.4.2 Vibrio harveyi ...... 19

2.5 Bacterial disease management ...... 20

2.5.1 Vaccines ...... 21

2.5.2 Prebiotics...... 22

2.5.3 Probiotics ...... 23

2.5.4 Bacteriophages ...... 26

2.5.5 Immunostimulants ...... 28

3. The Roseobacter clade ...... 28

3.1 Phaeobacter sp...... 32

3.1.1 The bioactive compound tropodithietic acid ...... 37

3.1.2 Probiotic potential ...... 44

4. Concluding remarks and future perspectives ...... 50

5. Acknowledgement...... 52

6. References ...... 53

List of articles

1. Grotkjær T, Bentzon-Tilia M, D'Alvise P, Dourala N, Nilsen KF and Gram L (2016) Isolation of TDA-producing Phaeobacter strains from sea bass larval rearing units and their probiotic effect against pathogenic Vibrio spp. in Artemia cultures. Systematic and Applied Microbiology, Volume 39, Issue 3, pp. 180-188. Doi: 10.1016/j.syapm.2016.01.005

2. Grotkjær T, Bentzon-Tilia M, D'Alvise P, Dierckens K, Bossier P and Gram L (2016) Phaeobacter inhibens as probiotic bacteria in non-axenic Artemia and algae cultures. Aquaculture, Volume 462, pp. 64-69. Doi: 10.1016/j.aquaculture.2016.05.001

3. Rasmussen BB, Grotkjær T, D'Alvise P, Yin G, Zhang F, Bunk B, Spröer C, Bentzon-Tilia M and Gram L (2016) Vibrio anguillarum is genetically and phenotypically unaffected by long- term continuous exposure to the antibacterial compound tropodithietic acid. Applied and Environmental Microbiology, Volume 82, Issue May, 99. 4802-4810. Doi: 10.1128/AEM.01047- 16

xii

1. Objectives and hypothesis

Aquaculture is the fastest growing agricultural industry providing healthy food for mankind. In addition, culture of high valued marine fish, crustacean and molluscan species is financially attractive. However, diseases at the larval stages causing large population crashes constitute a major bottleneck and cause economic losses to the industry. Vaccines are not effective at the larval stages and antibiotics are used for disease control, although there are serious concerns about development of bacterial antibiotic resistance and its transfer to human pathogenic bacteria. The World Health Organisation has argued that antibiotic resistance is one of the most serious threats to human health. This makes a strong argument for the need for development of non- antibiotic based disease control strategies, especially at the larval stages.

The main objective of this PhD project is to reduce the need for antibiotics in marine larviculture by using probiotic bacteria from the Roseobacter clade isolated from rearing units. Rearing of marine larvae is difficult as the larvae require live feed (Artemia, rotifers, copepods) that also require live feed (algae), and the live feed also serves as the vector for larval pathogens. We have during this PhD project isolated Vibrio-antagonistic Roseobacter from several aquaculture units. These were used in model challenge trials with live feed acquired from aquaculture against Vibrio fish pathogens (Paper 1 and 2). Almost all treatments using one pathogen-antagonising principle will select for

1 resistant pathogens and therefore prolonged exposure to the antibacterial compound from the probiont was studied (Paper 3). The virulence of pathogenic bacteria can be modulated by antibacterial compounds and any effects of the roseobacter antibacterial compound was elucidated by prolonged exposure of the compound (Paper 3). We hypothesise that by using rearing isolated probiotic bacteria the environmental impact will be minimized and the risk of new resistance or increase in virulence will be significantly reduced.

2. Aquaculture

Aquaculture is the farming and production of aquatic organisms such as fish, crustaceans, mollusks and aquatic plants and it is the fastest growing food-producing sector, and is projected to rise to 62 percent of providing fish for human food by 2030 (Benavete and Gatesoupe, 1988; Gatesoupe, 1991; Nicolas et al., 1989; FAO 2014). Farming implies some form of intervention in the rearing process to enhance production, such as regular stocking, feeding and protection from predators. In 1995 the world aquacultural production reached more than 20 million tonnes and was at 60 million tonnes in 2012 (FAO 2014) (Figure 1).

Figure 1. World capture fisheries and aquaculture production (FAO 2014)

The farming of fish is the most common form of aquaculture. It involves raising fish commercially in tanks, ponds, or ocean enclosures, usually for food (Edwards, 2015). A facility that releases juvenile fish into the

3 wild for recreational fishing or to supplement a species' natural numbers is generally referred to as a fish hatchery (Edwards, 2015).

2.2 Larviculture

Rearing fish in captivity provides access to embryos, which is a limiting factor in evolutionary developmental biology research. The ability to rear larvae of various species in captivity is essential for comparative and manipulative investigations of the genetic and developmental basis of diversity and novelty. However, larval rearing is usually carried out under intensive conditions and the larvae fed on cultured live food mainly rotifers and Artemia because formulated dry feed for larvae is not yet available (Olivotto et al., 2011).

Fish larvae can be reared under stagnant or open-system conditions. Generally, partial water changes are provided and microalgae are supplied to the rearing tanks during the initial stages of culture. Low exchange rates of water may affect the retention time of prey in the larval tanks and changes may occur in the biochemical composition of the prey before being consumed by the larvae (Reitan et al., 1993).

Recently, more attention has been paid to the importance of microbiology in larval rearing. It has been reported that the microflora associated with the food chain are detrimental to the larvae. Both microalgae and live feed have a high bacterial load, which in the later is dominated by Vibrionaceae and Pseudomonas (Austin and Allen, 1982; Nevejan et al., 2016; Olafsen, 2001; Taniguchi et al., 2016). The high

4 load of opportunistic pathogenic bacteria can cause complete system crashes, where whole stocks of larvae are killed (Figure 2) (Reid et al., 2009)

Figure 2. Mortality of cod (Gadus morhua) larvae in a Scottish hatchery from 46 to 95 days post hatch. All tanks initially contained 50,000 larvae, and total survival was 0 - 16 %. The mortalities in all tanks were caused by infections with Vibrio spp., and the mass mortality in tank N8 was due to Vibrio anguillarum (Reid et al., 2009).

2.3 Live feed

Aquatic invertebrates are natural food sources for fish larvae and are also co-inhabitants of larval ecosystems. This relationship will imply that the establishment of a larval microflora will also be influenced by the indigenous microflora of invertebrates, whether they are food organisms or co-inhabitants of larval ecosystems or rearing facilities. Aquatic invertebrates live in areas with an abundant microflora that may provide food particles for filter feeders, but also act as a potential 5 source of infection. They are, thus, continually confronted with infectious microorganisms (Haché et al., 2016b; Katan et al., 2016).

2.3.1 Micro algae

Figure 3. Microalgae tanks which supply a steady source of nutrients in aquaculture (http://www.worldfishing.net/news101/industry-news/us-releases-draft-aquaculture- policies).

Microalgae play a vital role in aquatic food chain and are widely used in rearing of aquatic animals such as mollusks, shrimps and fishes at different growth stages (Chakraborty et al., 2007). They are required for larvae nutrition during a brief period of life cycle and are used either for direct consumption or indirectly as prepared feed (Hemaiswarya et al., 2011; Roy and Pal, 2015). In most instances, the whole algae are used as feed or feed supplement to the live feed (Artemia, Rotifer and Copepod)

6 required for production of live feed (Figure 3). Microalgae affect the microbiology (Nicolas et al., 1989), nutrition (Brown et al., 1997; Howell, 1979; Roy and Pal, 2015; Scott and Middleton, 1979), feeding (Khatoon et al., 2016; Reitan et al., 1993) and behavior (Naas et al., 1992) of the larvae. The addition of microalgae to the tanks during early rearing of the larvae may affect rearing performance. For turbot, for instance, green water improves the appetite, initial growth rate, survival and viability of the larvae fry (Reitan et al., 1993; Øie et al., 1997).

2.3.2 Artemia

Figure 4. A seabass larvae feeding on Artemia nauplii (http://www.aquaculture.ugent.be/ About/history.htm).

Artemia is an important starter feed used in larviculture and is widely used as live feed in aquaculture (Figure 4). The Artemia nauplii satisfies the nutritional requirements for the early stage fish and crustacean larvae (Toi et al., 2013), although some enrichment is needed to complete the requirements of the larvae (Lanes et al., 2012).The hatching of Artemia cysts comes with an increase in the bacterial load (Haché et al., 2016a) and this is controlled by either disinfecting the cysts with hypochlorite or removing the shell of the Artemia cysts altogether through the process of decapsulation (Van Stappen, 1996). Another measure that is routinely taken to reduce the influx of bacteria into the larval rearing tanks is a thorough washing of the nauplii with clean (disinfected) water upon harvesting. Although these hygiene measures have been shown to reduce bacterial load to some extent, they are far from absolute. Bacteria numbers are known to increase dramatically during the Artemia hatching process. A subsequent enrichment step with lipid- and/or protein-rich products can further increase bacterial numbers. Moreover, bacteria are also taken up into the gut of the nauplii as demonstrated by Høj et al. (2009) (Høj et al., 2009). The study showed that a washing step was effective to remove the bacteria that are loosely attached to the external surface of the nauplii; however, most bacteria were localized in the nauplii gut and these were not removed. Also at the research level, several attempts have been made to disinfect Artemia nauplii (Gatesoupe, 2002). These include treatment of either cysts or the hatched nauplii with biocides (e.g. formaldehyde), UV, ozone or peroxide-based products. Apart from

8 the fact that these treatments might cause considerable mortality to the nauplii or reduce their vigor, they might also pose potential risk to the predator larvae they are fed to, because of residues or toxic by-products produced due to these treatments. In this PhD study it was shown that the probiotic bacteria Phaeobacter inhibens not only was able to colonize cultures of Artemia but also significantly improve the survival of the brine shrimp after Vibrio anguillarum and Vibrio harveyi infection (Grotkjær et al., 2016b) (Table 1).

Table 1. Survival of Artemia after four days [%]. Artemia were exposed to V. anguillarum or V. harveyi and co-inoculated with either of four Phaeobacter strains: DSM 17395, its TDA- negative mutant (MJG-G6) and two strains isolated from sea bass larval units: S26 and S60. Numbers are average of two independent trials.

Artemia survival (%)

Treatment V. anguillarum V. harveyi

Axenic control 54 ± 0

Pathogen only 8 ± 3.2 4 ± 1.1

Pathogen + P. inhibens DSM17395 89 ± 1.01 74 ± 13.6

Pathogen + P. inhibens MJG-G6 45 ± 12.2 12 ± 3.4

Pathogen + Phaeobacter S26 77 ± 11.4 74 ± 10.2

Pathogen + Phaeobacter S60 55 ± 3.8 73 ± 5.1

2.3.3 Copepods

Figure 5. Copepods for live feed in aquaculture (https://global.Britannica.com/animal/ copepod).

In nature, copepods constitute the major part of the diet of early life stages of many marine fish species (Figure 5). Copepods have been demonstrated to be a nutritionally adequate live prey for rearing larvae of many marine fish species (Busch et al., 2011; McEvoy et al., 1998; Payne and Rippingale, 2001; Shields et al., 1999; Støttrup, 2003). However, larval rearing in intensive commercial hatcheries relies exclusively, or almost exclusively, on traditional live prey; enriched rotifers and enriched nauplii of Artemia sp. Despite significant progress in the development of copepod culture techniques (Drillet et al., 2011;

Støttrup, 2003; Støttrup et al., 1986), the challenge still remains to develop an efficient, cost-effective large-scale production method that could match or compete with the established methods for culturing traditional live prey.

Copepod cultures in outdoor ponds or enclosed lagoons provide the basis for cost-effective, low technology extensive rearing of marine species such as turbot, flounder and cod. These systems rely on bloom culture of phytoplankton and zooplankton to which the newly hatched larvae are added (Rasdi and Qin, 2016; Souissi et al., 2014).

The use of freshly harvested copepods directly for fish larval rearing presents a risk of introducing parasites (Støttrup, 2003) and harmful bacteria to the larvae such as different Vibrio species (Gugliandolo et al., 2008). Using only the first generation nauplii of freshly harvested copepods can minimize risk of parasitic infections in cases where the copepod act as intermediary host between compulsory hosts. In general, bacterial numbers associated with copepods seem to be lower in copepods than in Artemia nauplii (Verner-Jeffreys et al., 2004). As copepods and nauplii seem to be well able to tolerate filtration, the copepod culture medium can be washed out thoroughly before concentrating and feeding the live prey to the fish, thus minimizing bacterial transfer. In some systems this is part of the harvest process, to minimize contaminants in the culture systems (Støttrup, 2003).

2.3.4 Rotifers

Figure 6. Rotifer plicatilis for live feed (https://www.primo.net.au/shop/Live-Food-Culture- and-Enrichment/rotifer-culture-products).

Rotifers, mainly belonging to the genus Brachionus (Brachionus plicatilis) have been essential live prey in larval rearing of marine fish species since the 1970s (Lubzens et al., 2001) (Figure 6). Two main species of rotifer have been used in European hatcheries: Brachionus plicatilis large size and Brachionus rotundiformis small size. It has been reported that the use of small sized rotifers significantly improves the initial feeding performance of different species of fish (Thépot et al., 2016) at the earlier developmental stages.

Rotifers, commonly used as the first prey in larvae artificial food chain, are major carriers of bacteria (Hamre, 2016). Most of bacteria in rotifers are not pathogenic but detrimental effects on fish larvae can be caused by the accumulation of bacteria in prey. Bacteria associated with rotifer cultures have been related to unexpected mortalities or to suppressed growth in rotifers as well as to low survival and growth in fish larvae (Benavete and Gatesoupe, 1988; Nicolas et al., 1989). The genus Vibrio has been found to be dominant in rotifers (Verdonck et al., 1997).

2.4 Disease in aquaculture

Intensive aquaculture implies that food is added at high concentration to water, which results in an excellent medium for growth of heterotrophic or opportunistic bacteria, and this will affect selection and growth of microorganisms in the nets or cages. Many fish diseases are caused by opportunistic pathogens. Thus, feed composition affects the commensal microflora and results in the proliferation of bacteria that normally have restricted opportunities to grow (Benavete and Gatesoupe, 1988; Kashulin et al., 2016; Uzun and Ogut, 2015).

The development and severity of a disease following exposure to a pathogen involves a complex web of variables such as: the virulence (ability to cause disease) of the pathogen; the immune, genetic and physiological condition of the host; stress; and population density. Different strains of the same pathogen may vary considerably in their

13 ability to cause disease (Frans et al., 2013; Li et al., 2015; Pedersen et al., 1997).

During the intensive hatching of eggs and rearing of marine larvae, various forms of interactions between bacteria and the biological surfaces may occur. This may result in the formation of an indigenous microflora or be the first step of infection (Bergh et al., 1992). In the aquatic environment, bacteria travel easily between habitats and hosts, and a better understanding of host–microbe interactions and natural defenses is imperative for the successful mass production of larvae. The microflora of marine invertebrates and plankton may be dominated by opportunistic or potentially pathogenic vibrios at certain times of the year, and marine invertebrates may harbor bacteria that are pathogenic to other organisms.

2.4.1 Vibrio anguillarum

Vibrio anguillarum is a Gram-negative bacterium belonging to the family Vibrionaceae with short rods, curved or straight, round ended, occurring singly or in pairs and is pleomorphic. The size of the bacterium is typically 0.5 to 0.8 µm in diameter and 1.4 to 2.6 µm in length with polar flagella. It is a halophilic and facultative anaerobe, capable of both fermentative and respiratory metabolism. It has rapid motility and is β- haemolysin positive. After 2 days of culture on agar, the colonies are 2mm and a glistening cream-color. The bacterium is catalase and oxidase positive. Optimal growth occurs at pH 7, salinity of 2% and at a temperature of 25°C (Larsen, 1984). 14

Vibrio anguillarum was reported and described for the first time in the start of the 20th century (Frans et al., 2011). The first atypical Vibrio anguillarum was described in 1969 (Pacha and Kiehn, 1969), and in the following years additional isolates were described as being closely related to the original strains but not identical (Harrell et al., 1976), which prompted the discrimination of Vibrio anguillarum into biotype 1 (original strains) and biotype 2 (related). In 1981 the advances in DNA methods highlighted more clearly the differences between biotype 1 and 2, especially regarding plasmid size and its cleavage, which prompted Schiewe and Crosa (1981) (Schiewe and Crosa, 1981) to reclassify Vibrio anguillarum biotype 2 under a new species, Vibrio ordalii. Nowadays, the distinction between these two species is well established (Grisez et al., 1991; Schiewe and Crosa, 1981). V. anguillarum is composed of 2 chromosomes (3.0 and 1.2 Mbp) with a whole genome size of about 4.2 Mbp and a G+C content of 43-46%. Most of the O1 serotype strains isolated harbor a plasmid which carries many virulence factors, such as genes affecting chemotaxis and motility, an iron uptake system, lipopolysaccharides (LPSs) and extracellular products with proteolytic or haemolytic activity (Frans et al., 2011)

The genome of the O1 serotype strain Vibrio anguillarum 775 was sequenced, where the annotation shows that the majority of the genes for essential cell functions and pathogenicity are located in chromosome 1. This chromosome also has 8 genomic islands, while chromosome 2 has only 2. It was found that some of these islands carried potential virulence genes (Naka et al., 2011). Vibrio anguillarum strains show 15 close similarities according to 16S rRNA sequences irrespective of time or place of isolation, suggesting 16S rRNA is highly uniform and stable (Tan et al., 2014), indicating further genetic fingerprinting tools such as enterobacterial repetitive element intergenic consensus (ERIC) PCR (Rodríguez et al., 2006), multi-locus sequence typing (MLST) (Enright et al., 2000), and pulsed-field gel electrophoresis (PFGE) (Skov et al., 1995) are needed for surveillance purposes, monitoring outbreaks and tracking pathogen spreading. Currently, 23 different O serotypes (O1- O23) within Vibrio anguillarum are discriminated, each single O serotype with its own different pathogenicity and host specificity. Among these, only serotypes O1, O2, and O3, have been reported to associate with vibriosis infection in fish (Toranzo and Barja, 1990). Within serotype O2, three host specific sub-serotypes, O2a, O2b and O2c, have been distinguished by Rasmussen (Rasmussen, 1987).

There are over 122 species of bacteria in the genus Vibrio (http://www.bacterio.cict.fr/, 2016), of which a dozen are reported pathogenic to fish and shellfish (Toranzo and Barja, 1990) such as V. anguillarum and V. ordalii (Schiewe et al., 1981), V. tubiashii (Hada et al., 1984), V. harveyi (Austin and Zhang, 2006), V. carchariae (Grimes et al., 1984), V. salmonicida (Egidius et al., 1986) and V. cholerae non-O1 (Rehulka et al., 2016).

V. anguillarum, as one of the most prominent fish pathogens, is associated with vibriosis infection in aquaculture (Frans et al., 2011; Larsen et al., 1994, 1988). Vibriosis, a highly fatal haemorrhagic

16 septicaemia, is one of the most prevalent fish diseases and has been found in more than 50 fresh and salt-water fish species including various species of economic importance to the larviculture and aquaculture industry, especially during the larval and juvenile stages. V. anguillarum has also been reported as a crustacean pathogen in live feed cultures of Artemia (Grotkjær et al., 2016b)

Most of the important economical species cultured in aquaculture are susceptible to Vibrio species, such as Atlantic salmon (Salmo salar), rainbow trout (Oncorhynchus mykiss), sea bream (Sparus aurata), sea bass (Dicentrarchus labrax), ayu (Plecoglossus altivelis), Atlantic cod (Gadus morhua), saithe (Pollachius virens) and turbot (Scophthalmus maximus) (Larsen et al., 1994).

Clinical signs of vibriosis were first described in eels in Norway in the 19th century (Frans et al., 2011) and later in the UK by (McCarthy, 1976). The disease manifested itself by the appearance of areas of macroscopic haemorrhaging, with swelling and boils. Destroyed epithelium is evident at the top of and around the lesion, while haemorrhaging can be present in the skin (Figure 7). The liver and intestine may also have haemorrhagic tissue and the heart lumen can be filled with fluid.

Figure 7. Clinical vibriosis in farmed Atlantic cod.

The triggering of an outbreak of vibriosis is the result of host-pathogen- environment interactions (Larsen et al., 1994), with the main predisposition being the stress level of the fish. Rise in temperatures has a known effect on the stress level of fish and corresponds well with disease outbreaks (Egidius and Andersen, 1977), especially in early to midsummer (Haastein and Holt, 1972), which corresponds also better with the optimal growth temperature of V. anguillarum.

2.4.2 Vibrio harveyi

Vibrio harveyi is curved rod shaped, Gram-negative, luminous and monotrichous belonging to the family Vibrionaceae. V. harveyi is a natural inhabitant of seawaterand is a facultative anaerobe. V. harveyi has the ability of bioluminescence, which is controlled by the regulatory gene luxR as well as population density-sensing (quorum sensing) autoinducer signaling molecules (Hong et al., 2016).

V. harveyi is known to regulate gene expression based on its population density by Quorum-Sensing. The interspecies communication of V. harveyi have major effects on the way the bacteria interact with the environment, whether it is luminescing, becoming virulent, or excreting nucleases for defense (Defoirdt et al., 2008). The persistence and survival of V. harveyi in shrimp hatcheries has been attributed to its ability to form biofilms with resistance to disinfectants and antibiotics (Gupta et al., 2016).

V. harveyi is a pathogen of fish and invertebrates, including sharks (Austin and Zhang, 2006), different fish species (Austin and Zhang, 2006; Pujalte et al., 2003; Zhang and Austin, 2000), seahorses (Alcaide et al., 2001), lobster (Diggles et al., 2000), sea cucumber (Becker et al., 2004) and prawn (Gupta et al., 2016). Its pathogenicity depends on the concentration of V. harveyi cells at a given time (Defoirdt et al., 2008). Diseases caused by V. harveyi include eye-lesions, gastro-enteritis, vasculitis and luminous vibriosis (Lee et al., 2002; Soffientino et al., 1999). Luminous vibriosis is a leading cause of death among 19 commercially farmed shrimp and other aquaculture with up to 75% mortalities occurred when the water temperatures were 20° and 23° C (Diggles et al., 2000). The infection, by V. harveyi, enters through the mouth and forms plaques, then spreads to the innards and the appendages. Loss of limb function and appendage degradation has been documented. Contamination can spread all the way to egg and larval tanks, thus causing an even bigger problem for shrimp farmers. V. harveyi have differences between isolates in terms of pathogenicity and proteases, phospholipase, haemolysins and/or other exotoxins may well exert significant roles in the pathogenicity of V. harveyi (Harris and Owens, 1999). Furthermore, the capability of the pathogen to attach to chitin by means of a specific protein-mediated mechanism may be of significance for the adhesion, colonization and subsequent infection of the host (Montgomery and Kirchman, 1993).

2.5 Bacterial disease management

In the aquacultural setting the different species are cultured in high densities and it is therefore difficult to determine individual doses of antimicrobial drugs, and this can result in an overuse of the drugs (Heuer et al., 2009). The indiscriminate use of antimicrobials may provide changes in the microbial community structure in the aquaculture systems and surrounding environment, resulting in the selection and persistency of antimicrobial resistant bacteria (Resende et al., 2012; Zhang et al., 2009) 20

Bacteria resistant to antibiotics have been found in the aquatic environment and higher level of antibiotic resistance was reported in bacteria associated with aquaculture farms than nearby coastal regions (Cabello, 2006). During the last decades, antibiotics have been used as a traditional strategy for fish diseases management and also for the improvement of growth and efficiency of feed conversion. However, the development and spread of antimicrobial resistant pathogens were well documented (Cabello et al., 2016). There is a risk associated with the transmission of resistant bacteria from aquaculture environments to humans, and risk associated with the introduction in the human environment of nonpathogenic bacteria, containing antimicrobial resistance genes, and the subsequent transfer of such genes to human pathogens (Cabello et al., 2016).

The immune capacity of fish larvae is not fully developed until they are several weeks old (Chantanachookhin et al., 1991), however this may vary among fish species. Until then, larvae probably rely on nonspecific defense mechanisms and phagocytosis as production of specific antibodies has not been observed in fish at this stage. Antibody response to bacteria has been observed 3 weeks post hatching (Dalmo et al., 1997).

2.5.1 Vaccines

Depending on the species, fish from 2–3 weeks may be vaccinated by injection, some protection may be obtained by immersion, while oral administration usually yields less protection (Sommerset et al., 2005). At 21 present, vaccines are available against many of the serious bacterial diseases causing problems for the fish farming industries (Brudeseth et al., 2013), but still only be administered after 2-3 weeks.

2.5.2 Prebiotics

Prebiotics are defined as non-digestible components that are metabolized by specific health-promoting bacteria such as Lactobacillus and Bifidobacterium (Roberfroid, 2005). It has been well established that the byproducts produced when beneficial commensal bacteria ferment prebiotics play a major role in improving host health (Chu and McLean, 2016; Ringø et al., 2010; Song et al., 2014). Prebiotics are carbohydrates, which can be classified according to their molecular size or degree of polymerization, into monosaccharides, oligosaccharides or polysaccharides (Mussatto and Mancilha, 2007). In this respect, prebiotics are used as energy sources for the gut bacteria (Roberfroid, 2005). The immunomodulatory activity of prebiotics is mediated through direct interactions with their receptors such as β-glucan receptors and dectin-1 receptors that are expressed on macrophages (Brown et al., 2002). Saccharides may also interact with PRRs in the form of microbe associated molecular patterns (MAMPs) such as the capsular polysaccharide of bacteria, triggering an immune response (Bron et al., 2011). In short the prebiotics works in two ways of stimulating the innate immune system: 1; by directly stimulation of the innate immune system or 2; by increasing the growth of commensal microbiota.

2.5.3 Probiotics

According to the currently adopted definition by Food and Agricultural Organization/World Health Organization, probiotics are live microorganisms which when administered in adequate amounts confer a health benefit on the host (FAO 2001). In aquaculture practices, probiotics are used for a quite long time but in last few years probiotics became an integral part of the culture practices for improving growth and disease resistance. This strategy offers innumerable advantages to overcome the limitations and side effects of antibiotics and other drugs and also leads to high production through enhanced growth and disease prevention (Das et al., 2008; Verschuere et al., 2000). In aquaculture, the range of probiotics evaluated for use is considerably wider than in terrestrial agriculture. Apart from the nutritional and other health benefits (Austin et al., 1995; Balcázar et al., 2006; Gram et al., 1999), certain probiotics as water additives can also play a significant role in decomposition of organic matter, reduction of nitrogen and phosphorus level as well as control of ammonia, nitrite, and hydrogen sulfide (Boyd and Massaut, 1999).

Numerous microbes have been identified as probiotics for aquaculture practices, many of which differ markedly in their mode of action. There are, however, some common mechanisms of action that have been reported for the majority of probiotic strains. Probiotics help in feed conversion efficiency and live weight gain (Al-Dohail et al., 2009; Sáenz De Rodrigánez et al., 2009; Syngai et al., 2016) and confer protection

23 against pathogens by competitive exclusion for adhesion sites (Chabrillón et al., 2005; Kesarcodi-Watson et al., 2008; Vine et al., 2004), production of organic acids, hydrogen peroxide and several other compounds such as antibiotics, bacteriocins, siderophores, lysozyme (Pandiyan et al., 2013; Yan et al., 2002) and also modulate physiological and immunological responses in fish (Balcázar et al., 2006; Nayak, 2010).

As it has been previously mentioned and well documented that live feed cultures act as vectors for bacteria that are harmful to the fish larvae during start-feeding, and various attempts at microbial control of the start-feeding cultures have been reported. An approach to this would be microbial control of the microflora of live feed organisms. Microbial control of Artemia juveniles has been achieved by preemptive colonization by selected probiotic bacterial strains (Grotkjær et al., 2016a, 2016b; Verschuere et al., 1999) and for rotifers (D’Alvise et al., 2012), and the resulting changes in the microflora appeared stable regarding colonization of the probiont in cultures of Artemia and microalgae (Figure 8 and 9 from Paper 2).

Figure 8. Counts of P. inhibens DSM 17395 GMr in axenic Artemia cultures. P. inhibens DSM 17395 GMr inoculated at 106 CFU/ ml alone (○), or in the presence of V. anguillarum (□), V. anguillarum and background strains separately (▲, ▼, ●, ■) or V. anguillarum and all background strains (Δ). Points are average of two biological replicates and error bars are standard deviation of the mean.

Figure 9. Counts of P. inhibens DSM 17395 GMr in antibiotic treated microalgae D. tertiolectra. P. inhibens DSM 17395 GMr inoculated at 106 CFU/ ml alone (○) or in the presence of V. anguillarum (□), V. anguillarum and background strains separately (▲, ▼, ●, ■) or V. anguillarum and all background strains (Δ). Counts of P. inhibens DSM 17395 GMr in D. tertiolectra with natural microbiota in presence of V. anguillarum and all of the background strains (). Points are average of two biological replicates and error bars are standard deviation of the mean.

2.5.4 Bacteriophages

Bacteriophages are bacterial viruses and therefore natural enemies of bacteria and they are extremely abundant in natural systems (Madhusudana Rao and Lalitha, 2015). The use of bacteriophages to control bacterial infections in aquatic food production system has the promising potential to address the twin problem of controlling bacterial infections and at the same time avoiding an environmental impact (Nakai and Park, 2002). 26

When the bacteriophages infect bacteria it adopts any of two lifestyle options. The most apparent lifestyle is the lytic cycle, a form of infection resulting in the destruction of the bacterial host. The lytic cycle comprises a series of events that occur between attachment of phage particle to a bacterial cell and its subsequent release of daughter phage particles. It consists of four phases: adsorption of phage to host cell, penetration of phage nucleic acid, intracellular development and final release of daughter phage particles (Madhusudana Rao and Lalitha, 2015). The lysogenic cycle, on the other hand comprises replication of phage nucleic acid together with the host genes for several generations without major metabolic consequences for the cell. This is a latent mode of infection and it occurs at a very low frequency. The phage genes in this state may occasionally revert to lytic cycle, leading to release of phages particles, this property is known as lysogeny (Madhusudana Rao and Lalitha, 2015).

Lytic phages have been tried for the control of bacterial infections in agriculture (Jones et al., 2012) and in meat and cheese products (Carlton et al., 2005). The establishment of the causative agent of the disease is the major step in bacteriophage therapy. Then, it is necessary to select the bacteriophages that can effectively infect the target bacteria and to evaluate the potential of those bacteriophages to control bacterial diseases in aquaculture (Madhusudana Rao and Lalitha, 2015).

2.5.5 Immunostimulants

The definition of an immunostimulant is a naturally occurring compound that modulates the immune system by increasing the host’s resistance against diseases that in most circumstances are caused by pathogens (Sakai, 1999). The immunostimulants can be divided into several groups depending on their sources: bacterial, algae-derived, animal-derived, nutritional factors as immunostimulants, and hormones/cytokines (Song et al., 2014). The perceived benefits of using immunostimulants in larval culture are numerous, but mainly theoretical (Song et al., 2014). Immunostimulants should up-regulate the innate defenses of a larval fish putting it in a more prepared state to meet and overcome an invading pathogen (Bricknell and Dalmo, 2005).

3. The Roseobacter clade

There is a great bacterial diversity present in the world’s oceans, whereas the majority of these fall into nine major clades (Rappé and Giovannoni, 2003). One of these nine groups is the Roseobacter clade, proven by 16S gene rRNA sequencing and is widespread distributed from coastal to open seas and from sea ice to sea floor (González and Moran, 1997; Gram et al., 2010; Selje et al., 2004), the subgroup Ruegeria mobilis are even being termed cosmopolitan because of their widespread distribution (Sonnenshein et al., 2016). The members of the clade have also been found in commensal relationships with marine phytoplankton, invertebrates and vertebrates. Culturable bacteria of the Roseobacter clade was first isolated by (Shiba et al., 1979), but were not 28 classified as the newly established Roseobacter clade until 1991 (Shiba, 1991). Members of the clade share more than 89% identity of the 16S rRNA gene (Buchan et al., 2005) and based on a survey of selected genes, they fall into five different clusters (Figure 10) (Luo and Moran, 2014). Members of the Roseobacter clade are Gram-negative rods and some produce pigments resulting in colored colonies on agar (Ruiz- Ponte et al., 1998; Shiba, 1991). With a few exceptions all Roseobacter clade members are strictly marine or hypersaline (Moran et al., 2007).

Figure 10. Survey of selected genes in 52 Roseobacter isolates genomes. % complete, estimate of genome completeness. The phylogenetic tree was constructed based on a concatenation of 50 single-copy conserved protein sequences using the RAxML software (Luo and Moran, 2014). ∼

Members of the clade have different pathways for energy metabolism. Some strains have the ability to obtain sulfur from dimethylsulphoniopropionate (DMSP), which originates from algae, either by the cleavage pathway where the end product is the gas dimethyl sulfide that may influence the global climate, or by the demethylation/demethiolation pathway where the sulfur is incorporated into biomass (Moran et al., 2003). Some of the Roseobacter species are able to form rosettes (Figure 11) which seems to be correlated with biofilm formation and thereby it may acts as an selective advantage when colonizing marine surfaces (Bruhn et al., 2007, 2005).

Figure 11. Scanning electron micrographs of Roseobacter strain 27-4 grown in MB under stagnant conditions (Bruhn et al., 2005)

3.1 Phaeobacter sp.

The genera Phaeobacter and Ruegeria of the Roseobacter clade consists of aggressive colonizers such as Phaeobacter inhibens, Phaeobacter gallaeciensis and Ruegeria mobilis (Dang and Lovell, 2000; Rao et al., 2006). This makes the bacteria from these genera an area of interest because of the remarkable ability to form and invade biofilms, their antifouling capacity, their antibacterial effect, and their potential application as probiotics in aquaculture (Brinkhoff et al., 2004; D’Alvise et al., 2014, 2012; Grotkjær et al., 2016a, 2016b; Hjelm et al., 2004a; Rao et al., 2006, 2005). In this PhD study different strains of Phaeobacter and Ruegeria was used (Table 2), but mainly Phaeobacter inhibens strain DSM 17395 and Phaeobacter sp. S26 and S60 was in focus as the probiont in paper 1 and 2 (Grotkjær et al., 2016a, 2016b).

Table 2. Potential probionts used in the PhD study.

Strain Paper Phaeobacter inhibens DSM 16374T 1 Phaeobacter inhibens DSM 17395 1 Phaeobacter gallaeciensis DSM 26640T 1 Phaeobacter sp. 27-4 1 Ruegeria mobilis F1926 1 Pseudovibrio sp. FO-BEG1 1 Phaeobacter sp. S26 1 Phaeobacter sp. S60 1 Phaeobacter inhibens DSM17395 GMr 1 and 2 Phaeobacter inhibens MJG-G6 1

In paper 1 (Grotkjær et al., 2016b) 3,100 bacterial colonies from a Greek rearing unit were analyzed in order to find potential probionts. Of these 3,100 colonies approximately 3% showed antagonistic activity against 32 the pathogen Vibrio anguillarum. Because of extensive in-house knowledge and previous work (D’Alvise et al., 2012, 2010; Porsby and Gram, 2016) the main focus was to find probionts relating to the Roseobacter clade. To identify the potential Roseobacter clade probionts 16S rRNA gene sequencing was used which revealed the potential probionts as being 99.7% identical to Phaeobacter inhibens and Phaeobacter gallaeciensis (Paper 1, Grotkjær et al., 2016b). A more in-depth molecular characterization was needed in order to determine the species. Therefore two representative strains were whole genome sequenced by Illumina MiSeq. Following the work of Buddruhs (2013) (Buddruhs et al., 2013) the V4 and V5 regions of the two strains were compared to the same regions of Phaeobacter inhibens and Phaeobacter gallaeciensis. Both strains had V4 regions corresponding to Phaeobacter gallaeciensis and V5 regions corresponding to Phaeobacter inhibens (Figure 12), giving no clear identification (Grotkjær et al., 2016b).

Figure 12. Alignment of the V4 (A) and V5 (B) regions of P. gallaeciensis DSM 26640T, Phaeobacter strains S60 and S26, and P. inhibens DSM 17395 and DSM 16374T (Paper 1) (Grotkjær et al., 2016b).

The average nucleotide identity (ANi) comparison suggested that Phaeobacter sp. S26 and Phaeobacter sp. S60 represented a single species with ANIb values of 98% identity. Both strains exhibited 91% identity with P. gallaeciensis DSM 26640T, 90% identity with the two P. inhibens reference strains (DSM 16374T and DSM 17395) (Figure 13) (Grotkjær et al., 2016b).

Figure 13. A) BLAST+ based average nucleotide identity analysis (ANIb) of the strains Phaeobacter sp. S26 and S60, and Phaeobacter inhibens DSM 16374T, Phaeobacter inhibens DSM 17395, Phaeobacter gallaeciensis DSM 26640T and Pseudophaeobacter articus DSM 23566. The scale bare represents percentages as quiniles. B) A pan-genomic dendrogram showing the phylogeny of the strains based on shared protein families (Grotkjær et al., 2016b). 35

The genomes of strains S26 and S60 were compared to P. inhibens (DSM 17395 and DSM 16374T), P. gallaeciensis (DSM 26640T) and P. arcticus (DSM 23566T) by in silico DNA-DNA hybridization using the web-page software tool using the Genome-to-Genome Distance Calculator (GGDC 2.0 http://ggdc.dsmz.de/) (Auch et al., 2010). Using the suggested formula 2, strains S26 and S60 had 85% identity to each other, but only 40% to P. inhibens DSM 17395 and P. inhibens DSM 16374T, and 44% to P. gallaeciensis DSM 26640T. Pseudophaeobacter arcticus DSM 23566T was included in the analysis as an outgroup (20% identity confirming it) (Table 3) (Grotkjær et al., 2016b). The extensive molecular characterization could not identify the two strains as either P. gallaeciensis or P. inhibens, but rather supported the proposal of a new Phaeobacter species.

Table 3. In silico DNA-DNA hybridization of Phaeobacter strains S26 and S60 with Phaeobacter inhibens DSM 17395, Phaeobacter inhibens DSM 16374T, Phaeobacter gallaeciensis DSM 26640T and Pseudophaeobacter arcticus DSM 23566T based on the DSMZ tool Genome-to- Genome Distance Calculator (GGDC 2.0, http://ggdc.dsmz.de/) (Grotkjær et al., 2016b).

DNA-DNA hybridization similarity (%) Formula Strain S26 S60 DSM DSM DSM DSM 23566 16374 17395 26640 2 S26 ------S60 85 - - - - - DSM 40 40 - - - - 16374 DSM 40 40 79 - - - 17395 DSM 44 44 38 38 - - 26640 DSM 20 20 20 20 20 - 23566

3.1.1 The bioactive compound tropodithietic acid

Phaeobacter sp. S60, P. inhibens and P. gallaeciensis, as well as a few other marine bacteria (i.e. Ruegeria mobilis and Pseudovibrio spp.), produce the potent antibacterial compound tropodithietic acid (TDA) (Figure 14)(Table 4) (Grotkjær et al., 2016b; Thiel et al., 2010). TDA inhibits both fish and human pathogenic bacteria, as well as many nonpathogenic bacteria (Brinkhoff et al., 2004; D’Alvise et al., 2012, 2010; Grotkjær et al., 2016a, 2016b; Hjelm et al., 2004b; Porsby et al., 2008) and potentially also has a role as an anticancer drug (Wilson et al., 2015).

Figure 14. The molecular structure of TDA (Thiel et al., 2010).

Table 4. TDA concentration after 72 hours stagnant at 25 °C. 13 Phaeobacter sp. strains isolated from sea bass larvae units. Reference strains Phaeobacter inhibens DSM 16374T, Phaeobacter inhibens DSM 17395, Phaeobacter gallaeciensis DSM 26640T, Phaeobacter sp. 27- 4, Ruegeria mobilis F1926, Pseudovibrio sp. FO-BEG1 and Pseudophaeobacter arcticus DSM 23566T are added for TDA concentration comparability. Numbers are averages of two independent replicates (Grotkjær et al., 2016b).

Strain TDA concentration (average µM ± s.d.)

20 0.47 ± 0.03 21 0.53 ± 0.01

22 0.63 ± 0.04 24 0.71 ± 0.07 26 0.92 ± 0.07 27 0.77 ± 0.08 28 0.86 ± 0.07 29 0.57 ± 0.02 30 0.49 ± 0.03 51 1.32 ± 0.03 53 0.31 ± 0.01 59 0.31 ± 0.03 60 0.95 ± 0.08 DSM 16374 0.74 ± 0.11 DSM 17395 0.54 ± 0.15 DSM 26640 0.34 ± 0.14 F1926 0.10 ± 0.11 FO-BEG1 0.24 ± 0.01 27-4 0.12 ± 0.01 DSM 23566 0 ± 0

When growing the bacteria the TDA production is followed by the production of a brown pigment whether it be in Marine Broth or on Marine Agar (Bruhn et al., 2005), and the pigment is not produced by mutants deficient in TDA production (D’Alvise et al., 2014). The production of is highly dependent on the cultivation conditions. In broth cultures highest TDA production is obtained in stagnant cultures (Bruhn et al., 2006). Another important condition for antibacterial active TDA is iron as no antibacterial effect is seen in cultures grown in iron-free media together with the loss of pigmentation (D’Alvise et al., 2015). Temperature also influences production of TDA when grown in Marine Broth. In Paper 1 (Grotkjær et al., 2016b) the inhibitory activity of the two Phaeobacter sp. strains and seven reference strains were grown at different temperatures and the supernatant was tested against V. anguillarum (Table 5) and V. harveyi (Table 6). The size of inhibition diameter was correlated to the strains ability to antagonize the two Vibrio strains at different temperatures and it seemed that there and in general the size of the zones increased in diameter with increasing growth temperature up to 25◦C after which it decreased slightly.

Table 5 Inhibition zone diameters against V. anguillarum 90-11-287 by supernatants of Phaeobacter sp. S26 and S60, Phaeobacter inhibens DSM 16374T, Phaeobacter inhibens DSM 17395, Phaeobacter gallaeciensis DSM 26640T, Phaeobacter sp. 27-4, Ruegeria mobilis F1926, Pseudovibrio sp. FO- BEG1 and Pseudophaeobacter arcticus DSM 23566T grown at different temperatures (Grotkjær et al., 2016b).

Inhibition diameter (mm) under different temperature/time conditions Strain 5°C/168 h 10°C/120 h 15°C/96 h 20°C/72 h 25°C/72 h 30°C/ 72 h S26 0 ± 0 1,2 ± 0 2,7 ± 0,1 2,7 ± 0,2 3,3 ± 0,1 2,9 ± 0,1 S60 0 ± 0 0,9 ± 0,2 2,9 ± 0,1 3,1 ± 0,1 3,3 ± 0,1 2,9 ± 0,1 DSM 16374 1,3 ± 0,1 1,9 ± 0,1 2,7 ± 0,2 3,2 ± 0,1 2,6 ± 0,4 2,8 ± 0,2 DSM 17395 1,1 ± 0,1 1,9 ± 0,2 2,9 ± 0,1 3,1 ± 0,2 2,9 ± 0,1 2,9 ± 0,1 DSM 26640 0,4 ± 0,4 1,6 ± 0,2 2,7 ± 0,2 3,2 ± 0 3,1 ± 0,1 2,6 ± 0 27-4 0 ± 0 1,4 ± 0,1 2,8 ± 0,1 2,8 ± 0,2 2,9 ± 0,1 2,4 ± 0,1 F1926 0 ± 0 1 ± 0,1 2,7 ± 0,3 2,6 ± 0 3,1 ± 0,1 2,7 ± 0,1 FO-BEG1 0 ± 0 0 ± 0 0 ± 0 1,4 ± 0,2 2,7 ± 0,1 2,4 ± 0,1 DSM 23566 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0 Sterile MB(control) 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0

Table 6. Inhibition zone diameters against V. harveyi VIB391 by supernatants of Phaeobacter sp. S26 and S60, Phaeobacter inhibens DSM 16374T, Phaeobacter inhibens DSM 17395, Phaeobacter gallaeciensis DSM 26640T, Phaeobacter sp. 27-4, Ruegeria mobilis F1926, Pseudovibrio sp. FO-BEG1 and Pseudophaeobacter arcticus DSM 23566T grown at different temperatures (Grotkjær et al., 2016b).

Inhibition diameter (mm) under different temperature/time conditions Strain 5°C/168 h 10°C/120 h 15°C/96 h 20°C/72 h 25°C/72 h 30°C/ 72 h

S26 0,4 ± 0,4 1 ± 0,1 2,2 ± 0,1 2,2 ± 0,1 2,2 ± 0,1 2,1 ± 0,1 S60 0 ± 0 0,4 ± 0,4 2,1 ± 0,1 2 ± 0 2,3 ± 0,1 2 ± 0,1 DSM 16374 1,3 ± 0,2 2,1 ± 0,2 2,1 ± 0 2,2 ± 0,1 2 ± 0,1 2 ± 0,1 DSM 17395 0,9 ± 0,3 1,8 ± 0 2,3 ± 0,1 2,2 ± 0,1 2,1 ± 0 2 ± 0 DSM 26640 0,5 ± 0,5 1,7 ± 0,4 2,3 ± 0,1 2,4 ± 0,1 2 ± 0 2 ± 0 27-4 0 ± 0 1,1 ± 0,2 1,9 ± 0,1 2,1 ± 0,1 2,1 ± 0,1 1 ± 0,1 F1926 0 ± 0 0,45 ± 0,2 1,9 ± 0,1 1,9 ± 0,1 2 ± 0,1 2 ± 0 FO-BEG1 0 ± 0 0 ± 0 0 ± 0 0 ± 0 1,9 ± 0,1 1,7 ± 0,1 DSM 23566 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0 Sterile MB (control) 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0 ± 0

Several genes has been identified for the biosynthesis pathway of TDA (Berger et al., 2011; Geng and Belas, 2010; Geng et al., 2008; Thole et al., 2012; Zhao et al., 2016), where six of the genes (tdaA-F) are located on a large plasmid (Geng et al., 2008; Thole et al., 2012). In Paper 1 (Grotkjær et al., 2016b) the organization of the tda-genes in the two Phaeobacter sp. strains was very similar to the TDA biosynthesis gene clusters that have previously been reported in P. inhibens and P. gallaeciensis, and concomitantly the tda genes of S26 and S60 exhibited high AA sequence similarity to the tda-genes of these species (Table 7). Compared to the Ruegeria mobilis, which is known to regulate TDA expression differently as compared to the Phaeobacter species (Porsby et al., 2008), the sequence similarities were lower. This was especially true for the tdaA gene which encodes the positive regulator of the TDA biosynthesis cluster.

Table 7. tda gene similarity between the strains Phaeobacter sp. S26 and S60, and Phaeobacter inhibens DSM 16374T, Phaeobacter inhibens DSM 17395, Phaeobacter gallaeciensis DSM 26640T, Ruegeria mobilis F1926 and Pseudovibrio sp. FO-BEG1 (Grotkjær et al., 2016b).

AA sequence similarity (%) Strain Gene S26 S60 DSM DSM DSM F1926 FO- 16374T 17395 26640T BEG1 S26 tdaA - 100 98 98 98 49 67 tdaB - 98 92 91 89 54 66 tdaC - 99 99 98 98 60 72 tdaD - 98 98 98 98 72 82 tdaE - 99 97 97 98 71 69 tdaF - 98 99 99 99 72 72 tdaH - 99 94 94 93 69 - S60 tdaA 100 - 98 98 98 49 67 tdaB 98 - 92 92 91 53 66 tdaC 99 - 98 97 98 59 72 tdaD 98 - 98 98 98 72 81 tdaE 99 - 97 97 98 71 69 tdaF 100 - 99 99 99 72 72 tdaH 99 - 94 94 93 69 -

Attempts to select for TDA resistance and tolerance in pathogenic bacteria have not been successful (Porsby et al., 2011; Rasmussen et al., 2016), suggesting that TDA resistance does not arise easily, likely due to a highly conserved target (or targets). A recent study demonstrated (Wilson et al., 2015), that cell membranes and the proton motive force (PMF) were targets of TDA, and TDA was suggested to function as an electroneutral proton antiporter, facilitating a one-to-one exchange of H+ for a single charged metal ion and thus disrupting the transmembrane proton gradient (ΔpH) (Wilson et al., 2015). Also, genes conferring resistance to TDA were found in Phaeobacter inhibens, and Escherichia coli became resistant when harboring tdaR on a conjugative plasmid (Wilson et al., 2015). 43

3.1.2 Probiotic potential

The microbial community in an aquaculture setting is very diverse and it includes pathogenic, innocuous and beneficial bacteria. The balance of this community can be a determining factor for a successful rearing environment. The use of probiotic bacteria can improve and help this balance, and to minimize the environmental impact it is very reasonable to use probiotic bacteria already occurring in the rearing environment. These probionts are already adapted to the environmental conditions and stress they will be exposed to.

In this PhD study, probiotic bacteria was isolated from a Greek rearing unit and by in-depth molecular work proposed as a new Phaeobacter species (Figure 12, 13 and Table 3) (Grotkjær et al., 2016b). These strains together with the well-studied Phaeobacter inhibens DSM 17395 and its TDA-negative mutant were inoculated into co-cultures of axenic Artemia in their exponential growth phase. To evaluate their probiotic potential, two strains of Vibrios were added separately. All the probiotic strains grew well in Artemia cultures, reaching densities of 107 to 108 CFU mL−1 in all the co-cultures (Grotkjær et al., 2016b). The numbers of V. anguillarum and its TDA-negative mutant V. harveyi decreased the presence of Phaeobacter sp. S26, S60 and P. inhibens (Figure 15). Reductions were in the order of four log units for V. anguillarum units and up to three log units for V. harveyi, as compared to the Artemia controls with only V. anguillarum or V. harveyi (Grotkjær et al., 2016b). It was proposed in this study that TDA-producing Phaeobacter strains

44 isolated from Mediterranean aquaculture units had potential as probiotics against V. harveyi and V. anguillarum in axenic cultures of Artemia in accordance with other findings (D’Alvise et al., 2012; Planas et al., 2006; Porsby and Gram, 2016). Furthermore it was proposed that the probiotic bacteria could be introduced at the live feed stage.

Figure 15. Reduction of A) V. anguillarum and B) V. harveyi in co-culture with Artemia by Phaeobacter spp. Colony-forming units of V. anguillarum (A) and V. harveyi (B) inoculated at 103 CFU/ml alone (Δ), in the presence of P. inhibens DSM 17395 (), in the presence of the P. inhibens DSM 17395 TDA negative mutant, MJG-G6 (), in the presence of Phaeobacter sp. S26 (), in the presence of Phaeobacter sp. S60 (○), and in the untreated controls (). Points are average of two biological replicates and error bars are standard deviation of the mean (Grotkjær et al., 2016b).

Introducing the probiotic bacteria at the live feed stage it is important to understand the complex interactions between the microbes and the live feed (Berland et al., 1970; Grossart and Simon, 2007; Nakase and Eguchi, 2007; Salvesen et al., 2000) and to determine if the probiotic bacteria are capable of pathogen suppression also in systems with a complex microbiota. These dense cultures allow the opportunistic pathogenic bacteria to proliferate and this raises the question whether they will interfere with the growth and colonization of the probiont and, more importantly, how they affect the antagonism of the pathogens by the

45 probionts. Also, the probiotic bacteria may affect the composition of the natural microbiota and since this community interacts in several ways with the algae (Grossart and Simon, 2007; Jensen et al., 1996; Seyedsayamdost et al., 2011), it is important to know how addition of probiotic bacteria potentially affects the natural microbiota. Most studies testing the potential of different probiotic strategies have been based on axenic systems. Therefore the need to move towards a more natural setting was needed to further evaluate the probiotic bacteria. In Paper 2 (Grotkjær et al., 2016a) the same challenge model of co-culture from Paper 1 (Grotkjær et al., 2016b) was used, except that only V. anguillarum was used as pathogen (Grotkjær et al., 2016a). First the natural setting of live feed was mimicked by the addition of four bacterial strains isolated from aquaculture. Thereafter non-axenic live feed samples of Artemia and micro algae from aquaculture was used. The probiont P. inhibens DSM 17395 GMr colonized all of the Artemia and micro algae cultures reaching densities of 106 to 108 CFU/ml after 48 hours irrespective of addition of a background microbiota. . The concentration of V. anguillarum decreased in the presence of P. inhibens in both culture settings (Figure 16). The reduction was in order of 4 log units in the non-axenic Artemia cultures and 3 log units in non-axenic micro algae cultures (Figure 16). The two studies from Paper 1 and 2 (Grotkjær et al., 2016a, 2016b) clearly demonstrates the high potential as a probiotic bacteria the strains from the Phaeobacter genus have when introduced I live feed cultures.

Figure 16. Counts of V. anguillarum in axenic Artemia (A), in D. tertiolecta cultures (B). Gfp tagged cells of V. anguillarum inoculated at 104 CFU/ml alone (○), or in the presence of P. r r inhibens 17395 GM (□), P. inhibens 17395 GM and background strains separately (▲, ▼, ●, ■), or in the presence of all strains (Δ). Counts of V. anguillarum in D. tertiolectra () with natural microbiota and added probiont. Counts of V. anguillarum in Artemia (C) and Tetraselmis suecica (D) with natural microbiota. V. anguillarum inoculated at 104 CFU/ml r alone (Δ), or in the presence of P. inhibens 17395 GM (■). Points are average of two biological replicates and error bars are standard deviation of the mean. Points are average of two biological replicates and error bars are standard deviation of the mean (Grotkjær et al., 2016a).

Introducing probiotics on a larger scale requires understanding of any potential side effects of long-term exposure of the pathogen to the probionts or any compounds they produce. Since TDA has pronounced antibacterial activity, the development of resistance to TDA could be a concern, especially following long-term use. The long-term effects of 47 other antibacterial compounds, such as antimicrobial peptides (AMPs), have been studied using adaptive laboratory evolution (ALE) experiments (Hein-Kristensen et al., 2013), and in Paper 3 (Rasmussen et al., 2016) that approach where chosen to evaluate the long-term exposure of Vibrio anguillarum is exposed to sublethal concentrations TDA for several hundred generations. Whether the long-term exposure of V. anguillarum to TDA affected the infectivity of the bacteria where also determined, measured in fish cell lines. No tolerance or resistance arose during the 90 days of exposure (Figure 17), and whole-genome sequencing of TDA-exposed lineages and clones revealed few mutational changes, compared to lineages grown without TDA (Table 2 in Paper 3 (Rasmussen et al., 2016)).

Figure 17. Attempt to select for TDA-tolerant or TDA-resistant V. anguillarum strains in an adaptive laboratory evolution experiment. Solid lines, individual lineages, of which 3 of 7 reached 1.5 times the wild-type MIC (18.75 µg/ml) and the rest reached 1.75 times the wild- type MIC (21.9 µg/ml). Dashed line, wild-type MIC (12.5 µg/ml TDA) (Rasmussen et al., 2016).

Amino acid-changing mutations were found in two to six different genes per clone; however, no mutations appeared unique to the TDA-exposed lineages or clones (Table 3 in Paper 3 (Rasmussen et al., 2016)). None of the virulence genes of V. anguillarum was affected, and infectivity assays using fish cell lines indicated that the TDA-exposed lineages and clones were less invasive than the wild-type strain (Rasmussen et al., 2016). Thus, long-term TDA exposure does not appear to result in TDA resistance and the physiology of V. anguillarum appears unaffected, supporting the application of TDA-producing roseobacters as probiotics in aquaculture (Rasmussen et al., 2016).

4. Concluding remarks and future perspectives

As the aquaculture industry continues to grow to be one of the biggest food-producing sectors, the environmental impact also grows. New sustainable technologies have to be developed in order to minimize the environmental impact of over usage of antibiotics. The use of antibiotics to treat and prevent bacterial diseases in aquaculture has to be limited to special cases and needs to be banned from the routine of aquaculture production. Juvenile and adult fish should be protected by multivalent vaccines. For fish larvae and invertebrates, which cannot be vaccinated, the main focus of development should be in terms of probiotic bacteria, bacteriophages and prebiotics.

This study focused on use of the probiotic strains from the Phaeobacter genus and where to insert them in the aquaculture system. Isolating potential probiotic bacteria from aquaculture not only revealed the presence of the well-studied Phaeobacter inhibens but also a potential new Phaeobacter species. The pathogenic bacteria that gets introduced in the system comes mainly from the live feed, and therefore it seems as a natural insertion point. The Phaeobacter strains were shown to reduce concentrations of the pathogens Vibrio anguillarum and Vibrio harveyi in Artemia and microalgae co-cultures. Furthermore it was shown that the Phaeobacter strains could colonize and reduce in axenic, mimicked and non-axenic cultures from aquaculture, which could be regarded as a proof of concept.

Production of the antibacterial compound tropodithietic acid (TDA) was identified as the major mechanism of action by including mutants deficient in TDA synthesis. A concern could rise in the terms of replacing an antibiotic with an antibacterial compound producing bacteria, and therefore resistance would be an issue again or affect the disease- causing ability of the fish pathogen. Exposing the fish pathogen Vibrio anguillarum to increasing TDA concentrations over 3 months did not reveal any tolerance or resistance to TDA, and the subsequent infection assays revealed that none of the TDA-exposed clones had increased virulence toward fish cells. Hence, again a study that shows the huge potential of using TDA-producing bacteria as non-risk based disease control measure in aquaculture.

A future perspective of the present PhD study could be to determine the probiotic effect in other live feed cultures such as rotifer and copepod. A two-way strategy of using probiotic bacteria and bacteriophages could further reduce the concentration of fish pathogens. The phage and probiont strategy could be tested in both live feed cultures and in fish larvae challenge trials.

5. Acknowledgement

I owe my supervisor Lone Gram a more than huge thank you, for not only giving me the chance to do a PhD in her group, but for being a fantastic mentor. The three years have been a rollercoaster of a ride, but her determination and constant believe in me has giving me confidence to finish the work proudly. I would also like to thank my two co-supervisors Mikkel Bentzon-Tilia and Paul D’Alvise for numerous of discussions, work and for the guidance through the three years. A big thanks to Kristian Fog Nielsen for his work on Paper 1 and to Bastian Barker Rasmussen for all his work on the third paper.

Many thanks to Peter Bossier, Kristof Dierckens, and the rest of the people from The Artemia Reference Center at the University of Ghent. It was three amazing months in Belgium. Not to forget my two favorite lab colleagues Ruy Santos and Navid Pormehr who made the stay in Belgium unforgettable.

An important thing in a good work environment is the colleagues; I have always felt welcome and appreciated. This is because of the amazing people of GramLab whom I had the pleasure of working with for three years.

I am very grateful for the support of my family and friends. Finally, I would like to thank my girlfriend Martine for her endless love and support, for always cheering through the long work days and for giving me the greatest gift ever in the shape of our soon to come daughter.

6. References

Alcaide, E., Gil-Sanz, C., Sanjuán, E., Esteve, D., Amaro, C., Silveira, L., 2001. Vibrio harveyi causes disease in seahorse, Hippocampus sp. J. Fish Dis. 24, 311–313. doi:10.1046/j.1365- 2761.2001.00297.x

Al-Dohail, M.A., Hashim, R., Aliyu-Paiko, M., 2009. Effects of the probiotic, Lactobacillus acidophilus, on the growth performance, haematology parameters and immunoglobulin concentration in African Catfish (Clarias gariepinus>, Burchell 1822) fingerling. Aquac. Res. 40, 1642–1652. doi:10.1111/j.1365-2109.2009.02265.x

Auch, A.F., Klenk, H.-P., Göker, M., 2010. Standard operating procedure for calculating genome- to-genome distances based on high-scoring segment pairs. Stand. Genomic Sci. 2, 142–8. doi:10.4056/sigs.541628

Austin, B., Allen, D.A., 1982. Microbiology of laboratory-hatched brine shrimp (Artemia). Aquaculture 26, 369–383. doi:10.1016/0044-8486(82)90170-3

Austin, B., Stuckey, L., Robertson, P., Effendi, I., Griffith, D., 1995. A probiotic strain of Vibrio alginolyticus effective in reducing diseases caused by Aeromonas salmonicida, Vibrio anguillarum and Vibrio ordalii. J. Fish Dis. 18, 93–96.

Austin, B., Zhang, X.-H., 2006. Vibrio harveyi: a significant pathogen of marine vertebrates and invertebrates. Lett. Appl. Microbiol. 43, 119–24. doi:10.1111/j.1472-765X.2006.01989.x

Balcázar, J.L., Blas, I. de, Ruiz-Zarzuela, I., Cunningham, D., Vendrell, D., Múzquiz, J.L., 2006. The role of probiotics in aquaculture. Vet. Microbiol. 114, 173–186. doi:10.1016/j.vetmic.2006.01.009

Becker, P., Gillan, D., Lanterbecq, D., Jangoux, M., Rasolofonirina, R., Rakotovao, J., Eeckhaut, I., 2004. The skin ulceration disease in cultivated juveniles of Holothuria scabra (Holothuroidea, Echinodermata). Aquaculture 242, 13–30. doi:10.1016/j.aquaculture.2003.11.018

Benavete, G., Gatesoupe, F., 1988. Bacteria Associated with Cultured Rotifers and Artemia are Detrimental to Larval Turbot, Scophthalmus Maximus L. Aquac. Eng. 7, 289–293.

Berger, M., Neumann, A., Schulz, S., Simon, M., Brinkhoff, T., 2011. Tropodithietic acid production in Phaeobacter gallaeciensis is regulated by N-acyl homoserine lactone- mediated quorum sensing. J. Bacteriol. 193, 6576–85. doi:10.1128/JB.05818-11

Bergh, Ø., Hansen, G., Taxt, R., 1992. Experimental infection of eggs and yolk sac larvae of halibut, Hippoglossus hippoglossus L. J. Fish Dis. 15, 379–391. doi:10.1111/j.1365- 2761.1992.tb01237.x

Berland, B.R., Bonin, D.J., Maestrini, S.Y., 1970. Study of bacteria associated with marine algae in culture. Mar. Biol. 3, 68–76. doi:10.1007/BF00698862

Boyd, C.E., Massaut, L., 1999. Risks associated with the use of chemicals in pond aquaculture. Aquac. Eng. 20, 113–132. doi:10.1016/S0144-8609(99)00010-2

Bricknell, I., Dalmo, R., 2005. The use of immunostimulants in fish larval aquaculture. Fish Shellfish Immunol. 19, 457–472. doi:10.1016/j.fsi.2005.03.008

Brinkhoff, T., Bach, G., Heidorn, T., Liang, L., Schlingloff, A., Simon, M., 2004. Antibiotic Production by a Roseobacter Clade-Affiliated Species from the German Wadden Sea and Its Antagonistic Effects on Indigenous Isolates. Appl. Environ. Microbiol. 70, 2560–2565. doi:10.1128/AEM.70.4.2560

Bron, P. a., van Baarlen, P., Kleerebezem, M., 2011. Emerging molecular insights into the interaction between probiotics and the host intestinal mucosa. Nat. Rev. Microbiol. 10, 66–78. doi:10.1038/nrmicro2690

Brown, G.D., Taylor, P.R., Reid, D.M., Willment, J. a., Williams, D.L., Martinez-Pomares, L., Wong, S.Y.C., Gordon, S., 2002. Dectin-1 Is A Major-Glucan Receptor On Macrophages. J. Exp. Med. 196, 407–412. doi:10.1084/jem.20020470

Brown, M.R., Jeffrey, S.W., Volkman, J.K., Dunstan, G.A., 1997. Nutritional properties of microaglae for mariculture. Aquaculture 151, 315–331. doi:10.1016/S0044- 8486(96)01501-3

Brudeseth, B.E., Wiulsrød, R., Fredriksen, B.N., Lindmo, K., Løkling, K.E., Bordevik, M., Steine, N., Klevan, A., Gravningen, K., 2013. Status and future perspectives of vaccines for industrialised fin-fish farming. Fish Shellfish Immunol. 35, 1759–1768.

doi:10.1016/j.fsi.2013.05.029

Bruhn, J.B., Gram, L., Belas, R., 2007. Production of antibacterial compounds and biofilm formation by Roseobacter species are influenced by culture conditions. Appl. Environ. Microbiol. 73, 442–450. doi:10.1128/AEM.02238-06

Bruhn, J.B., Haagensen, J.A., Bagge-Ravn, D., Gram, L., 2006. Culture conditions of Roseobacter strain 27-4 affect its attachment and biofilm formation as quantified by real-time PCR. Appl. Environ. Microbiol. 72, 3011–3015. doi:10.1128/AEM.72.4.3011

Bruhn, J.B., Nielsen, K.F., Hjelm, M., 2005. Ecology, inhibitory activity, and morphogenesis of a marine antagonistic bacterium belonging to the Roseobacter clade. Appl. Environ. Microbiol. 71, 7263–7270. doi:10.1128/AEM.71.11.7263

Buchan, A., González, J.M., Moran, M.A., 2005. Overview of the Marine Roseobacter Lineage. Appl. Environ. Microbiol. 71, 5665–5677. doi:10.1128/AEM.71.10.5665

Buddruhs, N., Pradella, S., Göker, M., Päuker, O., Pukall, R., Spröer, C., Schumann, P., Petersen, J., Brinkhoff, T., 2013. Molecular and phenotypic analyses reveal the non-identity of the Phaeobacter gallaeciensis type strain deposits CIP 105210T and DSM 17395. Int. J. Syst. Evol. Microbiol. 63, 4340–9. doi:10.1099/ijs.0.053900-0

Busch, K.E.T., Peruzzi, S., Tonning, F., Falk-Petersen, I.B., 2011. Effect of prey type and size on the growth, survival and pigmentation of cod (Gadus morhua L.) larvae. Aquac. Nutr. 17. doi:10.1111/j.1365-2095.2010.00800.x

Cabello, F.C., 2006. Heavy use of prophylactic antibiotics in aquaculture: a growing problem for human and animal health and for the environment. Environ. Microbiol. 8, 1137–44. doi:10.1111/j.1462-2920.2006.01054.x

Cabello, F.C., Godfrey, H.P., Buschmann, A.H., Dölz, H.J., 2016. Aquaculture as yet another environmental gateway to the development and globalisation of antimicrobial resistance. Lancet Infect. Dis. 16, e127–e133. doi:10.1016/S1473-3099(16)00100-6

Carlton, R.M., Noordman, W.H., Biswas, B., De Meester, E.D., Loessner, M.J., 2005. Bacteriophage P100 for control of Listeria monocytogenes in foods: Genome sequence, bioinformatic analyses, oral toxicity study, and application. Regul. Toxicol. Pharmacol. 43,

301–312. doi:10.1016/j.yrtph.2005.08.005

Chabrillón, M., Rico, R.M., Balebona, M.C., Moriñigo, M.A., 2005. Adhesion to sole, Solea senegalensis Kaup, mucus of microorganisms isolated from farmed fish, and their interaction with Photobacterium damselae subsp. piscicida. J. Fish Dis. 28, 229–237. doi:10.1111/j.1365-2761.2005.00623.x

Chakraborty, R.D., Chakraborty, K., Radhakrishnan, E. V., 2007. Variation in fatty acid composition of Artemia salina nauplii enriched with microalgae and baker’s yeast for use in larviculture. J. Agric. Food Chem. 55, 4043–4051. doi:10.1021/jf063654l

Chantanachookhin, C., Seikai, T., Tanaka, M., 1991. Comparative study of the ontogeny of the lymphoid organs in three species of marine fish. Aquaculture 99, 143–155.

Chu, W., McLean, R.J.C., 2016. Quorum Signal Inhibitors and Their Potential Use against Fish Diseases. J. Aquat. Anim. Health 28, 91–96. doi:10.1080/08997659.2016.1150907

D’Alvise, P.W., Lillebø, S., Prol-Garcia, M.J., Wergeland, H., Nielsen, K.F., Bergh, Ø., Gram, L., 2012. Phaeobacter gallaeciensis reduces Vibrio anguillarum in cultures of microalgae and rotifers, and prevents vibriosis in cod larvae. PLoS One 7, e43996. doi:10.1371/journal.pone.0043996

D’Alvise, P.W., Magdenoska, O., Melchiorsen, J., Nielsen, K.F., Gram, L., 2014. Biofilm formation and antibiotic production in Ruegeria mobilis are influenced by intracellular concentrations of cyclic dimeric guanosinmonophosphate. Environ. Microbiol. 16, 1252– 66. doi:10.1111/1462-2920.12265

D’Alvise, P.W., Melchiorsen, J., Porsby, C.H., Nielsen, K.F., Gram, L., 2010. Inactivation of Vibrio anguillarum by attached and planktonic Roseobacter cells. Appl. Environ. Microbiol. 76, 2366–2370. doi:10.1128/AEM.02717-09

D’Alvise, P.W., Phippen, C.B.W., Nielsen, K.F., Gram, L., 2015. Influence of iron on the production of the antibacterial compound tropodithietic acid and its non-inhibitory analog in Phaeobacter inhibens. Appl. Environ. Microbiol. AEM.02992–15. doi:10.1128/AEM.02992- 15

Dalmo, R.A., Ingebrigtsen, K., Bogwald, J., 1997. Non-specific defence mechanisms in fish, with

particular reference to the reticuloendothelial system (RES). J. Fish Dis. 20, 241–273. doi:10.1046/j.1365-2761.1997.00302.x

Dang, H., Lovell, C.R., 2000. Bacterial primary colonization and early succession on surfaces in marine waters as determined by amplified rRNA gene restriction analysis and sequence analysis of. Appl. Environ. Microbiol. doi:10.1128/AEM.66.2.467-475.2000.Updated

Das, S., Ward, L.R., Burke, C., 2008. Prospects of using marine actinobacteria as probiotics in aquaculture. Appl. Microbiol. Biotechnol. 81, 419–429. doi:10.1007/s00253-008-1731-8

Defoirdt, T., Boon, N., Sorgeloos, P., Verstraete, W., Bossier, P., 2008. Quorum sensing and quorum quenching in Vibrio harveyi: lessons learned from in vivo work. ISME J. 2, 19–26. doi:10.1038/ismej.2007.92

Diggles, B.K., Moss, G.A., Carson, J., Anderson, C.D., 2000. Luminous vibriosis in rock lobster Jasus verreauxi (Decapoda: Palinuridae) phyllosoma larvae associated with infection by Vibrio harveyi. Dis. Aquat. Organ. 43, 127–137. doi:10.3354/dao043127

Drillet, G., Frouël, S., Sichlau, M.H., Jepsen, P.M., Højgaard, J.K., Joardeer, A.K., Hansen, B.W., 2011. Status and recommendations on marine copepod cultivation for use as live feed. Aquaculture 315, 155–166. doi:10.1016/j.aquaculture.2011.02.027

Edwards, P., 2015. Aquaculture environment interactions: Past, present and likely future trends. Aquaculture 447, 2–14. doi:10.1016/j.aquaculture.2015.02.001

Egidius, E., Andersen, K., 1977. Norwegian reference strains of Vibrio anguillarum. Aquaculture 10, 215–219.

Enright, M.C., Day, N.P.J., Davies, C.E., Peacock, S.J., Spratt, B.G., 2000. Multilocus sequence typing for characterization of methicillin-resistant and methicillin-susceptible clones of Staphylococcus aureus. J. Clin. Microbiol. 38, 1008–1015.

FAO Fisheries and Aquaculture Department. The State of World Fisheries and Aquaculture 2014, n.d.

FAO, WHO. Report of a joint FAO/WHO expert consultation on evaluation of health and nutritional properties of probiotics in food including powder milk with live lactic acid 57

bacteria. 2001, n.d.

Frans, I., Dierckens, K., Crauwels, S., Assche, A. Van, Leisner, J., Larsen, M.H., Michiels, C.W., Willems, K.A., Lievens, B., Bossier, P., Rediers, H., 2013. Virulence Assessment of Vibrio anguillarum Using Sea Bass (Dicentrarchus labrax) Larvae Correspond with Genotypic and Phenotypic Characterization? PLoS One 8, e70477. doi:10.1371/journal.pone.0070477

Frans, I., Michiels, C.W., Bossier, P., Willems, K. a, Lievens, B., Rediers, H., 2011. Vibrio anguillarum as a fish pathogen: virulence factors, diagnosis and prevention. J. Fish Dis. 34, 643–61. doi:10.1111/j.1365-2761.2011.01279.x

Gatesoupe, F.J., 1991. The effect of three strains of lactic bacteria on the production rate of rotifers, Brachionus plicatilis, and their dietary value for larval turbot, Scophthalmus maximus. Aquaculture 96, 335–342. doi:10.1016/0044-8486(91)90162-Z

Gatesoupe, F.-J., 2002. Probiotic and formaldehyde treatments of Artemia nauplii as food for larval pollack, Pollachius pollachius. Aquaculture 212, 347–360. doi:10.1016/S0044- 8486(02)00138-2

Geng, H., Belas, R., 2010. Molecular mechanisms underlying roseobacter-phytoplankton symbioses. Curr. Opin. Biotechnol. 21, 332–8. doi:10.1016/j.copbio.2010.03.013

Geng, H., Bruhn, J.B., Nielsen, K.F., Gram, L., Belas, R., 2008. Genetic dissection of tropodithietic acid biosynthesis by marine roseobacters. Appl. Environ. Microbiol. 74, 1535–1545. doi:10.1128/AEM.02339-07

González, J.M., Moran, M.A., 1997. Numerical dominance of a group of marine bacteria in the alpha-subclass of the class Proteobacteria in coastal seawater. Appl. Environ. Microbiol. 63, 4237–4242.

Gram, L., Melchiorsen, J., Bruhn, J.B., 2010. Antibacterial activity of marine culturable bacteria collected from a global sampling of ocean surface waters and surface swabs of marine organisms. Mar. Biotechnol. 12, 439–51. doi:10.1007/s10126-009-9233-y

Gram, L., Melchiorsen, J., Spanggaard, B., Huber, I., Nielsen, T.F., 1999. Inhibition of Vibrio anguillarum by Pseudomonas fluorescens AH2 , a Possible Probiotic Treatment of Fish.

Appl. Environ. Microbiol. 65, 969–973.

Grimes, D.J., Stemmler, J., Hada, H., May, E.B., Maneval, D., Hetrick, F.M., Jones, R.T., Stoskopf, M., Colwell, R.R., 1984. Vibrio species associated with mortality of sharks held in captivity. Microb. Ecol. 10, 271–282. doi:10.1007/BF02010940

Grisez, L., Ceusters, R., Ollevier, F., 1991. The use of API 20E for the identification of Vibrio anguillarum and V. ordalii. J. Fish Dis. 14, 359–365. doi:10.1111/j.1365- 2761.1991.tb00833.x

Grossart, H., Simon, M., 2007. Interactions of planktonic algae and bacteria: effects on algal growth and organic matter dynamics. Aquat. Microb. Ecol. 47, 163–176. doi:10.3354/ame047163

Grotkjær, T., Bentzon-Tilia, M., D’Alvise, P., Dierckens, K., Bossier, P., Gram, L., 2016a. Phaeobacter inhibens as probiotic bacteria in non-axenic Artemia and algae cultures. Aquaculture 462, 64–69. doi:10.1016/j.aquaculture.2016.05.001

Grotkjær, T., Bentzon-tilia, M., Dourala, N., Nielsen, K.F., Gram, L., Street, N., 2016b. Isolation of TDA-producing Phaeobacter strains from sea bass larval rearing units and their probiotic effect against pathogenic Vibrio spp . in Artemia cultures. Syst. Appl. Microbiol. 1–34. doi:10.1016/j.syapm.2016.01.005

Gugliandolo, C., Irrera, G.P., Lentini, V., Maugeri, T.L., 2008. Pathogenic Vibrio, Aeromonas and Arcobacter spp. associated with copepods in the Straits of Messina (Italy). Mar. Pollut. Bull. 56, 600–606. doi:10.1016/j.marpolbul.2007.12.001

Gupta, A., Geetika, V., Gupta, P., 2016. Growth performance, feed utilization, digestive enzyme activity, innate immunity and protection against Vibrio harveyi of freshwater prawn, Macrobrachium rosenbergii fed diets supplemented with Bacillus coagulans. Aquac. Int. doi:10.1007/s10499-016-9996-x

Haché, R., Lanteigne, C., Hébert, Y., 2016a. Salt as a decontamination agent to control bacterial load in Artemia salina cultures. Aquaculture 452, 24–27. doi:10.1016/j.aquaculture.2015.10.017

Haché, R., Plante, S., Forward, B.S., Pernet, F., 2016b. Cod Larviculture Using High-density Rotifer

Production with Different Enrichments. J. World Aquac. Soc. doi:10.1111/jwas.12310

Hada, H.S., West, P. a., Lee, J. V., Stemmler, J., Colwell, R.R., 1984. Vibrio tubiashii sp. nov., a Pathogen of Bivalve Mollusks. Int. J. Syst. Bacteriol. 34, 1–4. doi:10.1099/00207713-34-1-1

Hamre, K., 2016. Nutrient profiles of rotifers (Brachionus sp.) and rotifer diets from four different marine fish hatcheries. Aquaculture 450, 136–142. doi:10.1016/j.aquaculture.2015.07.016

Harrell, L.W., Novotny, A.J., Schiewe, M.H., Hodgins, H.O., 1976. Isolation and description of two vibriosis pathogenic to Pacific salmon in Puget Sound, Washington. Fish. Bull. 74, 447– 449.

Harris, L.J., Owens, L., 1999. Production of exotoxins by two luminous Vibrio harveyi strains known to be primary pathogens of Penaeus monodon larvae. Dis. Aquat. Organ. 38, 11– 22. doi:10.3354/dao038011

Hein-Kristensen, L., Franzyk, H., Holch, A., Gram, L., 2013. Adaptive evolution of Escherichia coli to an α-peptide/β-peptoid peptidomimetic induces stable resistance. PLoS One 8, e73620. doi:10.1371/journal.pone.0073620

Hemaiswarya, S., Raja, R., Kumar, R.R., Ganesan, V., Anbazhagan, C., 2011. Microalgae: A sustainable feed source for aquaculture. World J. Microbiol. Biotechnol. 27, 1737–1746. doi:10.1007/s11274-010-0632-z

Heuer, O.E., Kruse, H., Grave, K., Collignon, P., Karunasagar, I., Angulo, F.J., 2009. Human health consequences of use of antimicrobial agents in aquaculture. Clin. Infect. Dis. 49, 1248– 1253. doi:10.1086/605667

Hjelm, M., Bergh, Ø., Riaza, A., Nielsen, J., Melchiorsen, J., Jensen, S., Duncan, H., Ahrens, P., Birkbeck, H., Gram, L., 2004a. Selection and Identification of Autochthonous Potential Probiotic Bacteria from Turbot Larvae (Scophthalmus maximus) Rearing Units. Syst. Appl. Microbiol. 27, 360–371.

Hjelm, M., Riaza, A., Formoso, F., Melchiorsen, J., Gram, L., 2004b. Seasonal incidence of autochthonous antagonistic Roseobacter spp. and Vibrionaceae strains in a turbot larva (Scophthalmus maximus) rearing system. Appl. Environ. Microbiol. 70, 7288–7294.

doi:10.1128/AEM.70.12.7288

Hong, N.T.X., Baruah, K., Vanrompay, D., Bossier, P., 2016. Characterization of phenotype variations of luminescent and non-luminescent variants of Vibrio harveyi wild type and quorum sensing mutants. J. Fish Dis. 39, 317–327. doi:10.1111/jfd.12365

Howell, B.R., 1979. Experiments of the rearing of larval turbot, Scophthalmus maximus L. Aquaculture 18, 215–225. doi:10.1016/0044-8486(79)90013-9

Høj, L., Bourne, D.G., Hall, M.R., 2009. Localization, abundance and community structure of bacteria associated with Artemia: Effects of nauplii enrichment and antimicrobial treatment. Aquaculture 293, 278–285. doi:10.1016/j.aquaculture.2009.04.024

Haastein, T., Holt, G., 1972. The occurrence of vibrio disease in wild Norwegian fish. J. Fish Biol. 4, 33–37.

Jensen, P., Kauffman, C., Fenical, W., 1996. High recovery of culturable bacteria from the surfaces of marine algae. Mar. Biol. 1–7.

Jones, J.B., Vallad, G.E., Iriarte, F.B., Obradović, A., Wernsing, M.H., Jackson, L.E., Balogh, B., Hong, J.C., Momol, M.T., 2012. Considerations for using bacteriophages for plant disease control. Bacteriophage 2, 208–214. doi:10.4161/bact.23857

Kashulin, A., Seredkina, N., Sørum, H., 2016. Cold-water vibriosis. The current status of knowledge. J. Fish Dis. 1–8. doi:10.1111/jfd.12465

Katan, T., Nash, G.W., Rise, M.L., Hall, J.R., Fernandes, J.M.O., Boyce, D., Johnsen, C.A., Gamperl, A.K., 2016. A little goes long way: Improved growth in Atlantic cod (Gadus morhua) fed small amounts of wild zooplankton. Aquaculture 451, 271–282. doi:10.1016/j.aquaculture.2015.09.014

Kesarcodi-Watson, A., Kaspar, H., Lategan, M.J., Gibson, L., 2008. Probiotics in aquaculture: The need, principles and mechanisms of action and screening processes. Aquaculture 274, 1– 14. doi:10.1016/j.aquaculture.2007.11.019

Khatoon, H., Banerjee, S., Syakir Syahiran, M., Mat Noordin, N.B., Munafi Ambok Bolong, A., Endut, A., 2016. Re-use of aquaculture wastewater in cultivating microalgae as live feed for aquaculture organisms. Desalin. Water Treat. 3994, 1–8. 61

doi:10.1080/19443994.2016.1156030

Lanes, C.F.C., Bolla, S., Fernandes, J.M.O., Nicolaisen, O., Kiron, V., Babiak, I., 2012. Nucleotide Enrichment of Live Feed: A Promising Protocol for Rearing of Atlantic Cod Gadus morhua Larvae. Mar. Biotechnol. 14, 544–558. doi:10.1007/s10126-012-9458-z

Larsen, J.L., 1984. Vibrio anguillavum : Influence of temperature , pH , NaCl concentration and incubation time on growth. J. Appl. Bacteriol. 57, 237–246.

Larsen, J.L., Pedersen, K., Dalsgaard, I., 1994. Vibrio anguillarum serovars associated with vibriosis in fish. J. Fish Dis. 17, 259–267. doi:10.1111/j.1365-2761.1994.tb00221.x

Larsen, J.L., Rasmussen, H.B., Dalsgaard, I., 1988. Study of Vibrio anguillarum strains from different sources with emphasis on ecological and pathobiological properties. Appl. Environ. Microbiol. 54, 2264–2267.

Lee, K.K., Liu, P.C., Chuang, W.H., 2002. Pathogenesis of gastroenteritis caused by Vibrio carchariae in cultured marine fish. Mar. Biotechnol. 4, 267–277. doi:10.1007/s10126-002- 0018-9

Li, X., Defoirdt, T., Bossier, P., 2015. Relation between virulence of Vibrio anguillarum strains and response to the host factors mucin, bile salts and cholesterol. J. Appl. Microbiol. 119, 25– 32. doi:10.1111/jam.12813

Lubzens, E., Zmora, O., Barr, Y., 2001. Biotechnology and aquaculture of rotifers. Hydrobiologia 446-447, 337–353. doi:10.1023/A:1017563125103

Luo, H., Moran, M.A., 2014. Evolutionary ecology of the marine Roseobacter clade. Microbiol. Mol. Biol. Rev. 78, 573–87. doi:10.1128/MMBR.00020-14

Madhusudana Rao, B., Lalitha, K. V., 2015. Bacteriophages for aquaculture: Are they beneficial or inimical. Aquaculture 437, 146–154. doi:10.1016/j.aquaculture.2014.11.039

McCarthy, D.H., 1976. Vibrio disease in eels. J. Fish Biol. 8, 317–320.

McEvoy, L.A., Naess, T., Bell, J.G., Lie, O., 1998. Lipid and fatty acid composition of normal and malpigmented Atlantic halibut (Hippoglossus hippoglossus) fed enriched Artemia: A comparison with fry fed wild copepods. Aquaculture 163, 237–250. doi:10.1016/S0044-

8486(98)00237-3

Montgomery, M.T., Kirchman, D.L., 1993. Role of chitin-binding protein in the specific attachment of the marine bacterium Vibrio harveyi to chitin. Appl. Environ. Microbiol. 59, 373–379.

Moran, M., Belas, R., Schell, M., González, J., Sun, F., Sun, S., Binder, B., Edmonds, J., Ye, W., Orcutt, B., Howard, E., Meile, C., Palefsky, W., Goesmann, A., Ren, Q., Paulsen, I., Ulrich, L.E., Thompson, L.S., Saunders, E., Buchan, A., 2007. Ecological genomics of marine Roseobacters. Appl. Environ. Microbiol. 73, 4559–69. doi:10.1128/AEM.02580-06

Moran, M.A., González, J.M., Kiene, R.P., 2003. Linking a Bacterial Taxon to Sulfur Cycling in the Sea: Studies of the Marine Roseobacter Group. Geomicrobiol. J. 20, 375–388. doi:10.1080/01490450303901

Mussatto, S.I., Mancilha, I.M., 2007. Non-digestible oligosaccharides: A review. Carbohydr. Polym. 68, 587–597. doi:10.1016/j.carbpol.2006.12.011

Naka, H., Dias, G.M., Thompson, C.C., Dubay, C., Thompson, F.L., Crosa, J.H., 2011. Complete genome sequence of the marine fish pathogen Vibrio anguillarum harboring the pJM1 virulence plasmid and genomic comparison with other virulent strains of V. anguillarum and V. ordalii. Infect. Immun. 79, 2889–900. doi:10.1128/IAI.05138-11

Nakai, T., Park, S.C., 2002. Bacteriophage therapy of infectious diseases in aquaculture. Res. Microbiol. 153, 13–18. doi:10.1016/S0923-2508(01)01280-3

Nakase, G., Eguchi, M., 2007. Analysis of bacterial communities in Nannochloropsis sp. cultures used for larval fish production. Fish. Sci. 73, 543–549. doi:10.1111/j.1444- 2906.2007.01366.x

Nayak, S.K., 2010. Probiotics and immunity: A fish perspective. Fish Shellfish Immunol. 29, 2–14. doi:10.1016/j.fsi.2010.02.017

Nevejan, N., De Schryver, P., Wille, M., Dierckens, K., Baruah, K., Van Stappen, G., 2016. Bacteria as food in aquaculture: do they make a difference? Rev. Aquac. 1–33. doi:10.1111/raq.12155

Nicolas, J.L., Robic, E., Ansquer, D., 1989. Bacterial flora associated with a trophic chain 63

consisting of microalgae, rotifers and turbot larvae: Influence of bacteria on larval survival. Aquaculture 83, 237–248. doi:10.1016/0044-8486(89)90036-7

Naas, I.L.E., Naess, T., Harboe, T., 1992. Enhanced first feeding of halibut larvae (Hippoglossus hippoglossus L.) in green water. Aquaculture 105, 143–156.

Olafsen, J. a., 2001. Interactions between fish larvae and bacteria in marine aquaculture. Aquaculture 200, 223–247. doi:10.1016/S0044-8486(01)00702-5

Olivotto, I., Planas, M., Simões, N., Holt, G.J., Avella, M.A., Calado, R., 2011. Advances in Breeding and Rearing Marine Ornamentals. J. World Aquac. Soc. 42, 135–166. doi:10.1111/j.1749-7345.2011.00453.x

Pacha, R.E., Kiehn, E.D., 1969. Characterization and relatedness of marine vibrios pathogenic to fish: physiology, serology, and epidemiology. J. Bacteriol. 100, 1242–1247.

Pandiyan, P., Balaraman, D., Thirunavukkarasu, R., George, E.G.J., Subaramaniyan, K., Manikkam, S., Sadayappan, B., 2013. Probiotics in aquaculture. Drug Invent. Today 5, 55–59. doi:10.1016/j.dit.2013.03.003

Payne, M.., Rippingale, R.., 2001. Intensive cultivation of the calanoid copepod Gladioferens imparipes. Aquaculture 201, 329–342. doi:10.1016/S0044-8486(01)00608-1

Pedersen, K., Gram, L., Austin, D., Austin, B., 1997. Pathogenicity of Vibrio anguillarum serogroup O1 strains compared to plasmids, outer membrane protein profiles and siderophore production. J. Appl. Microbiol. 82, 365–371.

Planas, M., Pérez-Lorenzo, M., Hjelm, M., Gram, L., Uglenes Fiksdal, I., Bergh, Ø., Pintado, J., 2006. Probiotic effect in vivo of Roseobacter strain 27-4 against Vibrio (Listonella) anguillarum infections in turbot (Scophthalmus maximus L.) larvae. Aquaculture 255, 323– 333. doi:10.1016/j.aquaculture.2005.11.039

Porsby, C., Webber, M., Nielsen, K.F., Piddock, L.J., Gram, L., 2011. Resistance and tolerance to tropodithietic acid, an antimicrobial in aquaculture, is hard to select. Antimicrob. Agents Chemother. 55, 1332–1337. doi:10.1128/AAC.01222-10

Porsby, C.H., Gram, L., 2016. Phaeobacter inhibens as biocontrol agent against Vibrio vulnificus in oyster models. Food Microbiol. 57, 63–70. doi:10.1016/j.fm.2016.01.005 64

Porsby, C.H., Nielsen, K.F., Gram, L., 2008. Phaeobacter and Rugeria species of the Roseobacter clade colonize separate niches in a Danish turbot (Scophthalmus maximus)-rearing farm and antagonize Vibrio anguillarum under different growth. Appl. Environ. Microbiol. 74, 7356–7364. doi:10.1128/AEM.01738-08

Pujalte, M.J., Sitjà-Bobadilla, A., Macián, M.C., Belloch, C., Alvarez-Pellitero, P., Pérez-Sánchez, J., Uruburu, F., Garay, E., 2003. Virulence and molecular typing of Vibrio harveyi strains isolated from cultured dentex, gilthead sea bream and European sea bass. Syst. Appl. Microbiol. 26, 284–92. doi:10.1078/072320203322346146

Rao, D., Webb, J.S., Kjelleberg, S., 2006. Microbial colonization and competition on the marine alga Ulva australis. Appl. Environ. Microbiol. 72, 5547–55. doi:10.1128/AEM.00449-06

Rao, D., Webb, J.S., Kjelleberg, S., 2005. Competitive Interactions in Mixed-Species Biofilms Containing the Marine Bacterium Pseudoalteromonas tunicata. Appl. Environ. Microbiol. 71, 1729–1736. doi:10.1128/AEM.71.4.1729

Rappé, M.S., Giovannoni, S.J., 2003. The uncultured microbial majority. Annu. Rev. Microbiol. 57, 369–394. doi:10.1146/annurev.micro.57.030502.090759

Rasdi, N.W., Qin, J.G., 2016. Improvement of copepod nutritional quality as live food for aquaculture: A review. Aquac. Res. 47, 1–20. doi:10.1111/are.12471

Rasmussen, B.B., Grotkjær, T., D’Alvise, P.W., Yin, G., Zhang, F., Bunk, B., Spröer, C., Bentzon- Tilia, M., Gram, L., 2016. Vibrio anguillarum is genetically and phenotypically unaffected by long-term continuous exposure to the antibacterial compound tropodithietic acid. Appl. Environ. Microbiol. 82, 4802–4810. doi:10.1128/AEM.01047-16

Rasmussen, H.B., 1987. Subgrouping of lipopolisaccharide O antigens from Vibrio anguillarum serogroup 02 by immunoelectrophoretic analyses. Curr. Microbiol. 16, 39–42.

Rehulka, J., Petras, P., Marejkova, M., Aldova, E., 2016. Vibrio cholerae non-O1/non-O139 infection in fish in the Czech Republic. Vet. Med. (Praha). 60, 16–22. doi:10.17221/7921- VETMED

Reid, H.I., Treasurer, J.W., Adam, B., Birkbeck, T.H., 2009. Analysis of bacterial populations in the gut of developing cod larvae and identification of Vibrio logei, Vibrio anguillarum and

Vibrio splendidus as pathogens of cod larvae. Aquaculture 288, 36–43. doi:10.1016/j.aquaculture.2008.11.022

Reitan, K.I., Rainuzzo, J.R., Øie, G., Olsen, Y., 1993. Nutritional effects of algal addition in first- feeding of turbot (Scophthalmus maximus L.) larvae. Aquaculture 118, 257–275. doi:10.1016/0044-8486(93)90461-7

Resende, J.A., Silva, V.L., Fontes, C.O., Souza-Filho, J.A., Rocha de Oliveira, T.L., Coelho, C.M., César, D.E., Diniz, C.G., 2012. Multidrug-resistance and toxic metal tolerance of medically important bacteria isolated from an aquaculture system. Microbes Environ. 27, 449–55. doi:10.1264/jsme2.ME12049

Ringø, E., Olsen, R., Gifstad, T., Dalmo, R., Amlud, H., Hemre, G., Bakke, A., 2010. Prebiotics in aquaculture: a review. Aquac. … 16, 117–136. doi:10.1111/j.1365-2095.2009.00731.x

Roberfroid, M.B., 2005. Introducing inulin-type fructans. Br. J. Nutr. 93, S13. doi:10.1079/BJN20041350

Rodríguez, J.M., López-Romalde, S., Beaz, R., Alonso, M.C., Castro, D., Romalde, J.L., 2006. Molecular fingerprinting of Vibrio tapetis strains using three PCR-based methods: ERIC- PCR, REP-PCR and RAPD. Dis. Aquat. Organ. 69, 175–183. doi:10.3354/dao069175

Roy, S. Sen, Pal, R., 2015. Microalgae in Aquaculture: A Review with Special References to Nutritional Value and Fish Dietetics. Proc. Zool. Soc. 68, 1–8. doi:10.1007/s12595-013- 0089-9

Ruiz-Ponte, C., Cilia, V., Lambert, C., Nicolas, J., 1998. Roseobacter gallaeciensis sp. nov., a new marine bacterium isolated from rearings and collectors of the scallop Pecten maximus. Int. J. Syst. Bacteriol. 48, 537–542.

Sáenz De Rodrigánez, M.A., Díaz-Rosales, P., Chabrillón, M., Smidt, H., Arijo, S., León-Rubio, J.M., Alarcón, F.J., Balebona, M.C., Morinigo, M.A., Cara, J.B., Moyano, F.J., 2009. Effect of dietary administration of probiotics on growth and intestine functionality of juvenile Senegalese sole (Solea senegalensis, Kaup 1858). Aquac. Nutr. 15, 177–185. doi:10.1111/j.1365-2095.2008.00581.x

Sakai, M., 1999. Current research status of fish immunostimulants. Aquaculture 172, 63–92.

doi:10.1016/S0044-8486(98)00436-0

Salvesen, I., Reitan, K.I., Skjermo, J., Øie, G., 2000. Microbial environments in marine larviculture: Impacts of algal growth rates on the bacterial load in six microalgae. Aquac. Int. 275–287.

Schiewe, M.H., Crosa, J., 1981. Molecular characteristic of Vibrio anguillarum biotype 2. Can.J.Microbiol. 27, 1011–1018.

Schiewe, M.H., Trust, T.J., Crosa, J.H., 1981. Vibrio ordalii sp. nov.: A causative agent of vibriosis in fish. Curr. Microbiol. 6, 343–348. doi:10.1007/BF01567009

Scott, A.P., Middleton, C., 1979. Unicellular algae as a food for turbot (Scophthalmus maximus L.) larvae - The importance of dietary long-chain polyunsaturated fatty acids. Aquaculture 18, 227–240. doi:10.1016/0044-8486(79)90014-0

Selje, N., Simon, M., Brinkhoff, T., 2004. A newly discovered Roseobacter cluster in temperate and polar oceans. Nature 427, 445–8. doi:10.1038/nature02272

Seyedsayamdost, M.R., Carr, G., Kolter, R., Clardy, J., 2011. Roseobacticides: small molecule modulators of an algal-bacterial symbiosis. J. Am. Chem. Soc. 133, 18343–9. doi:10.1021/ja207172s

Shiba, T., 1991. Roseobacter litoralis gen. nov., sp. nov., and Roseobacter denitrificans sp. nov., Aerobic Pink-Pigmented Bacteria which Contain Bacteriochlorophyll a. Syst. Appl. Microbiol. 14, 140–145.

Shiba, T., Simidu, U., Taga, N., 1979. Distribution of Aerobic Bacteria Which Contain Bacteriochlorophyll a. App 38, 43–45.

Shields, R.J., Bell, J.G., Luizi, F.S., Gara, B., Bromage, N.R., Sargent, J.R., 1999. Natural copepods are superior to enriched artemia nauplii as feed for halibut larvae (Hippoglossus hippoglossus) in terms of survival, pigmentation and retinal morphology: relation to dietary essential fatty acids. J. Nutr. 129, 1186–94.

Skov, M.N., Pedersen, K., Larsen, J.L., 1995. Comparison of Pulsed-Field Gel Electrophoresis, Ribotyping, and Plasmid Profiling for Typing of Vibrio anguillarum Serovar O1. Appl. Environ. Microbiol. 61, 1540–5.

Soffientino, B., Gwaltney, T., Nelson, D.R., Specker, J.L., Mauel, M., Gómez-Chiarri, M., 1999. Infectious necrotizing enteritis and mortality caused by Vibrio carchariae in summer flounder Paralichthys dentatus during intensive culture. Dis. Aquat. Organ. 38, 201–210. doi:10.3354/dao038201

Sommerset, I., Krossøy, B., Biering, E., Frost, P., 2005. Vaccines for fish in aquaculture. Expert Rev. Vaccines 4, 89–101. doi:10.1586/14760584.4.1.89

Song, S.K., Beck, B.R., Kim, D., Park, J., Kim, J., Kim, H.D., Ringø, E., 2014. Prebiotics as immunostimulants in aquaculture: A review. Fish Shellfish Immunol. 40, 40–48. doi:10.1016/j.fsi.2014.06.016

Sonnenshein, E.C., Nielsen, K.F., D’Alvise, P., Porsby, C.H., Melchiorsen, J., Heilmann, J., Kalatzis, P.G., López-Pérez, M., Bunk, B., Spöer, C., Middelboe, M., Gram, L., 2016. Global occurrence and heterogenecity of the Roseobacter-clade species Ruegeria mobilis. ISME J.

Souissi, A., Souissi, S., Hansen, B.W., 2014. Physiological improvement in the copepod Eurytemora affinis through thermal and multi- generational selection. Aquac. Res. 1365– 2109. doi:10.1111/are.12675

Støttrup, J.G., 2003. Production and Nutritional Value of Copepods. Live Feed. Mar. Aquac. 145– 205. doi:10.1002/9780470995143.ch5

Støttrup, J.G., Richardson, K., Kirkegaard, E., Pihl, N.J., 1986. The cultivation of Acartia tonsa Dana for use as a live food source for marine fish larvae. Aquaculture 52, 87–96. doi:10.1016/0044-8486(86)90028-1

Syngai, G.G., Gopi, R., Bharali, R., Dey, S., Lakshmanan, G.M.A., Ahmed, G., 2016. Probiotics - the versatile functional food ingredients. J. Food Sci. Technol. 53, 921–933. doi:10.1007/s13197-015-2011-0

Tan, D., Gram, L., Middelboe, M., 2014. Vibriophages and their interactions with the fish pathogen vibrio anguillarum. Appl. Environ. Microbiol. 80, 3128–3140. doi:10.1128/AEM.03544-13

Taniguchi, A., Aoki, R., Eguchi, M., 2016. Microbial communities in various waters used for fish larval rearing. Aquac. Res. 47, 370–378. doi:10.1111/are.12495

Thépot, V., Mangott, A., Pirozzi, I., 2016. Rotifers enriched with a mixed algal diet promote survival, growth and development of barramundi larvae, Lates calcarifer (Bloch). Aquac. Reports 3, 147–158. doi:10.1016/j.aqrep.2016.02.003

Thiel, V., Brinkhoff, T., Dickschat, J.S., Wickel, S., Grunenberg, J., Wagner-Döbler, I., Simon, M., Schulz, S., 2010. Identification and biosynthesis of tropone derivatives and sulfur volatiles produced by bacteria of the marine Roseobacter clade. Org. Biomol. Chem. 8, 234–46. doi:10.1039/b909133e

Thole, S., Kalhoefer, D., Voget, S., Berger, M., Engelhardt, T., Liesegang, H., Wollherr, A., Kjelleberg, S., Daniel, R., Simon, M., Thomas, T., Brinkhoff, T., 2012. Phaeobacter gallaeciensis genomes from globally opposite locations reveal high similarity of adaptation to surface life. ISME J. 6, 2229–44. doi:10.1038/ismej.2012.62

Toi, H.T., Boeckx, P., Sorgeloos, P., Bossier, P., Van Stappen, G., 2013. Bacteria contribute to Artemia nutrition in algae-limited conditions: A laboratory study. Aquaculture 388-391, 1– 7. doi:10.1016/j.aquaculture.2013.01.005

Toranzo, A.E., Barja, J.L., 1990. A review of the taxonomy and seroepizootiology of Vibrio anguillarum, with special reference to aquaculture in the Northwest of Spain. Dis. Aquat. Organ. 9, 73–82. doi:10.3354/dao009073

Uzun, E., Ogut, H., 2015. The isolation frequency of bacterial pathogens from sea bass (Dicentrarchus labrax) in the Southeastern Black Sea. Aquaculture 437, 30–37. doi:10.1016/j.aquaculture.2014.11.017

Van Stappen, G., 1996. Introduction, biology and ecology of Artemia. In: Lavens, P., Sorgeloos, P. (Eds.), Manual on the production and use of live food for aquaculture. FAO Tech. Pap. 295. doi:10.1007/s13398-014-0173-7.2

Verdonck, L., Grisez, L., Sweetman, E., Minkoff, G., Sorgeloos, P., Ollevier, F., Swings, J., 1997. Vibrios associated with routine productions of Brachionus plicatilis. Aquaculture 149, 203–214. doi:10.1016/S0044-8486(96)01451-2

Verner-Jeffreys, D.W., Shields, R.J., Bricknell, I.R., Birkbeck, T.H., 2004. Effects of different water treatment methods and antibiotic addition on larval survival and gut microflora development in Atlantic halibut (Hippoglossus hippoglossus L.) yolk-sac larvae. 69

Aquaculture 232, 129–143. doi:10.1016/S0044-8486(03)00525-8

Verschuere, L., Rombaut, G., Huys, G., Dhont, J., Sorgeloos, P., Verstraete, W., 1999. Microbial Control of the Culture of Artemia Juveniles through Preemptive Colonization by Selected Bacterial Strains. Appl. Environ. Microbiol. 65, 2527–2533.

Verschuere, L., Rombaut, G., Sorgeloos, P., Verstraete, W., 2000. Probiotic Bacteria as Biological Control Agents in Aquaculture. Microbiol. Mol. Biol. Rev. 64, 655–671.

Vine, N.G., Leukes, W.D., Kaiser, H., Daya, S., Baxter, J., Hecht, T., 2004. Competition for attachment of aquaculture candidate probiotic and pathogenic bacteria on fish intestinal mucus. J. Fish Dis. 27, 319–326. doi:10.1111/j.1365-2761.2004.00542.x

Wilson, M.Z., Wang, R., Gitai, Z., Seyedsayamdost, M.R., 2015. Mode of action and resistance studies unveil new roles for tropodithietic acid as an anticancer agent and the γ-glutamyl cycle as a proton sink. Proc. Natl. Acad. Sci. 113, 201518034. doi:10.1073/pnas.1518034113

Yan, L., Boyd, K.G., Grant Burgess, J., 2002. Surface attachment induced production of antimicrobial compounds by marine epiphytic bacteria using modified roller bottle cultivation. Mar. Biotechnol. 4, 356–366. doi:10.1007/s10126-002-0041-x

Zhang, X.-H., Austin, B., 2000. Pathogenicity of Vibrio harveyi to salmonids. J. Fish Dis. 23, 93– 102. doi:10.1046/j.1365-2761.2000.00214.x

Zhang, X.X., Zhang, T., Fang, H.H.P., 2009. Antibiotic resistance genes in water environment. Appl. Microbiol. Biotechnol. 82, 397–414. doi:10.1007/s00253-008-1829-z

Zhao, W., Dao, C., Karim, M., Gomez-Chiarri, M., Rowley, D., Nelson, D.R., 2016. Contributions of tropodithietic acid and biofilm formation to the probiotic activity of Phaeobacter inhibens. BMC Microbiol. 16, 1. doi:10.1186/s12866-015-0617-z

Øie, G., Makridis, P., Reitan, K.I., Olsen, Y., 1997. Protein and carbon utilization of rotifers (Brachionus plicatilis) in first feeding of turbot larvae (Scophthalmus maximus L.). Aquaculture 153, 103–122.

Paper 1

Isolation of TDA-producing Phaeobacter strains from sea bass larval rearing units and their probiotic effect against pathogenic Vibrio spp . in Artemia cultures

Systematic and Applied Microbiology 39 (2016) 180–188

Contents lists available at ScienceDirect

Systematic and Applied Microbiology

j ournal homepage: www.elsevier.de/syapm

Isolation of TDA-producing Phaeobacter strains from sea bass larval

rearing units and their probiotic effect against pathogenic Vibrio spp. in Artemia cultures

a,1 a,1 a,1,2 b

Torben Grotkjær , Mikkel Bentzon-Tilia , Paul D’Alvise , Nancy Dourala ,

c a,∗

Kristian Fog Nielsen , Lone Gram

Department of Systems Biology, Technical University of Denmark, Matematiktorvet Bldg. 301, DK-2800 Kgs. Lyngby, Denmark

Selonda SA, 30 Navarhou, Nikodemou Street, 10555 Athens, Greece

Department of Systems Biology, Technical University of Denmark, Søltofts Plads, Bldg 221, DK-2800 Kgs. Lyngby, Denmark

a r t i c l e i n f o a b s t r a c t

Article history: Fish-pathogenic Vibrio can cause large-scale crashes in marine larval rearing units and, since the use of

Received 14 October 2015

antibiotics can result in bacterial antibiotic resistance, new strategies for disease prevention are needed.

Received in revised form 28 January 2016

Roseobacter-clade bacteria from turbot larval rearing facilities can antagonize Vibrio anguillarum and

Accepted 29 January 2016

reduce mortality in V. anguillarum-infected cod and turbot larvae. In this study, it was demonstrated

that antagonistic Roseobacter-clade bacteria could be isolated from sea bass larval rearing units. In addi-

Keywords:

tion, it was shown that they not only antagonized V. anguillarum but also V. harveyi, which is the major

Artemia

bacterial pathogen in crustaceans and Mediterranean sea bass larvae cultures. Concomitantly, they sig-

Fish probiotics

Phaeobacter niﬁcantly improved survival of V. harveyi-infected brine shrimp. 16S rRNA gene sequence homology

Roseobacter identiﬁed the antagonists as Phaeobacter sp., and in silico DNA-DNA hybridization indicated that they

Vibrio anguillarum could belong to a new species. The genomes contained genes involved in synthesis of the antibacte-

Vibrio harveyi rial compound tropodithietic acid (TDA), and its production was conﬁrmed by UHPLC-TOFMS. The new

Phaeobacter colonized live feed (Artemia) cultures and reduced Vibrio counts signiﬁcantly, since they

4 −1 8 −1

reached only 10 CFU mL , as opposed to 10 CFU mL in non-Phaeobacter treated controls. Survival of

V. anguillarum-challenged Artemia nauplii was enhanced by the presence of wild type Phaeobacter com-

pared to challenged control cultures (89 ± 1.0% vs 8 ± 3.2%). In conclusion, TDA-producing Phaeobacter

isolated from Mediterranean marine larviculture are promising probiotic bacteria against pathogenic

Vibrio in crustacean live-feed cultures for marine ﬁsh larvae.

Introduction larviculture via live feed stock cultures, supply water, humans, or

brood stock [24]. Fish larvae requiring live feed are mostly fed

Aquaculture is the fastest growing agricultural industry, provid- rotifers (Brachionus plicatilis), Artemia and often also live microal-

ing high-value protein-rich food for the growing world population. gae. Due to the high concentration of organic matter in Artemia

Currently, 50% of ﬁsh consumed are reared in aquaculture [26] cultures [52], bacteria can grow to high levels and may include

and the culture of high value marine fish, crustacean and mollusk opportunistic fish pathogens, such as the prominent fish and shell-

species is ﬁnancially attractive. However, a major problem in aqua- ﬁsh pathogens Vibrio anguillarum and V. harveyi [3,29,70]. High

culture is represented by disease outbreaks caused by opportunistic loads of pathogenic bacteria in live feed will eventually lead to

pathogenic bacteria at the larval stage [6], where vaccination is infections that either spread rapidly and cause crashes of the larval

not applicable. Pathogenic bacteria can be introduced into marine population or induce a slow but steady mortality rate [63]. Such dis-

ease outbreaks constitute a major economic bottleneck in marine

larviculture.

Bacterial diseases in ﬁsh larvae may be controlled by using

∗

Corresponding author. Tel.: +45 45252586.

antibiotics, either prophylactically or as acute treatment. However,

E-mail address: [email protected] (L. Gram).

1 since continuous use of antibiotics can result in development and

Shared ﬁrst authorship.

2 spread of bacterial antibiotic resistance [13], new strategies for

Current address: University of Hohenheim, Institute for Animal Science, Popu-

lation Genomics Group, Garbenstr. 17, Room 008, 70599 Stuttgart, Germany. disease prevention are needed. Alternative treatment strategies

http://dx.doi.org/10.1016/j.syapm.2016.01.005

T. Grotkjær et al. / Systematic and Applied Microbiology 39 (2016) 180–188 181

−1 −1

include immunostimulants [7], use of bacterial quorum sensing France), 4 g L glucose, and 10 g L agar) in which V. anguillarum

inhibitors [20] or the use of probiotics, which are live microorgan- serotype O1 strain 90-11-287 [67] was embedded. A total of 120 ␮L

isms that exert a beneﬁcial effect on the health of a host [25]. Many of a V. anguillarum overnight culture, grown in marine broth (MB,

◦

studies have addressed the potential use of probiotics in aquacul- Difco 2216) at 25 C, was mixed with 120 mL of molten, cooled

◦

ture [22,72], however, a reproducible reduction of mortalities has (41.5 C) agar before pouring the plates. The replica plates were

◦

only occasionally been achieved [49]. Bacteria belonging to the incubated for 24 h at 25 C. Colonies causing clearing zones in the V.

marine Roseobacter clade have been isolated and/or detected in anguillarum-seeded agar were isolated from the original MA plate.

◦

marine larviculture systems [38,42,55], and Phaeobacter inhibens, All isolated strains were stored at −80 C in a freeze medium [33].

P. gallaeciensis and Ruegeria mobilis have been isolated due to The strains were re-tested for antibacterial activity, using Phaeobac-

their antagonism against pathogenic V. anguillarum. This antago- ter inhibens strain DSM 17395 as a positive control. The potential

nism is partially due to production of the antibacterial compound production of antibacterial compounds in liquid cultures was tested

tropodithietic acid (TDA) [17]. Interestingly, while TDA has a broad by a diffusion-inhibition assay: cultures grown for 2 days in MB

inhibitory spectrum, resistance or tolerance in target pathogenic were sterile-ﬁltered (0.2 ␮m) and 50 ␮L were added to 5 mm wells

bacteria have not yet been detected and cannot be easily pro- punched in V. anguillarum-seeded agar plates that were prepared

voked, as would be the case for most other antibacterial agents [56]. as for replica plating [38]. Plates were inspected for clearing zones

◦

Addition of P. inhibens to turbot or cod larvae infected with V. anguil- after 1 and 2 days incubation at 25 C.

larum has been shown to reduce mortality signiﬁcantly, compared

to infected larvae without P. inhibens addition [18,17,38].

Characterization of the antagonistic isolates

Since introducing probiotic bacteria directly into larvae basins

is challenging because of the large water volumes, and the fact that

Biochemical tests were used to tentatively identify the isolates

high pathogen concentrations are potentially already developing

that had antimicrobial activity. The Gram reaction was tested using

in the live feed cultures, we have suggested previously that the

3% KOH, catalase by exposure to 3% H2O2, and cytochrome oxi-

probiotics should be introduced already at the algae and rotifer TM TM

dase using BBL Dryslide Oxidase (Becton Dickson) on cultures

culture stage where they are used as live feed [17]. Therefore, high ◦

grown for 1 day on MA at 25 C. The ability to ferment or oxidize

pathogen concentrations in the live feed could be prevented and the

glucose was tested in OF basal medium (Merck 1.10282.0500) [39]

probiotics would be present in all trophic stages of the hatchery. ®

supplemented with 2% Instant Ocean sea salts. Shape, motility,

One purpose of the present study was to determine if pathogen

and the Phaeobacter indicative ability to form rosettes [10] were

reduction by Phaeobacter spp. would be effective in cultures of brine

examined by phase-contrast microscopy (1000-fold magniﬁcation

shrimp (Artemia), which are widely used as live feed for ﬁsh larvae ◦

using an Olympus BH2) of cultures grown in MB for 3 days at 25 C

and where high levels of organic material allow rapid growth of

under static conditions. 16S rRNA gene amplicons from a subset

potentially pathogenic Vibrio spp., especially V. harveyi.

of 13 putative Roseobacter-clade isolates were sequenced at GATC

Studies on Phaeobacter as a ﬁsh larval probiont have been carried

Biotech GA (Köln, Germany) and their phylogenetic afﬁliations

out in turbot and cod cultures infected with V. anguillarum. A range

were determined by a BLASTN search in the NCBI nucleotide col-

of other marine ﬁsh species, such as sea bream and sea bass, are

lection and through analyses of the V4-V5 region, as described by

reared under similar conditions using live feed, however, it is not

Buddruhs et al. [11]. The 16S rRNA gene sequences were deposited

known if roseobacters are part of the microbiota in this type of larval

at the NCBI under accession numbers KT884052–KT884064.

rearing. In sea bass and sea bream cultures, the temperature is often

Based on survival of Artemia challenged with V. harveyi and

higher than in turbot and cod cultures, and V. harveyi is often the

treated with the potential probionts, two isolates were chosen

disease-causing pathogen [1]. Although the ﬁsh species are reared

for further studies (Phaeobacter sp. S26 and S60; see below). To

under different temperatures and water chemistry conditions, the

investigate antagonism against V. anguillarum and V. harveyi at dif-

live feed cultures (Artemia, algae, rotifers) are grown under similar

ferent temperatures, the two strains were grown without shaking

conditions, which would allow cross-species development of feed ◦ ◦ ◦ ◦

in MB at 5 C for 168 h, 10 C for 120 h, 15 C for 96 h, 20 C for

probiotics. Therefore, a second purpose of the present study was to ◦ ◦

72 h, 25 C for 72 h and 30 C for 72 h together with seven reference

isolate potential probiotic bacteria from sea bass larval rearing units T

strains (Phaeobacter inhibens DSM 16374 , Phaeobacter inhibens

and evaluate if they would be effective against V. harveyi, which is T

DSM 17395, Phaeobacter gallaeciensis DSM 26640 , Phaeobacter sp.

the dominant pathogen of crustaceans, as well as sea bass and sea

27-4, Ruegeria mobilis F1926, Pseudovibrio sp. FO-BEG1 and Pseu-

bream larvae. T

dophaeobacter arcticus DSM 23566 ). Supernatants of these cultures

were obtained by centrifugation (10,000 × g, 5 min) and 40 ␮L were

tested in well-diffusion assays against the target organisms embed-

Materials and methods ®

ded in Instant Ocean agar plates. All assays were performed in

biological duplicates.

Isolation of culturable marine bacteria and testing for their

antimicrobial activity

Detection of tropodithietic acid (TDA)

Samples for microbial analyses were taken from two Greek sea

◦ ◦

bass aquaculture units in July 2013 (38 33 48 N 23 36 16.5 E). Marine broth culture samples of the 13 Phaeobacter sp. strains

Swab samples were collected from tank walls and outlets, and isolated from the Greek aquaculture unit and seven reference

water samples were taken from the outlet of the ﬁsh tanks and strains (Phaeobacter inhibens DSM 16374 , Phaeobacter inhibens

Artemia basins, as well as phytoplankton and rotifer cultures. Ster- DSM 17395, Phaeobacter gallaeciensis DSM 26640 , Phaeobacter sp.

ile gloves, swabs and tubes were used for sampling. All samples 27-4, Ruegeria mobilis F1926, Pseudovibrio sp. FO-BEG1 and Pseu-

were 10-fold serially diluted in sterile artiﬁcial sea water and dophaeobacter arcticus DSM 23566 ) were extracted with ethyl

plated onto marine agar (MA, Difco 2216). Plates were incubated acetate (HPLC grade) containing 1% formic acid (HPLC grade). The

◦ ◦

for 5 days at 25 C and sent by courier to Denmark. As previously organic phase was evaporated to dryness at 35 C with nitrogen

described [38], the MA plates were replica plated onto Instant ﬂow. The samples were re-dissolved in 85% acetonitrile, and liq-

® −1 ®

Ocean agar plates (30 g L Instant Ocean sea salts (Aquarium uid chromatographic-high resolution (UHPLC-TOFMS) analysis was

−1

Systems Inc., Sarrebourg, France), 3.33 g L casamino acids (Bacto, conducted in an Agilent 1290 UHPLC coupled to an Agilent 6550

182 T. Grotkjær et al. / Systematic and Applied Microbiology 39 (2016) 180–188

qTOF equipped with a dual electrospray source [40]. Separation as well as against the genomes of the three Phaeobacter reference

◦ T

was performed at 40 C in a 2.1 mm ID, 50 mm, 1.8 ␮m Eclipse Plus strains (P. inhibens DSM 17395, P. inhibens DSM 16374 , and P. gal-

C18 (Agilent) column using a water-acetonitrile gradient solvent laeciensis DSM 26640 ) using the CLC Main Workbench. Replicase

system, with both water and acetonitrile containing 20 mM formic and active partitioning genes of replicons that did not have any

−1

acid. Using a ﬂow of 0.8 mL min , the gradient was started at 10% apparent homolog in any of the reference strains were used as

acetonitrile and increased to 60% acetonitrile within 1.8 min, then query sequences in BLASTX-searches in the NCBI non-redundant

to 100% in 0.2 min, which was maintained for 0.8 min, returning to protein database. Contigs harboring replicases exclusively found in

15% acetonitrile in 0.2 min and equilibrating for 1.5 min. TDA was one of the isolates were re-annotated using Rapid Annotation with

+ +

determined in ESI mode from its [M + H] ion m/z 212.9674 ± 0.005 Subsystem Technology (RAST; [5]).

with the same retentions time (1.10 min) as the authentic standard

(BioViotica, Dransfeld, Germany).

Bacterial strains for Artemia co-culture assays

Genome sequencing and analyses

Two pathogenic Vibrio strains were included in the Artemia

experiments. V. anguillarum strain NB10 was isolated from the Gulf

Genomic DNA was obtained from Phaeobacter sp. strains S26 and

of Bothnia and has caused disease in rainbow trout [50,51]. The

S60 by successive phenol-chloroform puriﬁcation steps. Illumina

strain has been tagged by insertion of plasmid pNQFlaC4-gfp27 (cat,

MiSeq sequencing was carried out by the Novo Nordisk Center for

gfp) into an intergenic region on the chromosome, and was kindly

Biosustainability (DTU Biosustain, Hørsholm, Denmark). De novo

provided by D. Milton, University of Umeå [14]. V. harveyi strain

assembly was carried out using a CLC Genomic Workbench (CLC

VIB391 was isolated from aquacultured diseased shrimp [75]. In

Bio, Aarhus, Denmark), and the genomes were annotated using

an initial screening for potential probiotic effects, 13 Phaeobacter

the NCBI Prokaryotic Genome Annotation Pipeline (PGAP). The CLC

spp. were tested for their ability to reduce Vibrio counts in Artemia

Genomic Workbench was also used to BLAST-search the genomes

co-cultures and to enhance Artemia survival. The 13 strains were

of S26 and S60 for homologs of the P. inhibens DSM 17395 TDA

ranked by their ability to reduce mortality of Artemia after 3 days in

biosynthesis genes (tdaA-tdaH). 16S rRNA genes were retrieved

co-culture with V. harveyi strain VIB391 and the respective isolate.

from the genomes and the strains were identiﬁed to genus level

Based on the preliminary tests, strains S26 and S60 were chosen for

using the BLAST-based program LeBiBi [23] and the NCBI database.

further studies. Their potential probiotic effect was compared to P.

The V4 and V5 regions of the two isolates P. inhibens DSM 17395

T inhibens strain DSM 17395 [65] and a Tn5 mutant strain of DSM

and P. gallaeciensis DSM 26640 were identiﬁed using the CLC Main

17395, which carries a miniTn5-insertion in the tdaB gene and is

Workbench, according to Buddruhs et al. [11]. Genomes were sub-

unable to produce TDA [58].

mitted to the NCBI under accession numbers JSWJ00000000 and

JSWK00000000.

The average nucleotide identity (ANI) of Phaeobacter sp. S26, Artemia co-culture challenge

Phaeobacter sp. S60, P. inhibens DSM 16374 , P. inhibens DSM 17395,

P. gallaeciensis DSM 26640 , and Pseudophaeobacter arcticus DSM Combinations of the two ﬁsh pathogens and the potential

23566 was determined using a BLAST+-based approach (ANIb) probionts were introduced into axenic Artemia cultures. Axenic

through JSpeciesWS [64] using a signiﬁcance cutoff of 95% identity. Artemia cultures and Artemia cultures with only the pathogens were

The corresponding heatmap was created using CIMminer [74]. used as controls. For each culture, 20 mg Artemia cysts were disin-

In silico DNA-DNA hybridization was carried out using the fected by adding 1.5 mL 0.5% sodium hypochlorite and vortexing for

Genome-to-Genome Distance Calculator (GGDC 2.0, http://ggdc. 3 min [68]. The cysts were thereafter immediately separated from

dsmz.de/) [2] for comparing Phaeobacter sp. S26 and S60 to the the disinfection solution by ﬁltration and rinsed repeatedly with

strains of P. inhibens (DSM 17395, DSM 16374 ), P. gallaeciensis sterile water. The ﬁlter with cysts was transferred to 50 mL Falcon

T T

(DSM 26640 ) and Pseudophaeobacter arcticus (DSM 23566 ). GGDC tubes containing 20 mL artiﬁcial seawater (3% Instant Ocean sea

2.0 deploys three formulae for estimating DNA-DNA-hybridization salts; Aquarium Systems Inc., Sarrebourg, France). P. inhibens and

values, of which formula 2 must be applied for incomplete genome Vibrio strains were grown in ½ strength yeast-tryptone-sea salts

◦

sequences [2]. medium (YTSS) at 25 C with aeration (200 rpm) for 24 h. V. anguil-

Pan- and core genomes of Phaeobacter sp. S26, Phaeobacter sp. larum and V. harveyi pre-cultures were adjusted to OD600 = 0.5 by

S60, P. inhibens DSM 16374 , P. inhibens DSM 17395, and P. gal- diluting with ½ strength YTSS, and 10-fold serially diluted in artiﬁ-

T ␮

laeciensis DSM 26640 were analyzed using CMG-Biotools v.2.2 cial seawater (3% Instant Ocean sea salts). A total of 300 L of the

[73]. The existing genome annotations were of varying quality and appropriate Vibrio dilutions were added to the Artemia cultures, in

3 −1

hence gene ﬁnding was performed using prodigalrunner for all the order to provide 10 CFU mL as the initial concentration (veri-

raw genome sequences. Amino acid sequences of the predicted ﬁed by serial dilution and plating and found to be in the range of

2 2

× × ␮

open reading frames were used to deﬁne protein families (50/50 3 10 to 9 10 ). A total of 100 L of an overnight Phaeobacter cul-

6 −1

cutoffs as described by Vesth et al. [73]) encoded by the respec- ture was added to the Artemia cultures, giving 10 CFU mL as the

␮

tive genomes. To assess the phylogenetic relationship between initial concentration. A total of 100 L ½ strength YTSS was added

Phaeobacter sp. S26, Phaeobacter sp. S60, and the reference strains, to the axenic Artemia control cultures and to cultures with only

◦

a pan-genomic dendrogram was constructed. Using the pancore- Vibrio. The cultures were incubated at 25 C, lying ﬂat on a rotary

plot tree function, a distance matrix (Manhattan distances), based shaker at 200 rpm. The number of dead Artemia was determined

on shared protein families between the strains, was constructed daily using a Sedgewick-Rafter counting chamber. The number of

and these distances were used to build the dendrogram [73]). The surviving Artemia was determined at the end of the experiment.

speciﬁc protein families exclusively recovered from Phaeobacter sp. Bacterial concentrations were determined daily by plating 10-fold

S26 and Phaeobacter sp. S60 were identiﬁed using the pancore- serial dilutions. Half strength MA (27.6 g marine agar Difco 2216,

plot subsets function. A local Pfam domain search was carried out 15 g Instant Ocean sea salts, 7.5 g agar, 1 L deionized water) plates

on these protein families using the CLC Main Workbench with a were used for total bacterial counts, and tryptone soy agar (TSA,

−

signiﬁcance cutoff value of E ≤ 10 . Oxoid CM0131) plates were used for speciﬁc counts of the two

Plasmid replicons were extracted from the two draft genomes Vibrio species, since Phaeobacter does not grow on TSA. All com-

and individual replicases were blasted against the other genome binations were carried out with biological replicates.

T. Grotkjær et al. / Systematic and Applied Microbiology 39 (2016) 180–188 183

Table 1

The inﬂuence of 13 Phaeobacter strains isolated from sea bass larvae units on Vibrio harveyi-caused mortality in Artemia cultures and TDA concentration after 72 h static

◦ T T

growth at 25 C. Reference strains Phaeobacter inhibens DSM 16374 , Phaeobacter inhibens DSM 17395, Phaeobacter gallaeciensis DSM 26640 , Phaeobacter sp. 27-4, Ruegeria

mobilis F1926, Pseudovibrio sp. FO-BEG1 and Pseudophaeobacter arcticus DSM 23566 were included for TDA concentration comparability. Numbers are averages of two

independent replicates.

Hatchery Strain Dead Artemia (average count ± s.d.) TDA concentration (average ␮M ± s.d.)

1 20 334 ± 47 0.47 ± 0.03

1 21 339 ± 21 0.53 ± 0.01

1 22 248 ± 25 0.63 ± 0.04

1 24 157 ± 34 0.71 ± 0.07

1 26 122 ± 13 0.92 ± 0.07

1 27 367 ± 48 0.77 ± 0.08

1 28 477 ± 33 0.86 ± 0.07

1 29 404 ± 35 0.57 ± 0.02

1 30 228 ± 17 0.49 ± 0.03

2 51 330 ± 15 1.32 ± 0.03

2 53 496 ± 145 0.31 ± 0.01

2 59 469 ± 287 0.31 ± 0.03

2 60 167 ± 16 0.95 ± 0.08

Positive control (V. harveyi VIB293) 422 ± 504 ND

Negative control (non-infected) 32 ± 9ND

DSM 16374 ND 0.74 ± 0.11

DSM 17395 ND 0.54 ± 0.15

DSM 26640 ND 0.34 ± 0.14

F1926 ND 0.10 ± 0.11

FO-BEG1 ND 0.24 ± 0.01

27-4 ND 0.12 ± 0.01

DSM 23566 ND 0 ± 0

High deviation because of difﬁculty in counting at high mortality levels due to digestion caused by the chitinolytic enzymes produced by Vibrio harveyi.

ND: not determined.

Statistical analysis members of the Vibrionaceae family. Such indigenous vibrios are

often pathogenic to larvae [38], and therefore these strains were

Bacterial counts were log-transformed, and differences were not pursued further. Nineteen strains did not metabolize glucose,

tested for signiﬁcance using ANOVA in GraphPad Prism 5.00 and 13 of the 19 strains produced a brown pigment when grown

(GraphPad Software, San Diego CA). Tukey’s multiple comparison on MA or in MB. These strains also formed rosettes during static

test was used for pairwise comparisons. Values from day 0, which growth in MB. Collectively, the phenotypic analyses indicated that

were merely a proof of the inoculated concentrations, were omitted the 13 strains could be TDA-producing Roseobacter-clade bacteria

in the analysis. The end-point Artemia mortalities were compared (17, 37). 16S rRNA gene sequence analyses conﬁrmed this afﬁlia-

using ANOVA. tion and placed the isolates within the Phaeobacter genus. All of

these 13 strains were isolated from two bioﬁlm samples (9 strains

Results from one hatchery, 4 strains from the other) taken from the out-

let of the ﬁsh tanks. Liquid chromatography with high resolution

Isolation of culturable marine bacteria and screening for mass spectrometry conﬁrmed the presence of TDA in all 13 strains

antimicrobial activity (Table 1).

The strains were tested in a preliminary experiment for their

Samples were taken from two sea bass hatcheries within the ability to reduce mortality caused by V. harveyi in Artemia cultures.

same larviculture facility. Water samples were taken from the inlet, Eight of the 13 strains (61%) improved survival of the Artemia con-

ﬁsh tanks, Artemia cultures, phytoplankton cultures and rotifer cul- siderably, as compared to controls with only V. harveyi. Two strains,

tures. Bioﬁlm swab samples were taken from the ﬁsh tank outlets S26 and S60, were chosen for further studies based on the highest

and the ﬁsh tank walls. A total of 3100 colonies, on 43 plates from 11 survival of Artemia and their origin, as representatives of each of

samples, were replica plated on V. anguillarum-seeded agar plates, the two hatcheries sampled (Table 1). Furthermore, these isolates

and 96 colonies (3.1%) caused clearing zones in the Vibrio agar. The were among the best TDA producers, producing more TDA than any

colonies with antagonistic activity in the replica assay were iso- of the reference strains.

lated from the original MA plates, pure-cultured and re-tested for The inhibitory activity of the two strains and seven refer-

antagonistic activity against V. anguillarum. Of these 96 isolates, 74 ence strains grown at different temperatures was tested against V.

◦

retained antagonistic activity upon re-testing. Antagonistic bacte- anguillarum and V. harveyi. When grown at 5 C, no inhibitory effect

ria were present in all samples but, of the 74 antagonistic isolates, was observed against V. anguillarum for S26 and S60 (Table S1).

62% were isolated from the bioﬁlm samples. The 74 isolates were Similarly, no inhibition was observed for the four reference strains

ranked into three categories according to the size of the clearing Phaeobacter sp. 27-4, Ruegeria mobilis F1926, Pseudophaeobacter

zone (< 2 mm, < 5 mm, and > 5 mm) compared to the clearing zone arcticus DSM 23566 , and Pseudovibrio sp. FO-BEG1. In contrast, P.

of P. inhibens DSM 17395, which was used as a control. inhibens DSM 17395, P. inhibens DSM 16374 , and P. gallaeciensis

T ◦

DSM 26640 all exhibited inhibitory activity at 5 C. The same

Phenotypic characterization of antagonistic bacteria and results were obtained in inhibition assays against V. harveyi, except

sub-selection of isolates for S26 that displayed a small inhibition zone of 0.4 mm in one of

◦

the two duplicates (Table S2). At higher temperatures (10–30 C),

All of the 74 antagonistic strains were Gram-negative bacteria inhibitory activity was observed for all tested strains except

with positive catalase and oxidase reactions. Of these, 55 metabo- Pseudovibrio sp. FO-BEG1 (and the outgroup Pseudophaeobacter

lized glucose both fermentatively and oxidatively, and were likely arcticus DSM 23566 , which never exhibited any inhibition zones)

184 T. Grotkjær et al. / Systematic and Applied Microbiology 39 (2016) 180–188

against both pathogens. In general, the size of the zones increased

◦

in diameter with increasing growth temperature up to 25 C after

which it decreased slightly (Tables S1 and S2).

Molecular characterization of the isolated strains

The sequenced genomes of strains S26 and S60 were de novo

assembled into 60 and 55 contigs with a total genome size (includ-

ing plasmids) of 4.2 and 4.6 Mb. The coverage was 109 and 83 fold,

respectively. The 16S rRNA genes were 99.7% identical to both P.

inhibens and P. gallaeciensis. These two species are differentiated in

the V4 and V5 region of their 16S rRNA genes [11]. To determine

whether Phaeobacter sp. strains S26 and S60 exhibited V4 and V5

regions identical to P. inhibens or P. gallaeciensis they were com-

pared to P. inhibens (DSM 17395) and P. gallaeciensis (DSM 26640 ).

The differences were localized at base positions 614 and 626 in

the V4 region and at positions 835 and 851 in the V5 region (Fig.

S1). Both strains S26 and S60 had V4 regions corresponding to P.

gallaeciensis and V5 regions corresponding to P. inhibens.

ANIb comparison suggested that Phaeobacter sp. S26 and

Phaeobacter sp. S60 represented a single species with ANIb values

of 98% identity. Both strains exhibited 91% identity with P. gallae-

ciensis DSM 26640 , 90% identity with the two P. inhibens reference

strains (DSM 16374 and DSM 17395), and 74% identity with the

outgroup strain Pseudophaeobacter arcticus DSM 23566 (Fig. 1A).

Similarly, strains S26 and S60 were compared to P. inhibens (DSM

T T

17395 and DSM 16374 ), P. gallaeciensis (DSM 26640 ) and P. arcti-

cus (DSM 23566 ) by in silico DNA-DNA hybridization using the

web-page software tool GGDC 2.0 (http://ggdc.dsmz.de/). Based

on the appropriate formula 2, strains S26 and S60 had 85% iden-

tity to each other, but only 40% to P. inhibens DSM 17395 and P.

T T

inhibens DSM 16374 , and 44% to P. gallaeciensis DSM 26640 . Pseu-

dophaeobacter arcticus DSM 23566 (97.4% sequence similarity in

the 16S rRNA gene to strains S26 and S60) was included in the

analysis and had hybridization values of 20%, conﬁrming it as an

outgroup (Table S3).

The pan-genome of the ﬁve Phaeobacter strains represented

a total of 6056 protein families, of which 172 were exclusively

recovered from the S26 and S60 strains. Within these 172 pro-

tein families, 45 Pfam domains could be identiﬁed at the applied

−3

cutoff (E ≤ 10 ). These domains were mainly related to carbohy-

Fig. 1. (A) BLAST+ based average nucleotide identity analysis (ANIb) of the strains

drate and amino acid transportation, and transcriptional regulation T

Phaeobacter sp. S26 and S60, and Phaeobacter inhibens DSM 16374 , Phaeobacter

(Table S4). The phylogenetic analysis based on shared protein fam- inhibens DSM 17395, Phaeobacter gallaeciensis DSM 26640 and Pseudophaeobacter

ilies supported the whole-genome comparison data and placed arcticus DSM 23566 . The scale bar represents the percentages as quintiles. (B) A

pan-genomic dendrogram showing the phylogeny of the strains based on shared

Phaeobacter sp. S26 and Phaeobacter sp. S60 in a distinct sub-cluster

protein families.

(Fig. 1B).

Three plasmid-associated replicases were recovered from the

Phaeobacter sp. S26 draft genome. These were identical at the amino

acid (AA) level to three of seven plasmid-associated replicases (accession no. CP000831) and pDSHI04 (accession no. CP000834),

detected in the Phaeobacter sp. S60 genome. These three replicases sharing 93–94% AA sequence identity in the RepC protein and

exhibited a high degree of AA sequence similarity (96–99%) to the 98–99% AA sequence identity in the active partitioning genes. The

replicases of the three plasmids harbored by P. inhibens DSM 17395 third replicon that was not homologous to any of the reference

(Table S5), as well as to homolog plasmids in P. inhibens DSM 16374 strain replicons was most closely related to replication-related pro-

(96–98%). The highest degree of sequence similarity was, however, teins in Actibacterium atlanticum (93% RepC AA sequence similarity,

observed with the P. gallaeciensis DSM 26640 homologs (99–100%) accession no. WP 035252774) and Phaeobacter daeponensis (86%

for all three plasmids (Table S5). The four plasmid replicases from RepB AA sequence similarity, accession no. WP 027246247; 96%

Phaeobacter sp. S60, for which a homolog was not recovered in RepA AA sequence similarity, accession no. WP 036761612). Thus,

Phaeobacter sp. S26, were identical to the RepABC-type replicase the isolated strains displayed different plasmid proﬁles. Besides

from the P. gallaeciensis DSM 26640 plasmid pGalB134 (Table plasmid household regions, including Type IV secretion systems

S5). The remaining three RepABC type plasmid replicases that (T4SS/MOB regions) and stability systems, genomic features were

were observed exclusively in Phaeobacter sp. S60 did not have any related to several of the RAST pre-deﬁned subsystems, most of

apparent homologs in any of the three reference strains. BLASTX which were related to metabolism and regulation. In addition,

searches of the replicase and partitioning protein sequences in approximately 40 kb of the extended plasmid proﬁle of strain

the NCBI non-redundant protein database revealed that two of S60 was related to heavy metal resistance, and two plasmid con-

these replicons were most closely related to the replicons of tigs harbored genes involved in synthesis of bioactive secondary

the two Dinoroseobacter shibae DSM 16493 plasmids pDSHI01 metabolites (indigoidine and phenazine).

T. Grotkjær et al. / Systematic and Applied Microbiology 39 (2016) 180–188 185

Table 2

Survival of Artemia after four days [%]. Artemia were exposed to V. anguillarum or

V. harveyi and co-inoculated with either of four Phaeobacter strains: DSM 17395, its

TDA-negative mutant (MJG-G6) and two strains isolated from sea bass larval units:

S26 and S60. Numbers are averages of two independent trials.

Treatment Artemia survival (%)

V. anguillarum V. harveyi

Axenic control 54 ± 0

Pathogen only 8 ± 3.2 4 ± 1.1

Pathogen + P. inhibens DSM 17395 89 ± 1.01 74 ± 13.6

Pathogen + P. inhibens MJG-G6 45 ± 12.2 12 ± 3.4

Pathogen + Phaeobacter S26 77 ± 11.4 74 ± 10.2

Pathogen + Phaeobacter S60 55 ± 3.8 73 ± 5.1

counts were signiﬁcant in all Artemia cultures, compared to the

non-treated controls and the TDA-negative P. inhibens mutant

MJG-G6 (p < 0.001). The presence of MJG-G6 did reduce the growth

of V. anguillarum but not as markedly as the wild type. The maxi-

mum cell density of V. anguillarum was 1–2 log units lower in the

presence of MJG-G6 compared to the Vibrio control culture and,

although the reduction was not as pronounced as for the wild type,

it was statistically signiﬁcant (p = 0.02). In some of the Artemia

cultures challenged with V. anguillarum the pathogen increased in

abundance on the ﬁnal day (Fig. 2A).

The numbers of V. harveyi decreased signiﬁcantly in the pres-

ence of P. inhibens DSM 17395, Phaeobacter sp. S26, and S60

(p < 0.001). The reduction was in the order of two log units for

P. inhibens DSM 17395 and three log units for S26 and S60, as

Fig. 2. Reduction of (A) V. anguillarum and (B) V. harveyi in co-culture with Artemia

by Phaeobacter spp. Colony-forming units of V. anguillarum (A) and V. harveyi (B) compared to the Artemia controls with only V. harveyi. The TDA-

3 −1

inoculated at 10 CFU mL alone (), in the presence of P. inhibens DSM 17395 (᭿),

negative mutant did not signiﬁcantly reduce the numbers of V.

in the presence of the P. inhibens DSM 17395 TDA-negative mutant, MJG-G6 (ᮀ), in

harveyi (Fig. 2B).

the presence of Phaeobacter sp. S26 (), in the presence of Phaeobacter sp. S60 (),

The Artemia nauplii survived signiﬁcantly (p < 0.001) better in

and in the untreated controls (). Points represent the average of two biological

replicates and the error bars are the standard deviations of the means. cultures with the presence of P. inhibens DSM 17395, and Phaeobac-

ter sp. S26 and S60 compared to the cultures with no Phaeobacter

addition or to which the TDA-negative Phaeobacter mutant was

The brown pigmentation of S26 and S60 colonies on marine agar

added. In non-challenged, non-Phaeobacter treated Artemia the

indicated production of tropodithietic acid, and this was conﬁrmed

survival was 54 ± 0%. Challenge with V. anguillarum or V. harveyi

by UHPLC-TOFMS. Homologs of the tdaA to tdaH gene cluster of P.

caused a signiﬁcant (p < 0.001) decrease in survival to 8 and 4%,

inhibens DSM 17395 were found in both S26 and S60 on contig 2

respectively. Addition of DSM 17395 or S26 signiﬁcantly improved

and contig 28, respectively, and hence the TDA-biosynthesis gene

Artemia survival in the V. anguillarum-challenged Artemia to 89%

clusters were carried on the large plasmid (DnaA-like replicon). The

and 77%, respectively, which was even higher than the non-

organization of the tda genes in the isolates was very similar to the

challenged control. Addition of the TDA-negative mutant and S60

TDA biosynthesis gene clusters that have been reported previously

resulted in a survival comparable to the non-infected control, hence

in P. inhibens and P. gallaeciensis, and concomitantly the tda genes of

signiﬁcantly (p < 0.001) improving survival compared to the chal-

S26 and S60 exhibited high AA sequence similarity to the tda genes

lenged Artemia (Table 2).

of these species (Table S6). Compared to Ruegeria mobilis, which is

known to regulate TDA expression differently than the Phaeobacter Discussion

species [55], the sequence similarities were lower. This was espe-

cially true for the tdaA gene that encodes the positive regulator of

It was demonstrated in this study that TDA-producing Phaeobac-

the TDA biosynthesis cluster.

ter strains could be isolated from Mediterranean aquaculture units

and that they had potential as probiotics against V. harveyi in crus-

Antagonism in Artemia co-cultures tacean (Artemia) cultures. The Phaeobacter spp. isolated possibly

constituted a new species, signiﬁcantly reduced counts of the ﬁsh

P. inhibens DSM 17395, its TDA-negative mutant, and Phaeobac- pathogens V. anguillarum and V. harveyi, and increased the sur-

ter sp. S26 and S60 grew well in Artemia cultures, reaching densities vival of challenged Artemia. These Phaeobacter sp. isolates from

7 8 −1

of 10 to 10 CFU mL in all the co-cultures. All Phaeobacter grew sea bass hatcheries had an even more pronounced probiotic effect

to approximately equal concentrations in co-culture with V. harveyi than that of Phaeobacter inhibens DSM 17395 and produced more

and in co-culture with V. anguillarum. V. anguillarum and V. har- TDA than any of the tested reference strains. They closely resem-

veyi grew well in all Artemia cultures that were not co-inoculated bled the known Phaeobacter species P. inhibens and P. gallaeciensis

with Phaeobacter, and their concentrations reached approximately [11,27] based on biochemical and phenotypic tests, but the 16S

7 −1 8 −1

10 CFU mL after 24 h and 10 CFU mL after four days (Fig. 2). rRNA gene sequences, as well as whole-genome comparisons, sup-

The numbers of V. anguillarum decreased signiﬁcantly ported the proposal of a new Phaeobacter species. Congruently, it

(p < 0.001) in the presence of Phaeobacter sp. S26 and S60 (Fig. 2). was found that the new strains did not inhibit the growth of V.

V. anguillarum reductions were in the order of four log units, as anguillarum and V. harveyi at lower temperatures. This behavior

compared to the Artemia controls with only V. anguillarum. The was similar to R. mobilis and Phaeobacter sp. 27-4 but different

effects of Phaeobacter DSM 17395, S26 and S60 on V. anguillarum from P. inhibens and P. gallaeciensis. Furthermore, three plasmids

186 T. Grotkjær et al. / Systematic and Applied Microbiology 39 (2016) 180–188

were harbored by both isolates and the replicases of these plas- V. anguillarum is the single most important marine bacterial ﬁsh

mids were identical at the AA level. One isolate, Phaeobacter sp. pathogen and it also infects mollusks and other aquatic organisms

S60, had an extended plasmid proﬁle compared to Phaeobacter sp. [70]. Hence, it has been the target of probiotic treatment in many

S26, yet the proﬁle did not resemble any of the included reference studies [12,18,17,16]. V. anguillarum is not a common crustacean

strains (P. inhibens DSM 17395, P. inhibens DSM 16374 , P. gallae- pathogen, however, Defoirdt et al. [21] tested the inﬂuence of quo-

ciensis DSM 26640 ). This could have implications for their use as rum sensing on virulence in ﬁsh pathogens using V. anguillarum

probionts, since the extended plasmid proﬁle of Phaeobacter sp. NB10 in an Artemia setup. Artemia survival was only 6–8% after 2–5

S60 harbored genes involved in the synthesis of the non-ribosomal days, which is similar to the present ﬁndings. The killing of Artemia

peptide indigoidine and the N-containing heterocyclic secondary could be caused by oxygen depletion by the metabolizing Vibrio

metabolite phenazine. The former is important for the antagonism or by direct pathogenicity. Complete and non-degraded chitinous

of Vibrio ﬁscheri by Phaeobacter sp. strain Y4I [15], and the latter Artemia exuviae could be observed after four days, and V. anguil-

is important both as an antibacterial and in the association with larum was not chitinolytic on chitin plates, although the genome

eukaryotic hosts [45]. In addition, a range of general subsystems harbors chitinase genes [48]. V. anguillarum produces a range of

was identiﬁed in the extended plasmid proﬁle. However, the exact virulence factors [28], however, based on the present study, the spe-

biological implications of the presence of these subsystems were ciﬁc factors causing Artemia mortality could not be identiﬁed. In the

difﬁcult to extrapolate. The new isolates produced higher con- V. anguillarum-challenged Artemia cultures, the pathogen increased

centrations of TDA under the conditions shown in Table 1 and, in abundance on the ﬁnal day (Fig. 2A). This could be due to nutri-

therefore, could potentially be better ﬁsh larval probionts. How- ents released from the dying starved Artemia into the otherwise

ever, resolving the species afﬁliation as well as the probiont efﬁcacy nutrient-poor medium, thus facilitating pathogen proliferation.

will require further studies of both the phenotype and genotype. In this study, the target range of Phaeobacter spp. as probio-

TDA-producing Phaeobacter strains have frequently been iso- tic bacteria was expanded to include the important crustacean

lated from marine aquaculture units and we recently found a pathogen V. harveyi and it was shown that it could be antagonized

regular, seasonal occurrence of P. inhibens on submerged surfaces in by Phaeobacter in a crustacean model challenge system. V. harveyi

Danish harbors [36]. TDA-producing Phaeobacter sp. strains often is also a pathogen that occurs in sea bass rearing [59], and it is one

co-occur with algae, such as macroalgae or benthic diatoms, and of the major pathogens in marine ﬁsh and shrimp culture [69]. It

many other Roseobacter-clade species are often found in association causes large-scale economic losses and, as a consequence of treat-

with microalgae [34,35,37,46,55]. Microalgae are present in any ment efforts, can result in pollution due to antimicrobial agents,

aquaculture facility, either introduced on purpose, or as incidental especially in Asian penaeid shrimp cultures [19,62,71]. Artemia

colonizers of tank walls and water, and TDA-producing Phaeobac- nauplii are used as live feed for many ﬁsh and crustacean species

ter strains may be commensals of these algae [66]. TDA-producing and their cultures offer an ideal substrate for the propagation of

Phaeobacter are excellent bioﬁlm formers [9] and may inhibit set- V. harveyi. In our study, V. harveyi killed Artemia and survival was

tlement of other organisms or even replace an already established as low as 4% after four days. V. harveyi produces chitin-degrading

bioﬁlm [61,60]. enzymes [4] and, in the present study, complete degradation of

The Phaeobacter isolates from the Greek aquaculture units pro- the nauplii with only the eye remaining from the infected Artemia

duced TDA and were thus similar to isolates of P. inhibens, P. was observed. All the exoskeletons of the dead Artemia had been

gallaeciensis and R. mobilis [8,32,55]. TDA is also produced by strains degraded after four days. In a previous study [21] the mortality

of Pseudovibrio [31], which also belongs to the Rhodobacteraceae of gnotobiotically grown Artemia infected with V. harveyi (BB120)

family. The original discovery of TDA was made in an organism was 60–90% which is not as severe as the 96% reported in our study.

identiﬁed as Pseudomonas sp. [41], however, despite this ﬁnding, However, we used a strain isolated from diseased shrimp.

it appears that TDA production is a characteristic feature of only a The probiotic effect, measured as Artemia survival, was as efﬁ-

few marine bacterial species. In this study, TDA-negative mutants cient in V. harveyi-infected cultures as in V. anguillarum-infected

were not created in the new isolates, since the organization of cultures, even though the actual suppression of V. harveyi was not

the biosynthesis genes was identical to P. inhibens DSM 17395, as pronounced as that of V. anguillarum (Table 2). The higher density

but mortality and growth of the ﬁsh pathogens in Artemia co- of V. harveyi may be explained by its more rapid growth, since it had

cultured with a TDA-negative mutant of P. inhibens were compared. a generation time at room temperature of 1 h as opposed to 2.3 h for

In agreement with results from algal and rotifer cultures [17], the V. anguillarum (data not shown). Additionally, there may be other

TDA-negative mutant was not as effective for V. harveyi inhibition unknown factors involved in the survival of the Artemia besides the

as TDA-producing strains, indicating TDA production was also a antagonism of vibrios, since increased survival in cultures supple-

major mechanism of probiotic action in the Artemia system. mented with both vibrios and probionts was observed compared

The present study demonstrated that P. inhibens DSM 17395 and to the axenic controls. Equivalent observations have been reported

the two Phaeobacter sp. strains (S26 and S60) antagonized V. anguil- in ﬁsh larvae systems [18,17].

larum NB10 in an Artemia model system. Planas et al. [54] demon- A key parameter in the successful use of probiotics in aqua-

strated that mortality in rotifer-fed turbot larvae infected with V. culture is the identiﬁcation of when and where to introduce the

anguillarum could be reduced by Phaeobacter sp. 27-4 when both probionts into the system. One rationale would be to introduce

the pathogen and the probiont were enclosed in rotifers and fed to the probionts at, or prior to, the step where pathogens are most

the larvae. However, in spite of feed delivery, the probiont was only prevalent. The intake water could potentially harbor pathogens,

found in the lumen of the larval gut and did not colonize the intesti- however, in the present Greek system sampled, the intake water

nal epithelium. In contrast to this approach, the present study did was ﬁltered bore hole water and CFU counts were very low (data

not aim to produce a probiotic effect in the intestinal tract of the not shown). In larval rearing relying on live feed (algae, rotifers,

larvae, but assessed the potential of Phaeobacter to eliminate the Artemia), the live feed itself is a well-known potential entry and

pathogen before it came into contact with the larvae (i.e. in the feed propagation point for ﬁsh pathogens [24,47,52,63], and introducing

cultures where pathogenic bacteria could otherwise grow to high the probiont at this stage would appear to be a logical step. Several

concentrations). It should be noted that the present study was car- studies have demonstrated that probiotic bacteria can antagonize

ried out using axenic systems in order to rule out the inﬂuence of the ﬁsh pathogenic bacteria in live feed co-cultures [17,43,44]. There-

inherent microbiota of Artemia cultures. Thus, future studies should fore, the live feed organisms have been suggested as an introduction

also address the probiotic effect in non-axenic Artemia cultures. vector for probionts into the system [30,54,53,57]. This study, as

T. Grotkjær et al. / Systematic and Applied Microbiology 39 (2016) 180–188 187

well as previous work [17], demonstrated that probiotic bacteria marine roseobacter Phaeobacter sp. strain Y4I. Appl. Environ. Microbiol. 78 (14),

4771–4780, http://dx.doi.org/10.1128/AEM. 00297-12.

could be introduced at the live feed stage and had a pronounced

[16] D’Alvise, P.W., Melchiorsen, J., Porsby, C.H., Nielsen, K.F., Gram, L. (2010) Inacti-

pathogen-reducing effect, which could potentially limit the subse-

vation of Vibrio anguillarum by attached and planktonic Roseobacter cells. Appl.

quent use of antibiotics for control of pathogenic bacteria. Environ. Microbiol. 76 (7), 2366–2370, http://dx.doi.org/10.1128/AEM. 02717-

09.

Acknowledgements [17] D’Alvise, P.W., Lillebø, S., Prol-Garcia, M.J., Wergeland, H., Nielsen, K.F., Bergh,

Ø., Gram, L. (2012) Phaeobacter gallaeciensis reduces Vibrio anguillarum in cul-

tures of microalgae and rotifers, and prevents vibriosis in cod larvae. PLoS ONE

7 (8), e43996, http://dx.doi.org/10.1371/journal.pone.0043996.

The present study was ﬁnanced by the Technical University of

[18] D’Alvise, P., Lillebø, S., Wergeland, H., Gram, L., Bergh, Ø. (2013) Protection

Denmark (TG PhD grant) and by a grant from the Danish Council

of cod larvae from vibriosis by Phaeobacter spp.: a comparison of strains and

for Strategic Research, Programme Commission on Health, Food introduction times. Aquaculture 384–387, 82–86, http://dx.doi.org/10.1016/j.

and Welfare (12-132390) to LG, and by grant VKR023285 from the aquaculture.2012.12.013.

[19] de la Pena,˜ L.D., Lavilla-Pitogo, C.R., Paner, M.G. (2001) Luminescent

Villum Kann Rasmussen Foundation to LG. We thank Drs. Anna Koza

vibrios associated with mortality in pond-cultured shrimp Penaeus

and Pep Charusanti for help with the genome sequencing of S26 and monodon in the Philippines: species composition. Fish Pathol. 36 (3),

S60. We are grateful to Agilent technologies for the Thought Leader 133–138.

[20] Defoirdt, T., Boon, N., Bossier, P., Verstraete, W. (2004) Disruption of bacte-

Donation of the UHPLC-qTOF system.

rial quorum sensing: an unexplored strategy to ﬁght infections in aquaculture.

Aquaculture 240 (1–4), 69–88, http://dx.doi.org/10.1016/j.aquaculture.2004.

Appendix A. Supplementary data 06.031.

[21] Defoirdt, T., Bossier, P., Sorgeloos, P., Verstraete, W. (2005) The impact of

mutations in the quorum sensing systems of Aeromonas hydrophila, Vibrio

Supplementary data associated with this article can be found,

anguillarum and Vibrio harveyi on their virulence towards gnotobiotically cul-

in the online version, at http://dx.doi.org/10.1016/j.syapm.2016.01. tured Artemia franciscana. Environ. Microbiol. 7 (8), 1239–1247, http://dx.doi.

005. org/10.1111/j.1462-2920.2005.00807.x.

[22] Defoirdt, T., Sorgeloos, P., Bossier, P. (2011) Alternatives to antibiotics for

the control of bacterial disease in aquaculture. Curr. Opin. Biotechnol. 14 (3),

References 251–258, http://dx.doi.org/10.1016/j.mib.2011.03.004.

[23] Devulder, G., Perrière, G., Baty, F., Flandrois, J.P. (2003) BIBI, a bioinformatics

bacterial identiﬁcation tool. J. Clin. Microbiol. 41 (4), 1785–1787, http://dx.doi.

[1] Arias, C., Macián, M., Aznar, R., Garay, E., Pujalte, M. (1999) Low incidence of Vib-

org/10.1128/JCM.41.4.1785.

rio vulniﬁcus among Vibrio isolates from sea water and shellﬁsh of the western

[24] Eddy, S.D., Jones, S.H. (2002) Microbiology of summer ﬂounder Paralichthys

Mediterranean coast. J. Appl. Microbiol. 86 (1), 125–134.

dentatus ﬁngerling production at a marine ﬁsh hatchery. Aquaculture 211 (1–4),

[2] Auch, A.F., Klenk, H.-P., Göker, M. (2010) Standard operating procedure for cal-

9–28, http://dx.doi.org/10.1016/S0044-8486(01)00882-1.

culating genome-to-genome distances based on high-scoring segment pairs.

[25] FAO, WHO, 2001 Report of a joint FAO/WHO expert consultation on evaluation

Stand. Genomic Sci. 2 (1), 142–148, http://dx.doi.org/10.4056/sigs.541628.

of health and nutritional properties of probiotics in food including powder milk

[3] Austin, B. (2010) Vibrios as causal agents of zoonoses. Vet. Microbiol. 140 (3–4),

with live lactic acid, bacteria.

310–317, http://dx.doi.org/10.1016/j.vetmic.2009.03.015.

[26] FAO Fisheries and Aquaculture Department, 2014 The State of World Fisheries

[4] Austin, B., Zhang, X.-H. (2006) Vibrio harveyi: a signiﬁcant pathogen of marine

and Aquaculture.

vertebrates and invertebrates. Lett. Appl. Microbiol. 43 (2), 119–124, http://dx.

[27] Frank, O., Pradella, S., Rohde, M., Scheuner, C., Klenk, H.-P., Göker, M., Petersen,

doi.org/10.1111/j.1472-765X. 2006.01989.x.

J. (2014) Complete genome sequence of the Phaeobacter gallaeciensis type strain

[5] Aziz, R.K., Bartels, D., Best, A.A., DeJongh, M., Disz, T., Edwards, R.A., Formsma,

CIP 105210(T) (= DSM 26640(T) = BS107(T)). Stand. Genomic Sci. 9 (3), 914–932,

K., Gerdes, S., Glass, E.M., Kubal, M., Meyer, F., Olsen, G.J., Olson, R., Osterman,

http://dx.doi.org/10.4056/sigs.5179110.

A.L., Overbeek, R.A., McNeil, L.K., Paarmann, D., Paczian, T., Parrello, B., Pusch,

[28] Frans, I., Michiels, C.W., Bossier, P., Willems, K.A., Lievens, B., Rediers, H. (2011)

G.D., Reich, C., Stevens, R., Vassieva, O., Vonstein, V., Wilke, A., Zagnitko, O.

Vibrio anguillarum as a ﬁsh pathogen: virulence factors, diagnosis and preven-

(2008) The RAST server: rapid annotations using subsystems technology. BMC

tion. J. Fish Dis. 34 (9), 643–661, http://dx.doi.org/10.1111/j.1365-2761.2011.

Genomics 9 (1), 75, http://dx.doi.org/10.1186/1471-2164-9-75.

01279.x.

[6] Bolinches, J., Egidius, E. (1987) Heterotrophic bacterial communities associated

[29] Gatesoupe, F. (1999) The use of probiotics in aquaculture. Aquaculture 180

with the rearing of halibut Hippoglossus hippoglossus with special reference to

(1–2), 147–165.

Vibrio spp. J. Appl. Ichthyol. 3 (1985), 165–173.

[30] Gatesoupe, F.-J. (2002) Probiotic and formaldehyde treatments of Artemia nau-

[7] Bricknell, I., Dalmo, R. (2005) The use of immunostimulants in ﬁsh larval aqua-

plii as food for larval pollack, Pollachius pollachius. Aquaculture 212 (1–4),

culture. Fish Shellﬁsh Immunol. 19 (5), 457–472, http://dx.doi.org/10.1016/j.

fsi.2005.03.008. 347–360, http://dx.doi.org/10.1016/S0044-8486(02)00138-2.

[31] Geng, H., Belas, R. (2010) Expression of tropodithietic acid biosynthesis is con-

[8] Bruhn, J.B., Nielsen, K.F., Hjelm, M. (2005) Ecology, inhibitory activity, and mor-

trolled by a novel autoinducer. J. Bacteriol. 192 (17), 4377–4387, http://dx.doi.

phogenesis of a marine antagonistic bacterium belonging to the Roseobacter

org/10.1128/JB. 00410-10.

clade. Appl. Environ. Microbiol. 71 (11), 7263–7270, http://dx.doi.org/10.1128/

AEM.71.11.7263. [32] Geng, H., Bruhn, J.B., Nielsen, K.F., Gram, L., Belas, R. (2008) Genetic dissec-

tion of tropodithietic acid biosynthesis by marine roseobacters. Appl. Environ.

[9] Bruhn, J.B., Haagensen, J.A., Bagge-Ravn, D., Gram, L. (2006) Culture conditions

Microbiol. 74 (5), 1535–1545, http://dx.doi.org/10.1128/AEM. 02339-07.

of Roseobacter strain 27-4 affect its attachment and bioﬁlm formation as quan-

[33] Gibson, L.F., Khoury, J. (1986) Storage and survival of bacteria by ultra-freeze.

tiﬁed by real-time PCR. Appl. Environ. Microbiol. 72 (4), 3011–3015, http://dx.

doi.org/10.1128/AEM.72.4.3011. Lett. Appl. Microbiol. 3 (6), 127–129, http://dx.doi.org/10.1111/j.1472-765X.

1986.tb01565.x.

[10] Bruhn, J.B., Gram, L., Belas, R. (2007) Production of antibacterial compounds

[34] González, J.M., Moran, M.A. (1997) Numerical dominance of a group of marine

and bioﬁlm formation by Roseobacter species are inﬂuenced by culture con-

bacteria in the alpha-subclass of the class Proteobacteria in coastal seawater.

ditions. Appl. Environ. Microbiol. 73 (2), 442–450, http://dx.doi.org/10.1128/

Appl. Environ. Microbiol. 63 (11), 4237–4242.

AEM. 02238-06.

[35] González, J.M., Simó, R., Massana, R., Covert, J.S., Casamayor, E.O., Pedrós-alió,

[11] Buddruhs, N., Pradella, S., Göker, M., Päuker, O., Pukall, R., Spröer, C., Schu-

C., Pedrós-Alió, C., Moran, M.A. (2000) Bacterial community structure associ-

mann, P., Petersen, J., Brinkhoff, T. (2013) Molecular and phenotypic analyses

ated with a dimethylsulfoniopropionate-producing North Atlantic algal bloom.

reveal the non-identity of the Phaeobacter gallaeciensis type strain deposits CIP

Appl. Environ. Microbiol. 66 (10), 4237–4246, http://dx.doi.org/10.1128/AEM.

105210T and DSM 17395. Int. J. Syst. Evol. Microbiol. 63 (Pt 11), 4340–4349,

http://dx.doi.org/10.1099/ijs.0.053900-0. 66.10.4237-4246.2000.

[36] Gram, L., Rasmussen, B.B., Wemheuer, B., Bernbom, N., Ng, Y.Y., Porsby, C.H.,

[12] Burbank, D.R., Shah, D.H., LaPatra, S.E., Fornshell, G., Cain, K.D. (2011) Enhanced

Breider, S., Brinkhoff, T. (2015) Phaeobacter inhibens from the Roseobacter clade

resistance to coldwater disease following feeding of probiotic bacterial strains

has an environmental niche as a surface colonizer in harbors. Syst. Appl. Micro-

to rainbow trout (Oncorhynchus mykiss). Aquaculture 321 (3–4), 185–190,

http://dx.doi.org/10.1016/j.aquaculture.2011.09.004. biol. 38, 483–493.

[37] Hahnke, S., Brock, N.L., Zell, C., Simon, M., Dickschat, J.S., Brinkhoff, T. (2013)

[13] Cabello, F.C. (2006) Heavy use of prophylactic antibiotics in aquaculture: a

Physiological diversity of Roseobacter clade bacteria co-occurring during a phy-

growing problem for human and animal health and for the environment. Envi-

toplankton bloom in the North Sea. Syst. Appl. Microbiol. 36 (1), 39–48, http://

ron. Microbiol. 8 (7), 1137–1144, http://dx.doi.org/10.1111/j.1462-2920.2006.

01054.x. dx.doi.org/10.1016/j.syapm.2012.09.004.

[38] Hjelm, M., Bergh, Ø., Riaza, A., Nielsen, J., Melchiorsen, J., Jensen, S., Duncan,

[14] Croxatto, A., Lauritz, J., Chen, C., Milton, D.L. (2007) Vibrio anguillarum

H., Ahrens, P., Birkbeck, H., Gram, L. (2004) Selection and identiﬁcation of

colonization of rainbow trout integument requires a DNA locus involved

autochthonous potential probiotic bacteria from turbot larvae (Scophthalmus

in exopolysaccharide transport and biosynthesis. Environ. Microbiol. 9 (2),

maximus) rearing units. Syst. Appl. Microbiol. 27 (3), 360–371.

370–382, http://dx.doi.org/10.1111/j.1462-2920.2006.01147.x.

[39] Hugh, R., Leifson, E. (1953) The taxonomic signiﬁcance of fermentative versus

[15] Cude, W.N., Mooney, J., Tavanaei, A.A., Hadden, M.K., Frank, A.M., Gulvik,

oxidative metabolism of carbohydrates by various Gram negative bacteria. J.

C.A., May, A.L., Buchan, A. (2012) Production of the antimicrobial secondary

Bacteriol. 66 (1), 24–26.

metabolite indigoidine contributes to competitive surface colonization by the

188 T. Grotkjær et al. / Systematic and Applied Microbiology 39 (2016) 180–188

[40] Kildgaard, S., Mansson, M., Dosen, I., Klitgaard, A., Frisvad, J.C., Larsen, T.O., Vibrio in microalgae, rotifer, Artemia and ﬁrst feeding turbot (Psetta maxima) lar-

Nielsen, K.F. (2014) Accurate dereplication of bioactive secondary metabolites vae. J. Appl. Microbiol. 106 (4), 1292–1303, http://dx.doi.org/10.1111/j.1365-

from marine-derived fungi by UHPLC-DAD-QTOFMS and a MS/HRMS library. 2672.2008.04096.x.

Mar. Drugs 12 (6), 3681–3705, http://dx.doi.org/10.3390/md12063681. [58] Prol García, M., D’Alvise, P., Rygaard, A., Gram, L. (2014) Bioﬁlm formation is

[41] Kintaka, K., Ono, H., Tsubotani, S., Harada, S., Okazaki, H. (1984) Thiotropocin, not a prerequisite for production of the antibacterial compound tropodithietic

a new sulfur-containing 7-membered-ring antibiotic produced by a Pseu- acid in Phaeobacter inhibens DSM17395. J. Appl. Microbiol. 117, 1592–1600.

domonas sp. J. Antibiot. (Tokyo) 37 (11), 1294–1300. [59] Pujalte, M.J., Sitjà-Bobadilla, A., Macián, M.C., Belloch, C., Álvarez-Pellitero, P.,

[42] Lauzon, H., Gudmundsdóttir, S., Pedersen, M., Budde, B., Gudmundsdóttir, Pérez-Sánchez, J., Uruburu, F., Garay, E. (2003) Virulence and molecular typing

B. (2008) Isolation of putative probionts from cod rearing environment. of Vibrio harveyi strains isolated from cultured dentex, gilthead sea bream and

Vet. Microbiol. 132 (3–4), 328–339, http://dx.doi.org/10.1016/j.vetmic.2008. European sea bass. Syst. Appl. Microbiol. 26 (2), 284–292, http://dx.doi.org/10.

05.014. 1078/072320203322346146.

[43] Marques, A., Ollevier, F., Verstraete, W., Sorgeloos, P., Bossier, P. (2006) Gno- [60] Rao, D., Webb, J.S., Kjelleberg, S. (2005) Competitive interactions in mixed-

tobiotically grown aquatic animals: opportunities to investigate host-microbe species bioﬁlms containing the marine bacterium Pseudoalteromonas tunicata.

interactions. J. Appl. Microbiol. 100 (5), 903–918, http://dx.doi.org/10.1111/j. Appl. Environ. Microbiol. 71 (4), 1729–1736, http://dx.doi.org/10.1128/AEM.

1365-2672.2006.02961.x. 71.4.1729.

[44] Marques, A., Thanh, T.H., Sorgeloos, P., Bossier, P. (2006) Use of microalgae [61] Rao, D., Webb, J.S., Holmström, C., Case, R., Low, A., Steinberg, P., Kjelleberg,

and bacteria to enhance protection of gnotobiotic Artemia against differ- S. (2007) Low densities of epiphytic bacteria from the marine alga Ulva aus-

ent pathogens. Aquaculture 258 (1–4), 116–126, http://dx.doi.org/10.1016/j. tralis inhibit settlement of fouling organisms. Appl. Environ. Microbiol. 73 (24),

aquaculture.2006.04.021. 7844–7852, http://dx.doi.org/10.1128/AEM. 01543-07.

[45] Mavrodi, D.V., Peever, T.L., Mavrodi, O.V., Parejko, J.A., Raaijmakers, J.M., [62] Rattanama, P., Srinitiwarawong, K., Thompson, J.R., Pomwised, R., Supamattaya,

Lemanceau, P., Mazurier, S., Heide, L., Blankenfeldt, W., Weller, D.M., K., Vuddhakul, V. (2009) Shrimp pathogenicity, hemolysis, and the presence of

Thomashow, L.S. (2010) Diversity and evolution of the phenazine biosynthesis hemolysin and TTSS genes in Vibrio harveyi isolated from Thailand. Dis. Aquat.

pathway. Appl. Environ. Microbiol. 76 (3), 866–879, http://dx.doi.org/10.1128/ Organ. 86 (2), 113–122, http://dx.doi.org/10.3354/dao02119.

AEM. 02009-09. [63] Reid, H.I., Treasurer, J.W., Adam, B., Birkbeck, T.H. (2009) Analysis of bacterial

[46] Moran, M.A., González, J.M., Kiene, R.P. (2003) Linking a bacterial taxon to sulfur populations in the gut of developing cod larvae and identiﬁcation of Vibrio logei,

cycling in the sea: studies of the marine Roseobacter group. Geomicrobiol. J. 20 Vibrio anguillarum and Vibrio splendidus as pathogens of cod larvae. Aquaculture

(4), 375–388, http://dx.doi.org/10.1080/01490450303901. 288 (1–2), 36–43, http://dx.doi.org/10.1016/j.aquaculture.2008.11.022.

[47] Munro, P., Barbour, A., Birkbeck, T. (1995) Comparison of the growth and sur- [64] Richter, M., Rosselló-Móra, R., Glöckner, F.O., Peplies, J. (2015) JSpeciesWS:

vival of larval turbot in the absence of culturable bacteria with those in the a web server for prokaryotic species circumscription based on pairwise

presence of Vibrio anguillarum, Vibrio alginolyticus or a marine Aeromonas sp. genome comparison. Bioinformatics (November), http://dx.doi.org/10.1093/

Appl. Environ. Microbiol. 61, 4425–4428. bioinformatics/btv681.

[48] Naka, H., Dias, G.M., Thompson, C.C., Dubay, C., Thompson, F.L., Crosa, J.H. (2011) [65] Ruiz-Ponte, C., Cilia, V., Lambert, C., Nicolas, J. (1998) Roseobacter gallaeciensis

Complete genome sequence of the marine ﬁsh pathogen Vibrio anguillarum sp. nov., a new marine bacterium isolated from rearings and collectors of the

harboring the pJM1 virulence plasmid and genomic comparison with other vir- scallop Pecten maximus. Int. J. Syst. Bacteriol. 48, 537–542.

ulent strains of V. anguillarum and V. ordalii. Infect. Immun. 79 (7), 2889–2900, [66] Seyedsayamdost, M.R., Case, R.J., Kolter, R., Clardy, J. (2011) The Jekyll-and-

http://dx.doi.org/10.1128/IAI. 05138-11. Hyde chemistry of Phaeobacter gallaeciensis. Nat. Chem. 3 (4), 331–335, http://

[49] Nogami, K., Maeda, M. (1992) Bacteria as biocontrol agents for rearing larvae dx.doi.org/10.1038/nchem.1002.

of the crab Portunus trituberculatus. Can. J. Fish. Aquat. Sci. 49, 2373–2376. [67] Skov, M.N., Pedersen, K., Larsen, J.L. (1995) Comparison of pulsed-ﬁeld gel elec-

[50] Norqvist, A., Hagström, Å., Wolf-Watz, H. (1989) Protection of rainbow trout trophoresis, ribotyping, and plasmid proﬁling for typing of Vibrio anguillarum

against vibriosis and furunculosis by the use of attenuated strains of Vibrio serovar O1. Appl. Environ. Microbiol. 61 (4), 1540–1545.

anguillarum. Appl. Environ. Microbiol. 55 (6), 1400–1405. [68] Sorgeloos, P., Bossuytz, E., Lavina,˜ E., Baeza-Mesa, M., Persoone, G. (1977)

[51] Norqvist, A., Norrman, B., Wolf-Watz, H. (1990) Identiﬁcation and characteri- Decapsulation of Artemia cysts: a simple technique for the improvement of

zation of a zinc metalloprotease associated with invasion by the ﬁsh pathogen the use of brine shrimp in aquaculture. Aquaculture 12, 311–315.

Vibrio anguillarum. Infect. Immun. 58 (11), 3731–3736. [69] Thompson, F.L., Iida, T., Swings, J. (2004) Biodiversity of vibrios. Microbiol. Mol.

[52] Olafsen, J.A. (2001) Interactions between ﬁsh larvae and bacteria in marine Biol. Rev. 68 (3), http://dx.doi.org/10.1128/MMBR.68.3.403.

aquaculture. Aquaculture 200 (1–2), 223–247, http://dx.doi.org/10.1016/ [70] Toranzo, A.E., Magarinos,˜ B., Romalde, J.L. (2005) A review of the main bacterial

S0044-8486(01)00702-5. ﬁsh diseases in mariculture systems. Aquaculture 246 (1–4), 37–61, http://dx.

[53] Planas, M., Vázquez, J.A., Marqués, J., Pérez-Lomba, R., González, M.P., Murado, doi.org/10.1016/j.aquaculture.2005.01.002.

M. (2004) Enhancement of rotifer (Brachionus plicatilis) growth by using ter- [71] Vandenberghe, J., Li, Y., Verdonck, L., Li, J., Sorgeloos, P., Xu, H., Swings, J. (1998)

restrial lactic acid bacteria. Aquaculture 240 (1–4), 313–329, http://dx.doi.org/ Vibrios associated with Penaeus chinensis (Crustacea: Decapoda) larvae in Chi-

10.1016/j.aquaculture.2004.07.016. nese shrimp hatcheries. Aquaculture 169, 121–132.

[54] Planas, M., Pérez-Lorenzo, M., Hjelm, M., Gram, L., Uglenes Fiksdal, I., Bergh, Ø., [72] Verschuere, L., Rombaut, G., Sorgeloos, P., Verstraete, W. (2000) Probiotic bacte-

Pintado, J. (2006) Probiotic effect in vivo of Roseobacter strain 27-4 against Vibrio ria as biological control agents in aquaculture. Microbiol. Mol. Biol. Rev. 64 (4),

(Listonella) anguillarum infections in turbot (Scophthalmus maximus L.) lar- 655–671.

vae. Aquaculture 255 (1–4), 323–333, http://dx.doi.org/10.1016/j.aquaculture. [73] Vesth, T., Lagesen, K., Acar, Ö., Ussery, D. (2013) CMG-biotools, a free workbench

2005.11.039. for basic comparative microbial genomics. PLOS ONE 8 (4), e60120, http://dx.

[55] Porsby, C.H., Nielsen, K.F., Gram, L. (2008) Phaeobacter and Ruegeria species of doi.org/10.1371/journal.pone.0060120.

the Roseobacter clade colonize separate niches in a Danish turbot (Scophthal- [74] Weinstein, J.N., Myers, T., Buolamwini, J., Raghavan, K., Van Osdol, W., Licht, J.,

mus maximus)-rearing farm and antagonize Vibrio anguillarum under different Viswanadhan, V.N., Kohn, K.W., Rubinstein, L.V., Koutsoukos, A.D., Monks, A.,

growth conditions. Appl. Environ. Microbiol. 74 (23), 7356–7364, http://dx.doi. Scudiero, D.A., Zaharevitz, D., Chabner, B.A., Anderson, N.L., Grever, M.R., Paull,

org/10.1128/AEM. 01738-08. K.D. (1994) Predictive statistics and artiﬁcial intelligence in the U.S. National

[56] Porsby, C., Webber, M., Nielsen, K.F., Piddock, L.J., Gram, L. (2011) Resistance Cancer Institute’s drug discovery program for cancer and AIDS. Stem Cells 12

and tolerance to tropodithietic acid, an antimicrobial in aquaculture, is hard (1), 13–22.

to select. Antimicrob. Agents Chemother. 55 (4), 1332–1337, http://dx.doi.org/ [75] Zhang, X.-H., Austin, B. (2000) Pathogenicity of Vibrio harveyi to salmonids. J.

10.1128/AAC.01222-10. Fish Dis. 23 (2), 93–102, http://dx.doi.org/10.1046/j.1365-2761.2000.00214.x.

[57] Prol, M.J., Bruhn, J.B., Pintado, J., Gram, L. (2009) Real-time PCR detection and

quantification of fish probiotic Phaeobacter strain 27-4 and fish pathogenic Paper 2

Phaeobacter inhibens as probiotic bacteria in non-axenic Artemia and algae cultures

Aquaculture 462 (2016) 64–69

Contents lists available at ScienceDirect

Aquaculture

journal homepage: www.elsevier.com/locate/aquaculture

Phaeobacter inhibens as probiotic bacteria in non-axenic Artemia and algae cultures

Torben Grotkjær a, Mikkel Bentzon-Tilia a, Paul D'Alvise a,1, Kristof Dierckens b, Peter Bossier b,LoneGrama,⁎ a Department of Systems Biology, Technical University of Denmark, Matematiktorvet Bldg. 301, DK-2800 Kgs. Lyngby, Denmark b Laboratory of Aquaculture & Artemia Reference Center, Ghent University, Rozier 44, B-9000, Belgium article info abstract

Article history: The growing aquaculture industry is in need for non-antibiotic based disease control strategies to reduce risk Received 2 April 2016 of bacteria developing and spreading antibiotic resistance. We have previously, in axenic model systems of live Received in revised form 2 May 2016 larval feed, demonstrated that bacteria from the Roseobacter clade can antagonize fish pathogens such as Vibrio Accepted 2 May 2016 anguillarum and Vibrio harveyi and that they can reduce larval mortality in challenge trials. However, in the Available online 4 May 2016 aquaculture production, a natural microbiota is present at all stages and may affect the efficacy of the probiotic Keywords: bacteria. The purpose of the present study was to determine if marine roseobacters in non-axenic systems fi Live feed were capable of antagonizing sh pathogenic vibrios. We added a controlled background microbiota of four Artemia bacterial strains to axenic Artemia and algae (Duniella) and these bacteria had a marginal but significant reducing Microalgae effect on inoculated Vibrio anguillarum that grew to 107 in control samples but to a level 1–2loglowerinsamples Fish probiotics with background microbiota. The addition of the Roseobacter-clade bacteria, Phaeobacter inhibens,causeda Phaeobacter significant reduction in growth of the pathogen that reached levels 3–4 log lower than in the control. In non- Roseobacter axenic natural Artemia and algae (Tetraselmis) received from an aquaculture unit, Vibrio anguillarum grew to Vibrio anguillarum 107 CFU/ml but only reached 104 CFU/ml when P. inhibens was also added. P. inhibens was added at a concentration 106 CFU/ml in all systems and remained at this concentration at the end of the study, irrespective of the background microbiota. We therefore conclude that P. inhibens are indeed promising as probiotic bacteria in marine larvi-culture where it in natural live feed can suppress fish larval pathogens. Statement of relevance: We and others have in several studies demonstrated the potential effect of probiotic bacteria in axenic or gnotobiotic models. Here, we for the first time present data that demonstrate that probiotic roseobacters also are efficient in non-axenic systems; even in commercial algae and Artemia from a commercial fish farmer. This adds further promise to probiotics as disease control in aquaculture. © 2016 Elsevier B.V. All rights reserved.

1. Introduction clade bacteria can antagonize fish pathogenic bacteria in axenic systems of live larval feed and they can also, in controlled model systems, A major constraint in aquaculture is the sudden reduction (crash) in improve the survival of infected fish larvae (D'Alvise et al., 2012, a number of fish larval cultures caused by pathogenic bacteria (Reid 2013; Grotkjær et al., 2016). The roseobacters are marine alpha- et al., 2009). Antibiotics can control these crashes, however, widespread Proteobacteria that are widely distributed across climate zones use of antibiotics is not sustainable because it selects for antibiotic (Buchan et al., 2005; Segev et al., 2015; Sonnenshein et al., 2016). Espe- resistant bacteria, and this resistance can be transferred to human path- cially the genera Phaeobacter and Ruegeria have repeatedly been isolat- ogens (Rhodes et al., 2000). Several more sustainable disease control ed from aquaculture units (Grotkjær et al., 2016; Hjelm et al., 2004b; measures are being explored in aquaculture. This includes vaccination, Porsby et al., 2008) and recently, Phaeobacter inhibens was repeatedly immune-stimulants, prebiotics and probiotics. Vaccination requires a isolated from biofilms in Danish habour areas (Gram et al., 2015). functioning immune system with memory, which is not yet developed To be able to control and manipulate the composition of bacteria in at the fish larval stage. Non-antibiotic-based strategies include aquaculture live feed it is important to understand the complex interac- immune-stimulating compounds, quorum sensing inhibiting com- tions between the microbes and the live feed (Berland et al., 1970; pounds, bacteriophages and fish probiotic bacteria. Marine Roseobacter Grossart and Simon, 2007; Nakase and Eguchi, 2007; Salvesen et al., 2000) and to determine if the probiotic bacteria are capable of pathogen suppression also in systems with a complex microbiota. Bacteria grow ⁎ Corresponding author. well in algal cultures as they are supplied with organic compounds E-mail address: [email protected] (L. Gram). 1 Current address: University of Hohenheim, Institute for Animal Science, Population from the algal organisms (Cole, 1982). Also, bacteria may be beneficial genomics group, Garbenstr. 17, Room 008, 70599 Stuttgart, Germany. for the algae by decomposing organic material or by producing

http://dx.doi.org/10.1016/j.aquaculture.2016.05.001 0044-8486/© 2016 Elsevier B.V. All rights reserved. T. Grotkjær et al. / Aquaculture 462 (2016) 64–69 65 secondary metabolites (Berland et al., 1970; Seyedsayamdost et al., Table 1 2011a). The use of live feed such as algae and rotifers in aquaculture Bacterial strains used in the study. improves the growth and health of fish larvae during their first feeding Strain Isolated from Source or reference since the nutritional quality is improved (Reitan et al., 1993)and Vibrio anguillarum NB10 Oncorhynchus mykiss Croxatto et al. (2007) because of the bacterial composition (Skjermo and Vadstein, 1993). Phaeobacter inhibens Pecten maximus D'Alvise et al. (2012) The microbiota of fish larvae is changed when algae are added (Bergh DSM17395 GMr et al., 1994), where a gradual shift of nonfermentative bacteria of the LT3 (Bacillus sp.) Penaeus vannamei Defoirdt et al. (2011) Cytophaga/Flexibacter/Flavobacteriurn group is to be dominated by the B7 (Acinetobacter sp.) Dicentrarchus labrax Liu et al. (2010) fi M13(Ochrobactrum sp) Macrobrachium rosenbergi Liu et al. (2010) fermentative Vibrio/Aeromonas group. Newly hatched sh larvae ingest M15(Ochrobactrum sp) Macrobrachium rosenbergi Liu et al. (2010) algae but are often not able to digest the cell contents and therefor need bacteria and the extracellular material produced by the algae as a significant source of nutrition (Daume, 2006). The nauplii of the brine shrimp kindly provided by D. Milton, University of Umeå (Croxatto et al., Artemia are the most commonly used live feed organism in aquaculture 2007). The gfp-tagged Vibrio anguillarum NB10 was counted using a industry (Van Stappen, 1996). Unfortunately the Artemia nauplii are flow cytometer (BD Accuri C6, Becton, Dickinson and Company) and also a vector for introducing bacteria into the hatchery systems on selective medium (Tryptone Soy Agar (TSA, Difco 212185) supple- (Austin and Allen, 1982; Benavete and Gatesoupe, 1988) herein oppor- mented with 10 mg/l chloramphenicol). The Phaeobacter inhibens tunistic pathogenic bacteria (Gomez-Gil et al., 1994; López-Torres and DSM17395 gentamicin resistant (GMr) has been tagged chromosomally Lizárraga-Partida, 2001). The bacterial community of newly hatched with a miniTn7(Gm)PA1/04/03DsRedExpress-a cassette, using a mini- nauplii is dominated by uncultured members of Gammaproteobacteria Tn7 tagging system (D'Alvise et al., 2012). The four bacterial strains and Planctomycetales situated in the gut rather than at their external that were used as a controlled microbiota were isolated from Artemia surfaces (Høj et al., 2009), and can reach densities of up to 107 within cultures and shrimp or fish in aquaculture and were provided by The 24 h (Austin and Allen, 1982). Artemia Reference Center. These had been identified by 16S rRNA The use of Roseobacter clade bacteria as probiotic bacteria in larvi- gene sequence analyses. Bacteria from frozen stock cultures (−80 °C) culture requires consideration of how to introduce the bacteria to the were streaked on MA and MA with 25 mg/l gentamicin was used for aquaculture system. It is not feasible to produce the quantities of counting P. inhibens DSM17395 GMr. The bacterial pre-cultures for the probiotics that would be needed to add them on a continuous basis to Artemia and algae experiments were grown in 20 ml of 1/2 YTSS (2 g the larvae tanks, and we and others (D'Alvise et al., 2012, 2013; Bacto Yeast extract, 1.25 g Bacto Tryptone, 20 g Sigma Sea Salts, 1 l de- Grotkjær et al., 2016) have suggested introducing the probiont via the ionized water) (Sobecky et al., 1997)at25°Cunderaeratedconditions. live feed which would require lower quantities of probiont culture. Also, the live feed is potentially the source of pathogens and thus a 2.2. Preparation of Artemia for pathogen-probiont experiments with logic step in the process to intervene. Usually, pathogens are not defined background bacteria introduced with the intake water (Olafsen, 2001; Pintado et al., 2010; Tinh et al., 2007), but are likely carried by the live feed where concentra- Axenic Artemia were prepared by disinfecting 1 g of Artemia cysts tions of nutrients from microalgae or Artemia are high. These dense (Yik Sung et al., 2007) and decapsulating them in 90 ml deionized cultures allow the opportunistic pathogenic bacteria to proliferate. In water for 1 h. 3.3 ml 30% NaOH, 50 ml NaOCl and 70 ml 1% Na2S2O3 addition to pathogens thriving in these live feed cultures, other were added for 2 min. The Artemia cysts were filtered and washed commensal background bacteria also proliferate (Grossart and Simon, with 1 l 3.5% Instant Ocean sea saltwater (3% Instant Ocean sea salts; 2007; Grossart et al., 2005), and this raises the question whether they Aquarium Systems Inc., Sarrebourg, France). The cysts were transferred will interfere with the growth and colonization of the probiont and, to a 1-liter blue cap bottle with 1 l 3.5% IO sea saltwater and incubated more importantly, how they affect the antagonism of the pathogens for 24 h, aerated at 30 °C. For bacterial mono- or co-culture combina- by the probionts. Also, the probiotic bacteria may affect the composition tions, 20 individual live Artemia were pipetted into a 50 ml falcon tube of the natural microbiota and since this community interacts in several containing 20 ml 3.5% IO. For the non-axenic setup, cultures with ways with the algae (Grossart and Simon, 2007; Jensen et al., 1996; 20 mg of non-disinfected Artemia cysts were hatched in 20 ml 3.5% IO Seyedsayamdost et al., 2011a), it is important to know how addition and incubated at 20 °C shaken at 200 rpm. The cysts hatched at this tem- of a probiotic bacteria potentially affects the natural microbiota. perature within approx. 24 h. Most studies testing the potential of different probiotic strategies have been based on axenic systems. The purpose of the present study 2.3. Preparation of axenic Dunaliella tertiolecta for pathogen-probiont was to determine if the probiotic effect by roseobacters that we have experiments with defined background microbiota previously demonstrated (D'Alvise et al., 2012, 2013; Grotkjær et al., 2016; Planas et al., 2006; Porsby et al., 2008) could be transferred to a The Dunaliella tertiolecta algae stock was grown in 20 ml algae non-axenic system. We first mimick a natural live-feed setup by adding, medium (1 ml WALNE's medium for algal cultures, 0.1 ml vitamin solu- in a controlled manner, bacterial strains isolated from aquaculture to tion, 1 l autoclaved seawater) (Walne, 1970). The algae culture was in- Artemia in parallel to the addition of pathogen and probionts. We finally cubated at 20 °C in an algae growth room on a rotator at 50 rpm for 48 h. demonstrate that in natural, non-axenic live feed cultures, the probiotic The Dunaliella tertiolecta was treated with a cocktail of five antibiotics Phaeobacter inhibens is capable of inhibiting the fish pathogen Vibrio (gentamicin, tetracycline, chloramphenicol, streptomycin and penicillin anguillarum. G) of 50 mg/l for 48 h at 20 °C in an algae growth room on a rotator. The antibiotic treated and an untreated Dunaliella tertiolecta culture were 2. Material and methods centrifugedat6000×g and washed with sterile 3.5% IO three times to remove residual antibiotics. Bacterial concentrations of the cultures 2.1. Bacterial strains and media were determined after 48 h of incubation by plating on MA plates.

All bacterial strains are listed in Table 1. The pathogenic Vibrio 2.4. Preparation of non-axenic Tetraselmis suecica for anguillarum strain NB10 was isolated from the Gulf of Bothnia and has pathogen-probiont experiments caused disease in rainbow trout (Norqvist et al., 1989, 1990). The strain used is a variant that has been tagged by insertion of plasmid pNQFlaC4- Non-axenic Tetraselmis suecica cultures in f/2 medium (Guillard and gfp27 (cat, gfp) into an intergenic region on the chromosome, and was Ryther, 1962) in 3% IO was acquired from a commercial aquaculture 66 T. Grotkjær et al. / Aquaculture 462 (2016) 64–69 unit. The concentration of Tetraselmis suecica was determined using a Neubauer-improved counting chamber. Tetraselmis suecica was added to 50 ml falcon tubes with 20 ml 3.5% IO aiming at 105 cells/ml as the initial concentration. All cultures were incubated at 20 °C shaken at 200 rpm with a light intensity on the tubes of 2000 lx.

2.5. Pathogen-probiont experiments in Artemia and Dunaliella tertiolecta cultures

Pre-cultures of V. anguillarum NB10 and the microbiota strains

(Table 1) were adjusted to OD600 = 0.5 by diluting with 1/2 YTSS, and 10-fold serially diluted in artificial seawater (3.5% Instant Ocean sea salts). Fifty μl of the appropriate bacterial dilutions were added to the 4 Artemia and Dunaliella tertiolecta cultures, aiming at 10 CFU/ml as the Fig. 1. Counts of P. inhibens DSM 17395 GMr in axenic Artemia cultures. P. inhibens DSM initial concentration. One hundred μl of overnight P. inhibens 17395 GMr inoculated at 106 CFU/ml alone (○), or in the presence of V. anguillarum (□), DSM17395 GMr culture was added to the Artemia and Dunaliella V. anguillarum and background strains separately (▲, ▼, ●, ■)orV. anguillarum and all Δ tertiolecta cultures, aiming at 106 CFU/ml as the initial concentration. background strains ( ). Points are average of two biological replicates and errorbars are standard deviation of the mean. One hundred μl of 1/2 YTSS medium was added to the axenic Artemia control cultures and to the monoxenic Vibrio and background controls. The Artemia cultures were incubated at 30 °C on a rotary shaker at 200 rpm. The Dunaliella tertiolecta cultures were incubated at 20 °C, and 2). V. anguillarum grew in Artemia cultures that were inoculated on a rotary shaker at 50 rpm in an algae growth room. All challenge cul- with the background bacteria separately and reached densities of 105 tures were done in independent duplicates. Bacterial concentrations to 106 (Fig. 3). The four background bacteria were enumerated as the were determined daily. P. inhibens DSM17395 GMr numbers were non-fluorescent counts and all were inoculated at approx. 103 cells/ml determined on selective medium (MA with 25 mg/l gentamicin). Sam- and grew in two days to 106 cells/ml (data not shown). ples from cultures without P. inhibens DSM17395 GMr were also plated The background microbiota strains caused a slight reduction approx- as controls. Concentrations of the gfp tagged V. anguillarum NB10 were imately one log unit of the V. anguillarum as compared to the control determined by using a flow cytometer (BD Accuri C6, Becton, Dickinson without background microbiota strains. Addition of P. inhibens 17395 and Company), using a 533 ± 30 filter for green fluorescence. For none GMr, caused a significant (p b 0.001) reduction of V. anguillarum of 3 fluorescent protein tagged strains SYBR green staining was used togeth- log units in all combinations (Figs. 4 and 5) as compared to the Artemia er with the 533 ± 30 filter. and Dunaliella tertiolecta where only V. anguillarum was added.

2.6. Pathogen-probiont experiments in non-axenic Artemia and 3.2. Non-axenic Artemia and Tetraselmis suercica Tetraselmis suecica cultures pathogen-probiont experiments

V. anguillarum NB10 pre-cultures were adjusted to OD600 = 0.5 by Attempts were made to use FACS to enumerate V. anguillarum,how- diluting with 1/2 YTSS, and 10-fold serially diluted in 3.5% IO. Fifty μl ever, numbers were too low compared to the level of the overall back- of the appropriate bacterial dilutions were added to the non-axenic ground microbiota and despite several attempts to adjust the gating Artemia and Tetraselmis suecica cultures, aiming at 104 CFU/ml as the ini- parameters, it was not possible to get reliable counts. Therefore, tial concentration. One hundred μl of overnight P. inhibens DSM17395 V. anguillarum were enumerated on chloramphenicol plates. GMr culture was added to the non-axenic Artemia and Tetraselmis V. anguillarum NB10 was able to colonize and grew well in non-axenic suecica cultures, aiming at 106 CFU/ml as the initial concentration. One Artemia and Tetraselmis cultures and reached concentrations of 107 hundred μl of 1/2 YTSS medium was added to the non-axenic Artemia (Figs. 6 and 7). The concentration of V. anguillarum NB10 decreased sig- and Tetraselmis suecica control cultures and to the cultures with niﬁcantly (p b 0.001) in the presence of P. inhibens DSM 17395 GMr in V. anguillarum NB10 together with the non-axenic Artemia or the non- axenic Tetraselmis suecica. Bacterial concentrations of V. anguillarum NB10 were determined daily by plating on selective medium (TSA with 10 mg/l chloramphenicol).

2.7. Statistical analysis

Bacterial counts were log-transformed, and differences e.g. between V. anguillarum counts under different treatments were tested for signiﬁcance using ANOVA on GraphPad Prism 5.00 (GraphPad Software, San Diego CA). Tukey's multiple comparisons test was used for pairwise comparisons. Values from day 0, which are merely a proof of the inoculated concentrations, were omitted in the analysis.

3. Results

3.1. Artemia and Dunaliella tertiolecta pathogen-probiont experiments Fig. 2. Counts of P. inhibens DSM 17395 GMr in antibiotic treated D. tertiolecta. P. inhibens with deﬁned background bacteria DSM 17395 GMr inoculated at 106 CFU/ml alone (○) or in the presence of V. anguillarum (□), V. anguillarum and background strains separately (▲, ▼, ●, ■)orV. anguillarum and all background strains (Δ). Counts of P. inhibens DSM 17395 GMr in D. tertiolecta P. inhibens DSM 17395 GMr colonized all of the Artemia and with natural microbiota in presence of V. anguillarum and all of the background strains 6 8 Dunaliella tertiolecta cultures reaching densities of 10 to 10 CFU/ml (▽). Points are average of two biological replicates and errorbars are standard deviation after 48 h irrespective of addition of a background microbiota (Figs. 1 of the mean. T. Grotkjær et al. / Aquaculture 462 (2016) 64–69 67

Fig. 3. Counts of V. anguillarum NB10 in axenic Artemia. Gfp tagged cells of V. anguillarum Fig. 5. Counts V. anguillarum in D. tertiolecta cultures. Gfp tagged cells of V. anguillarum 4 inoculated at 104 CFU/ml alone (○), or in the presence of background strain LT3 (▲), inoculated in antibiotic treated D. tertiolecta at 10 CFU/ml alone (○), or in the presence r r background strain B7 (▼), background strain M13 (●), or background strain M15 (■). of P. inhibens 17395 GM (□), or P. inhibens 17 395 GM andbackgroundstrains Points are average of two biological replicates and errorbars are standard deviation of separately (▲, ▼, ●, ■), or in the presence of all the strains (Δ). Counts of V. anguillarum the mean. in D. tertiolecta (▽) with natural microbiota and added probiont. Points are average of two biological replicates and errorbars are standard deviation of the mean. both culture settings (Figs. 6 and 7). The reduction was in order of 4 log units in the non-axenic Artemia cultures and 3 log units in non-axenic 1996). P. inhibens have previously been shown to colonize axenic Tetraselmis suercica cultures as compared to the non-axenic controls algae (Seyedsayamdost et al., 2011b, 2014)aswellasotherlive with only added V. anguillarum NB10. In both the Artemia and the feed subjects such as Artemia and rotifers (D'Alvise et al., 2012; Tetraselmis suercica with no added pathogen or probiont, an increase Grotkjær et al., 2016), but our data suggest that this is also the case in chloramphenicol resistant bacteria was seen in the order of almost when P. inhibens is co-cultured with an existing microbiota. Irrespec- 2logunits(Figs. 6 and 7). tive of whether one or more of the background strains were added, P. inhibens maintained cell densities of around 106 CFU/ml, or grew 4. Discussion an additional order of magnitude. This is congruent with the fact that members of the Phaeobacter genus, as well as other genera of To circumvent the need for prophylactic use of antibiotics in the Roseobacter clade, are readily isolated from the aquaculture aquaculture, the application of probiotics seems to hold great potential environments (Grotkjær et al., 2016; Hjelm et al., 2004a). and previous studies have demonstrated a clear probiotic effect of the We observed the largest increase in P. inhibens density in non-axenic Roseobacter clade bacterium Phaeobacter inhibens in axenic model D. tertiolecta cultures compared to the non-axenic Artemia cultures. This systems (D'Alvise et al., 2012, 2013; Grotkjær et al., 2016; Prol-García may be due to nutritional differences in the respective media since and Pintado, 2013). Our results demonstrate that moving P. inhibens D. tertiolecta were grown in WALNE's medium, which is rich in ammo- from an axenic model system to a defined non-axenic, and ultimately nium, iron, and phosphate, while the experiments with P. inhibens and to a completely non-axenic, live feed model system, does not interfere Artemia were performed in pure Instant Ocean®. However, it has been with the establishment of the potential probiont, nor with its antagonis- suggested that P. inhibens lives in association with phytoplankton in tic effects on the fish pathogenic bacterium Vibrio anguillarum. its natural environment (Seyedsayamdost et al., 2011a), and hence In the application of probiotics it is imperative that the probiont live feed in the form of algae may be particularly suited to accommodate is able to establish itself in the aquaculture environment. This may Phaeobacter probionts. potentially be a problem when supplying probionts through the live Similarly, the pathogenic V. anguillarum strain NB10 was able to feed as different kinds of live feed already have different established establish itself in the Artemia cultures where it swiftly increased three microbiotas (Austin and Allen, 1982; Berland et al., 1970; Grossart orders of magnitude. This is in line with previous findings where the or- and Simon, 2007; López-Torres and Lizárraga-Partida, 2001; ganism grows well in axenic algae, rotifer and Artemia cultures (D'Alvise Nakase and Eguchi, 2007; Salvesen et al., 2000; Van Stappen, et al., 2012, 2013; Grotkjær et al., 2016). However, when each of the background bacterial strains were added in conjunction with the

Fig. 4. Counts of V. anguillarum in axenic Artemia. Gfp tagged cells of V. anguillarum inoculated at 104 CFU/ml alone (○), or in the presence of P. inhibens 17395 GMr (□), Fig. 6. Counts of V. anguillarum in Artemia with natural microbiota. V. anguillarum P. inhibens 17395 GMr and background strains separately (▲, ▼, ●, ■), or in the inoculated at 104 CFU/ml alone (Δ), or in the presence of P. inhibens 17395 GMr (■). presence of all strains (Δ). Points are average of two biological replicates and errorbars Points are average of two biological replicates and errorbars are standard deviation of are standard deviation of the mean. the mean. 68 T. Grotkjær et al. / Aquaculture 462 (2016) 64–69

References

Austin, B., Allen, D.A., 1982. Microbiology of laboratory-hatched brine shrimp (Artemia). Aquaculture 26, 369–383. Benavete, G., Gatesoupe, F., 1988. Bacteria associated with cultured rotifers and Artemia are detrimental to larval turbot, Scophthalmus maximus L. Aquac. Eng. 7, 289–293. Bergh, Ø., Naas, K., Harboe, T., 1994. Shift in the intestinal microflora of Atlantic halibut (Hippoglossus hippoglossus)larvaeduringfirst feeding. Can. J. Fish. Aquat. Sci. 0–4. Berland, B.R., Bonin, D.J., Maestrini, S.Y., 1970. Study of bacteria associated with marine algae in culture. Mar. Biol. 3, 68–76. Buchan, A., González, J.M., Moran, M.A., 2005. Overview of the marine Roseobacter lineage. Appl. Environ. Microbiol. 71, 5665–5677. Cole, J.J., 1982. Interactions between bacteria and algae in aquatic ecosystems. Annu. Rev. Ecol. Syst. 13, 291–314. Croxatto, A., Lauritz, J., Chen, C., Milton, D.L., 2007. Vibrio anguillarum colonization of rainbow trout integument requires a DNA locus involved in exopolysaccharide transport and biosynthesis. Environ. Microbiol. 9, 370–382. Fig. 7. Counts of V. anguillarum in Tetraselmis suecica with natural microbiota. D'Alvise, P.W., Lillebø, S., Prol-Garcia, M.J., Wergeland, H., Nielsen, K.F., Bergh, Ø., Gram, L., V. anguillarum inoculated at 104 CFU/ml alone (Δ), or in the presence of P. inhibens 2012. Phaeobacter gallaeciensis reduces Vibrio anguillarum in cultures of microalgae 17395 GMr (■). Points are average of two biological replicates and errorbars are and rotifers, and prevents vibriosis in cod larvae. PLoS One 7, e43996. standard deviation of the mean. D'Alvise, P., Lillebø, S., Wergeland, H., Gram, L., Bergh, Ø., 2013. Protection of cod larvae from vibriosis by Phaeobacter spp.: a comparison of strains and introduction times. Aquaculture 384-387, 82–86. Daume, S., 2006. The roles of bacteria and micro and macro algae in abalone aquaculture: a review. J. Shellfish Res. 25, 151–157. pathogen, its growth was slightly impaired. Despite the fact that this Defoirdt, T., Thanh, L.D., Van Delsen, B., De Schryver, P., Sorgeloos, P., Boon, N., Bossier, P., reduction in the number of vibrios was not significantly different from 2011. N-acylhomoserine lactone-degrading Bacillus strains isolated from aquaculture control cultures where V. anguillarum was colonizing Artemia alone, animals. Aquaculture 311, 258–260. these data may suggest that the natural microbiota present in the Gomez-Gil, B., Abreu-Grobois, F.A., Romero-Jarero, J., de los Herrera-Vega, M., 1994. Chemical disinfection of Artemia nauplii. J. World Aquacult. Soc. 25, 579–583. aquaculture environment impose some kind of restriction on growth Gram, L., Rasmussen, B.B., Wemheuer, B., Bernbom, N., Ng, Y.Y., Porsby, C.H., Breider, S., of pathogens. Interestingly, Bacillus sp. strain LT3 has previously been Brinkhoff, T., 2015. Tropodithietic acid producing Phaeobacter inhibens from the shown to have probiotic effects in Artemia cultures both affecting Roseobacter-clade have a natural environmental niche as surface colonizer in Danish harbors. Syst. Appl. Microbiol. 38, 483–493. the pathogen (V. campbellii)andtheinnateimmunesystemof Grossart, H., Simon, M., 2007. Interactions of planktonic algae and bacteria: effects on the brine shrimp (Niu et al., 2014). Ochrobactum sp. strains M13 algal growth and organic matter dynamics. Aquat. Microb. Ecol. 47, 163–176. and M15 have also been shown to have probiotic effects, yet not Grossart, H.-P., Levold, F., Allgaier, M., Simon, M., Brinkhoff, T., 2005. Marine diatom species harbour distinct bacterial communities. Environ. Microbiol. 7, 860–873. through antagonism of pathogens, but rather through degradation of Grotkjær, T., Bentzon-tilia, M., Dourala, N., Nielsen, K.F., Gram, L., Street, N., 2016. Isolation poly-β-hydroxybutyrate (Liu et al., 2010), which was not supplied in of TDA-producing Phaeobacter strains from sea bass larval rearing units and their pro- our experiments. The effect of the background microbiota was most biotic effect against pathogenic Vibrio spp. in Artemia cultures. Syst. Appl. Microbiol. 1–34. pronounced after 24 h of incubation and then declined and we attribute Guillard, R.R.L., Ryther, J.H., 1962. Studies of marine planktonic diatoms. Can. J. Microbiol. the slight reduction rather to an interspecies competition for nutrients 8, 229–239. and not to a direct bactericidal effect. The difference after 48 h is the Hjelm, M., Bergh, Ø., Riaza, A., Nielsen, J., Melchiorsen, J., Jensen, S., Duncan, H., Ahrens, P., Birkbeck, H., Gram, L., 2004a. Selection and identification of autochthonous potential same as the start inoculation at time 0 corresponding with the expected probiotic bacteria from turbot larvae (Scophthalmus maximus) rearing units. Syst. outcome derived from former studies (Liu et al., 2010; Niu et al., 2014). Appl. Microbiol. 27, 360–371. In the live feed cultures obtained from aquaculture units, both the Hjelm, M., Riaza, A., Formoso, F., Melchiorsen, J., Gram, L., 2004b. Seasonal incidence of autochthonous antagonistic Roseobacter spp. and Vibrionaceae strains in a turbot larva pathogen and the probiotic bacteria when added as mono-cultures (Scophthalmus maximus) rearing system. Appl. Environ. Microbiol. 70, 7288–7294. established themselves and grew well. This demonstrates that if patho- Høj, L., Bourne, D.G., Hall, M.R., 2009. Localization, abundance and community structure of gens are introduced into the live feed with no control measures, they are bacteria associated with Artemia: effects of nauplii enrichment and antimicrobial – able to grow rapidly to high densities and are a likely source of infection. treatment. Aquaculture 293, 278 285. Jensen, P., Kauffman, C., Fenical, W., 1996. High recovery of culturable bacteria from the P. inhibens antagonized the pathogen in these non-axenic systems and surfaces of marine algae. Mar. Biol. 1–7. numbers remained at the inoculum level. In previous studies (D'Alvise Liu, Y., De Schryver, P., Van Delsen, B., Maignien, L., Boon, N., Sorgeloos, P., Verstraete, W., et al., 2012; Grotkjær et al., 2016), we have seen a probiont mediated Bossier, P., Defoirdt, T., 2010. PHB-degrading bacteria isolated from the gastrointesti- nal tract of aquatic animals as protective actors against luminescent vibriosis. FEMS killing of V. anguillarum independently of the initial inoculum Microbiol. Ecol. 74, 196–204. (D'Alvise et al., 2012). It is possible that the natural microbiota had a López-Torres, M.A., Lizárraga-Partida, M.L., 2001. Bacteria isolated on TCBS media associ- protective effect on the pathogen and potential absorbed some of the ated with hatched Artemia cysts of commercial brands. Aquaculture 194, 11–20. Nakase, G., Eguchi, M., 2007. Analysis of bacterial communities in Nannochloropsis sp. antibacterial compound, tropodithietic acid, produced by the probiont. cultures used for larval fish production. Fish. Sci. 73, 543–549. In conclusion, we have demonstrated that the potential probiotic Niu, Y., Defoirdt, T., Baruah, K., Van de Wiele, T., Dong, S., Bossier, P., 2014. Bacillus sp. LT3 bacterium, P. inhibens, is capable of antagonizing V. anguillarum in sever- improves the survival of gnotobiotic brine shrimp (Artemia franciscana)larvae challenged with Vibrio campbellii by enhancing the innate immune response and by al live feed systems with natural or controlled microbiota. This is prom- decreasing the activity of shrimp-associated vibrios. Vet. Microbiol. 173, 279–288. ising as it brings the probiotic concept, and hence the reduction in Norqvist, A., Hagström, Å., Wolf-Watz, H., 1989. Protection of rainbow trout against antibiotic usage, a step forward. vibriosis and furunculosis by the use of attenuated strains of Vibrio anguillarum. Appl. Environ. Microbiol. 55, 1400–1405. Norqvist, A., Norrman, B., Wolf-Watz, H., 1990. Identification and characterization of a zinc metalloprotease associated with invasion by the fish pathogen Vibrio Acknowledgements anguillarum. Infect. Immun. 58, 3731–3736. Olafsen, J.a., 2001. Interactions between fish larvae and bacteria in marine aquaculture. Aquaculture 200, 223–247. The present study was financed by the Technical University of Pintado, J., Pérez-Lorenzo, M., Luna-González, A., Sotelo, C.G., Prol, M.J., Planas, M., 2010. Denmark (TG PhD grant), by a grant from The Danish Council for Monitoring of the bioencapsulation of a probiotic Phaeobacter strain in the rotifer Strategic Research | Programme Commission on Health, Food and Brachionus plicatilis using denaturing gradient gel electrophoresis. Aquaculture 302, 182–194. Welfare (12-132390) to LG, and by grant VKR023285 from the Villum Planas, M., Pérez-Lorenzo, M., Hjelm, M., Gram, L., Uglenes Fiksdal, I., Bergh, Ø., Pintado, J., Kann Rasmussen foundation to LG. The research leading to these results 2006. Probiotic effect in vivo of Roseobacter strain 27-4 against Vibrio (Listonella) has received funding from the European Community's Seventh anguillarum infections in turbot (Scophthalmus maximus L.) larvae. Aquaculture 255, 323–333. Framework Programme (FP7/2007-2013) under the grant agreement Porsby, C.H., Nielsen, K.F., Gram, L., 2008. Phaeobacter and Rugeria species of the n° FP7-262336-AQUAEXCEL. Roseobacter clade colonize separate niches in a Danish turbot (Scophthalmus T. Grotkjær et al. / Aquaculture 462 (2016) 64–69 69

maximus)-rearing farm and antagonize Vibrio anguillarum under different growth. Seyedsayamdost, M.R., Wang, R., Kolter, R., Clardy, J., 2014. Hybrid biosynthesis of Appl. Environ. Microbiol. 74, 7356–7364. Roseobacticides from algal and bacterial precursor molecules. J. Am. Chem. Soc. Prol-García, M., Pintado, J., 2013. Effectiveness of probiotic Phaeobacter bacteria grown 136, 15150–15153. in biofilters against Vibrio anguillarum infections in the rearing of turbot (Psetta Skjermo, J., Vadstein, O., 1993. Characterization of the bacterial flora of mass cultivated maxima) larvae. Mar. Biotechnol. 15, 726–738. Brachionus plicatilis. Hydrobiologia 255-256, 185–191. Reid, H.I., Treasurer, J.W., Adam, B., Birkbeck, T.H., 2009. Analysis of bacterial populations Sobecky, P.A., Mincer, T.J., Chang, M.C., Helinski, D.R., 1997. Plasmids isolated from in the gut of developing cod larvae and identification of Vibrio logei, Vibrio marine sediment microbial communities contain replication and incompatibility anguillarum and Vibrio splendidus as pathogens of cod larvae. Aquaculture 288, 36–43. regions unrelated to those of known plasmid groups. Appl. Environ. Microbiol. 63, Reitan, K.I., Rainuzzo, J.R., Øie, G., Olsen, Y., 1993. Nutritional effects of algal addition in 888–895. first-feeding of turbot (Scophthalmus maximus L.) larvae. Aquaculture 118, 257–275. Sonnenshein, E.C., Nielsen, K.F., D'Alvise, P., Porsby, C.H., Melchiorsen, J., Heilmann, J., Rhodes, G., Huys, G., Swings, J., McGann, P., Hiney, M., Smith, P., Pickup, R.W., 2000. Dis- Kalatzis, P.G., López-Pérez, M., Bunk, B., Spöer, C., Middelboe, M., Gram, L., 2016. tribution of oxytetracycline resistance plasmids between aeromonads in hospital Global occurrence and heterogenecity of the Roseobacter-clade species Ruegeria and aquaculture environments: implication of Tn1721 in dissemination of the tetra- mobilis. ISME J. (revision submitted). cycline resistance determinant Tet A. Appl. Environ. Microbiol. 66, 3883–3890. Tinh, N.T.N., Dierckens, K., Sorgeloos, P., Bossier, P., 2007. A review of the functionality of Salvesen, I., Reitan, K.I., Skjermo, J., Øie, G., 2000. Microbial environments in marine probiotics in the larviculture food chain. Mar. Biotechnol. 10, 1–12. larviculture: impacts of algal growth rates on the bacterial load in six microalgae. Van Stappen, G., 1996. Introduction, biology and ecology of Artemia. In: Lavens, P., Aquac. Int. 275–287. Sorgeloos, P. (Eds.), Manual on the Production and Use of Live Food for Aquaculture. Segev, E., Tellez, A., Vlamakis, H., Kolter, R., 2015. Morphological heterogeneity and at- FAO Tech. Pap. Vol. 295. tachment of Phaeobacter inhibens. PLoS One 10, e0141300. Walne, P., 1970. Studies on the food value of nineteen genera of algae to juvenile bivalves Seyedsayamdost, M.R., Carr, G., Kolter, R., Clardy, J., 2011a. Roseobacticides: small mole- of the genera Ostrea, Crassostrea, Mercenaria,andMytilis. Fish. Investig. 26, 1–62. cule modulators of an algal-bacterial symbiosis. J. Am. Chem. Soc. 133, 18343–18349. Yik Sung, Y., Van Damme, E.J.M., Sorgeloos, P., Bossier, P., 2007. Non-lethal heat shock Seyedsayamdost, M.R., Case, R.J., Kolter, R., Clardy, J., 2011b. The Jekyll-and-Hyde chemis- protects gnotobiotic Artemia franciscana larvae against virulent vibrios. Fish Shellfish try of Phaeobacter gallaeciensis. Nat. Chem. 3, 331–335. Immunol. 22, 318–326.

Paper 3

Vibrio anguillarum is genetically and phenotypically unaffected by long-term continuous exposure to the antibacterial compound tropodithietic acid.

crossmark

Vibrio anguillarum Is Genetically and Phenotypically Unaffected by Long-Term Continuous Exposure to the Antibacterial Compound Tropodithietic Acid Downloaded from

Bastian Barker Rasmussen,a Torben Grotkjær,a Paul W. D’Alvise,a* Guangliang Yin,b Faxing Zhang,b Boyke Bunk,c Cathrin Spröer,c Mikkel Bentzon-Tilia,a Lone Grama Department of Systems Biology, Technical University of Denmark, Kongens Lyngby, Denmarka; Beijing Genomics Institute Europe, Copenhagen, Denmarkb; Leibniz Institute, DSMZ-German Collection of Microorganisms and Cell Cultures, Braunschweig, Germanyc

ABSTRACT Minimizing the use of antibiotics in the food production chain is essential for limiting the development and spread of antibiotic- resistant bacteria. One alternative intervention strategy is the use of probiotic bacteria, and bacteria of the marine Roseobacter http://aem.asm.org/ clade are capable of antagonizing fish-pathogenic vibrios in fish larvae and live feed cultures for fish larvae. The antibacterial compound tropodithietic acid (TDA), an antiporter that disrupts the proton motive force, is key in the antibacterial activity of several roseobacters. Introducing probiotics on a larger scale requires understanding of any potential side effects of long-term exposure of the pathogen to the probionts or any compounds they produce. Here we exposed the fish pathogen Vibrio anguillarum to TDA for several hundred generations in an adaptive evolution experiment. No tolerance or resistance arose during the 90 days of exposure, and whole-genome sequencing of TDA-exposed lineages and clones revealed few mutational changes, compared to lineages grown without TDA. Amino acid-changing mutations were found in two to six different genes per clone; however, no mutations appeared unique to the TDA-exposed lineages or clones. None of the virulence genes of V. anguillarum was on July 30, 2016 by TECH KNOWLEDGE CTR OF DENMARK affected, and infectivity assays using fish cell lines indicated that the TDA-exposed lineages and clones were less invasive than the wild-type strain. Thus, long-term TDA exposure does not appear to result in TDA resistance and the physiology of V. anguillarum appears unaffected, supporting the application of TDA-producing roseobacters as probiotics in aquaculture.

IMPORTANCE It is important to limit the use of antibiotics in our food production, to reduce the risk of bacteria developing antibiotic resistance. We showed previously that marine bacteria of the Roseobacter clade can prevent or reduce bacterial diseases in fish larvae, acting as probiotics. Roseobacters produce the antimicrobial compound tropodithietic acid (TDA), and we were concerned regarding whether long-term exposure to this compound could induce resistance or affect the disease-causing ability of the fish pathogen. Therefore, we exposed the fish pathogen Vibrio anguillarum to increasing TDA concentrations over 3 months. We did not see the development of any resistance to TDA, and subsequent infection assays revealed that none of the TDA-exposed clones had increased virulence toward fish cells. Hence, this study supports the use of roseobacters as a non-risk-based disease control measure in aquaculture.

quaculture has been one of the fastest growing protein-pro- ter and Ruegeria, are especially promising as probiotics for marine Aducing sectors for decades, providing high-quality protein to larviculture (12–16). Phaeobacter and Ruegeria strains have re- the growing world population (1); however, a key constraint in the peatedly been isolated from aquaculture units and are harmless to aquaculture industry is microbial infectious diseases. More than fish larvae and their live feed (12–16). The probiotic (antibacte- 50% of such diseases are caused by bacteria, of which members of rial) effects of Phaeobacter and Ruegeria are predominantly caused the Vibrio genus are some of the most prevalent (2, 3). The species by their production of the dual-sulfur tropone-derived com- Vibrio anguillarum alone is known to infect more than 50 different fish species (4). Antibiotics are used in the treatment of bacterial infections in aquaculture, and they have also been used as prophy- Received 6 April 2016 Accepted 24 May 2016 lactic measures, especially in smaller aquaculture facilities and in Accepted manuscript posted online 27 May 2016 developing countries (5, 6). However, increases in antibiotic re- Citation Rasmussen BB, Grotkjær T, D’Alvise PW, Yin G, Zhang F, Bunk B, Spröer C, sistance and the associated health risks have led to a search for Bentzon-Tilia M, Gram L. 2016. Vibrio anguillarum is genetically and phenotypically alternative means of treatment (6). Vaccination of juvenile and unaffected by long-term continuous exposure to the antibacterial compound adult fish has reduced the use of antibiotics in the production of tropodithietic acid. Appl Environ Microbiol 82:4802–4810. doi:10.1128/AEM.01047-16. several finfish (7); however, it is not possible to vaccinate fish Editor: E. G. Dudley, Pennsylvania State University larvae, due to their immature immune systems, which makes Address correspondence to Lone Gram, [email protected]. them particularly vulnerable to infectious diseases (8). * Present address: Paul W. D’Alvise, Institute for Animal Science, University of The use of probiotic bacteria as sustainable controls for patho- Hohenheim, Stuttgart, Germany. genic bacteria in larviculture has been studied for several years Copyright © 2016, American Society for Microbiology. All Rights Reserved. (9–11), and two genera from the Roseobacter clade, i.e., Phaeobac-

4802 aem.asm.org Applied and Environmental Microbiology August 2016 Volume 82 Number 15 TDA Exposure of Vibrio anguillarum

pound tropodithietic acid (TDA) (16–19). TDA-producing Phae- serotype O1, isolated from diseased rainbow trout (Oncorhynchus mykiss) obacter and Ruegeria strains can prevent vibriosis in cod larvae from a Danish ﬁsh farm. The strain was chosen due to its high level of (16) and reduce mortality rates for turbot larvae (15) and Artemia pathogenicity for cod, halibut, and turbot larvae (A. Rønneseth, P. (20) challenged with V. anguillarum and V. harveyi, respectively. D’Alvise, Ø. Tønnesen, G. Haugland, T. Grotkjær, K. Engell-Sørensen, L. TDA is a broad-spectrum antibacterial that is bactericidal and Nørremark, Ø. Bergh, L. Gram, and H. I. Wergeland, unpublished data). inhibits the growth of several human and ﬁsh pathogens (16, 21– Strain 90-11-286 was grown in half-strength YTSS (½YTSS) (0.2% yeast extract, 0.125% tryptone, 2% Sigma sea salts) broth (41). Streaking and 23) but has very low toxicity in Caenorhabditis elegans and Artemia

plating were performed on ½YTSS-agar plates at 25°C. Tropodithietic Downloaded from (24). Recently, it was demonstrated that TDA is cytotoxic against acid (BioViotica, Dransfeld, Germany) was dissolved in dimethyl sulfox- N2a cells and OLN-93 cells of the mammalian nervous system and ide (DMSO) (1.5 mg/ml) by vortex-mixing for 10 s and kept at Ϫ80°C. several cancer cell lines, respectively (25, 26). Since TDA has pro- Daily working solutions were prepared from the TDA frozen stock solu- nounced antibacterial activity, the development of resistance to tion by diluting the solution in sterile ½YTSS medium by vortex-mixing TDA could be a concern, especially following long-term use. At- for 10 s, sonicating the mixture with ultrasound twice for 30 s, and vortex- tempts to select for TDA resistance and tolerance in pathogenic mixing for 10 s. Stock cultures of the wild-type strain and isolates from the bacteria have not been successful (22), suggesting that TDA resis- TDA-exposed lineage were stored at Ϫ80°C in a mixture of 4% (wt/vol) tance does not arise easily, likely due to a highly conserved target glycerol, 0.5% (wt/vol) glucose, 2% (wt/vol) skim milk powder, and 3% (or targets). Very recently, Wilson et al. demonstrated that cell (wt/vol) tryptone soy powder (42), whereas lineage populations were http://aem.asm.org/ Ϫ membranes and the proton motive force (PMF) were targets of stored at 80°C in 25% (wt/vol) glycerol. TDA, and TDA was suggested to function as an electroneutral Continuous-exposure experiment. Continuous exposure to TDA ϩ was carried out in an adaptive evolution experimental setup, with the aim proton antiporter, facilitating a one-to-one exchange of H for a of increasing TDA concentrations to induce resistance or tolerance. single charged metal ion and thus disrupting the transmembrane ⌬ Twenty microliters of a bacterial suspension of a single colony of V. an- proton gradient ( pH) (26). Also, genes conferring resistance to guillarum 90-11-286 was inoculated into 10 parallel 2-ml cultures. For 5 TDA were found in Phaeobacter inhibens, and Escherichia coli be- days, the lineages were diluted (1%) in fresh medium every 24 h. Seven of came resistant when harboring tdaR on a conjugative plasmid the lineages were subsequently diluted in ½YTSS medium supplemented (26). with TDA, and the remaining three were unexposed control lineages on July 30, 2016 by TECH KNOWLEDGE CTR OF DENMARK While those recent studies clearly indicated that resistance to (DMSO was added instead of TDA solution). After five dilutions at a TDA does not develop easily, e.g., by single point mutations, it is constant TDA concentration (exposed lineages), the concentration was not known how long-term exposure to TDA can affect target cells. increased. The starting concentration was 1/16 of the MIC, i.e., 0.8 ␮g Ϫ1 The long-term effects of other antibacterial compounds, such as ml , and this was increased to 1.75ϫ MIC during the course of the ␮ Ϫ1 ϫ antimicrobial peptides (AMPs), have been studied using adaptive experiment. The concentration steps were 0.8 gml (1/16 MIC), 3.1 ␮gmlϪ1 (1/4ϫ MIC), 6.25 ␮gmlϪ1 (1/2ϫ MIC), 12.5 ␮gmlϪ1 (1ϫ laboratory evolution (ALE) experiments (27–31), and we chose Ϫ Ϫ MIC), 15.6 ␮gml 1 (1.25ϫ MIC), 18.75 ␮gml 1 (1.5ϫ MIC), and 21.9 such an approach to evaluate the long-term exposure of vibrios to Ϫ ␮gml 1 (1.75ϫ MIC). Reinoculations were not performed for cultures TDA. Several studies have demonstrated that antibiotics can in- exposed to TDA concentrations above 1ϫ MIC if growth was not ob- duce mutations in exposed strains (32, 33) and affect gene expres- served. In those cases, the reinoculations were postponed until turbidity sion and phenotypes in exposed bacteria, e.g., inducing biofilm was observed. Aliquots of the 10 lineages were deposited at Ϫ80°C prior to formation or the expression of virulence genes in pathogens (32). an increase in the TDA concentration. If a lineage failed to grow at an Trimethoprim, which at high concentrations is lethal for Burk- increased TDA concentration, even after repeated reinoculations from its holderia thailandensis E264, functioned as a global activator of its last deposited stock, then it was terminated. The total numbers of passages secondary metabolism, activating at least five biosynthetic gene were 33 to 42 for the 10 lineages, which were equivalent to approximately clusters when added at subinhibitory concentrations (34). In sev- 220 to 280 generations, assuming six to seven generations per passage. At eral pathogenic bacteria, subinhibitory concentrations of antibi- the end of the experiment, all lineages were plated to allow isolation of individual clones from each lineage. Five colony isolates were randomly otics affect virulence genes and behavior (35–37); for example, ϫ ϫ subinhibitory concentrations of nafcillin were found to increase selected from each lineage population at 1.5 MIC to 1.75 MIC and were preserved as frozen stocks. toxin production in Staphylococcus aureus, whereas clindamycin Determination of MICs. MICs were determined using the microdilu- and linezolid decreased toxin production (38). Infection of fish by tion method (43).A50␮gmlϪ1 TDA stock solution in DMSO was pre- V. anguillarum depends on a series of virulence factors, including pared and 1:2 serial dilutions were made, giving a final concentration chemotaxis, motility, adhesion, and siderophore and exotoxin range of 25 to 0.4 ␮gmlϪ1 in the wells. A control DMSO dilution series production (4); therefore, in the present study, we also deter- was prepared accordingly. MIC testing was performed with three biolog- mined whether long-term exposure of V. anguillarum to TDA ical replicates. The bacteria (lineages or clones) were revived from frozen affected the infectivity of the bacteria, as measured in fish cell lines. cultures and grown overnight in ½YTSS medium at 20°C. Therefore, we exposed the fish pathogen V. anguillarum Stability of TDA tolerance. Cultures were revived from frozen stocks Ϫ1 strain 90-11-286 to sublethal concentrations of TDA over the in ½YTSS medium with 6.75 ␮gml TDA (one-half the wild-type MIC) course of 220 to 280 generations, in an adaptive laboratory at 20°C. The stability of the TDA tolerance was assessed by determining evolution experiment. Subsequently, we assessed phenotypes the MICs of bacterial populations as described above but with increased resolution. Four TDA-½YTSS working solutions were prepared, with 50 (resistance and infectivity) as well as genetic changes (single- Ϫ Ϫ Ϫ Ϫ ␮gml 1, 43.75 ␮gml 1, 37.5 ␮gml 1, and 31.25 ␮gml 1 TDA, and 1:2 nucleotide polymorphisms [SNPs]) and deletion-insertion serial dilutions were made from the working solutions to give a final range polymorphisms [DIPs]). of 25 to 4.7 ␮gmlϪ1 in the wells. The MIC assay results were read after 24 h and 72 h. MATERIALS AND METHODS Whole-genome sequencing. Twenty-eight isolates from TDA-ex- Bacterial strains and culture conditions. All adaptive evolution experi- posed lineages (7 lineages and 21 clones) and 12 isolates from control ments were performed with Vibrio anguillarum strain 90-11-286 (39, 40), lineages (3 lineages and 9 clones) were subjected to whole-genome se-

August 2016 Volume 82 Number 15 Applied and Environmental Microbiology aem.asm.org 4803 Rasmussen et al.

quencing. Genomic DNA was extracted from each using phenol-chloro- bation, when a complete confluent cell layer had formed, the cells were form-isoamyl alcohol and then was precipitated with isopropanol. Sam- washed three times with fresh Hanks’ balanced salt solution (HBSS) to ples were treated with RNase before quantification and quality analysis remove the antibiotic-antimycotic solution. TDA-exposed isolates were using a 1% agarose gel, a NanoDrop spectrophotometer (Saveen Werner, grown for 24 h. Subsequently, the medium was removed, bacteria were Sweden), and a Qubit 2.0 analyzer (Invitrogen, United Kingdom). washed with HBSS, and a bacterial suspension at an optical density at 600 Ϫ4 The genome of the ancestral wild-type V. anguillarum strain 90-11- nm (OD600) of 1 was prepared. Three milliliters of a 10 HBSS dilution 286 was sequenced using a combined sequencing approach with PacBio of the bacterial suspension was transferred into each of the wells, produc- ϫ 4 Ϫ1 and Illumina platforms, allowing the assembly of a fully closed high-qual- ing a bacterial cell density of approximately 3 10 CFU ml . After 24 h Downloaded from ity genome. A SMRTbell template library was prepared according to the of incubation at 20°C, the suspension was removed and the cells were instructions from Pacific Biosciences (Menlo Park, CA, USA), following trypsinized with 1 ml trypsin-EDTA. After 10 min, 250 ␮l FBS was added the procedure and checklist for 20-kb template preparation using the to inactivate the trypsin-EDTA, and the cells were dislodged. Cells were BluePippin size selection system. Briefly, for preparation of 15-kb librar- then washed three times in phosphate-buffered saline (PBS) (400 ϫ g for ies, 8 ␮g of genomic DNA was sheared using g-Tubes from Covaris 1 min at 5°C), and the pellet was resuspended in 500 ␮l PBS. Cell numbers (Woburn, MA, USA), according to the manufacturer’s instructions. DNA were determined in a 100-␮m Neubauer improved counting chamber was end repaired and ligated overnight to hairpin adapters by using com- (Assistent) and live/dead stained using a Cellstain double staining kit (Sig- ponents from DNA/polymerase binding kit P6 (Pacific Biosciences). Re- ma-Aldrich), as described by the manufacturer. To assess CHSE-214 kill- actions were carried out according to the manufacturer’s instructions. ing, microscopy of the live/dead-stained cells was performed at a magni- http://aem.asm.org/ ϫ BluePippin size selection to 7,000 kb was performed according to the fication of 100, using an Olympus BX51 fluorescence microscope. A manufacturer’s instructions (Sage Science, Beverly, MA, USA). Condi- counting chamber was used to define the area. All infection trials were tions for annealing of sequencing primers and binding of polymerase to performed in triplicate, assessing 70 to 400 cells per sample. ϭ ϭ the purified SMRTbell template were assessed with the calculator in RS Statistics. One-way analysis of variance (ANOVA) tests (n 3, df ␣ Յ Remote (Pacific Biosciences). SMRT sequencing was carried out with the 12, 0.05) were performed with the data obtained from each of the PacBio RS II system (Pacific Biosciences), taking one 240-min movie for three infection assays, to estimate the significance of any observed differ- each SMRT cell. In total, two SMRT cells were run and a total of 56,839 ences between the treated strains (TDA-exposed or control strains) and reads, with a mean read length of 13,006 bp, were obtained. SMRT cell the wild-type strain. Gaussian distributions were confirmed prior to the data were assembled using the RS_HGAP_Assembly.3 protocol included tests, and the analyses were performed using GraphPad Prism 6. on July 30, 2016 by TECH KNOWLEDGE CTR OF DENMARK in SMRT Portal version 2.3.0, using default parameters. The assembly Accession number(s). The complete genome of V. anguillarum strain revealed one circular chromosome. The validity of the assembly was 90-11-286 was deposited in NCBI GenBank under accession number checked using the RS_Bridgemapper.1 protocol. Each of the replicons was CP011460 for chromosome I and accession number CP011461 for chro- circularized independently, and artificial redundancies at the ends of the mosome II. The whole-genome shotgun project was deposited in NCBI SRA under accession number SRP072381. contigs were removed; the chromosome was adjusted to dnaA as the first gene. Subsequently, each genome was error corrected by mapping of Illu- RESULTS mina reads onto finished genomes using BWA software (44), with subse- Attempts to select for TDA-resistant V. anguillarum. The adap- quent variant and consensus calling using CLC Genomics Workbench 7 (Qiagen, Aarhus, Denmark). A consensus concordance of QV60 could tive evolution process was started at a TDA concentration of 1/16 ␮ Ϫ1 be confirmed for the genome. Finally, annotation was generated using times the wild-type MIC, which was measured to be 12.5 gml , Prokka 1.8 (45). and lineages were slowly taken to higher concentrations. After 33 Five-hundred-base-pair insertion libraries were prepared for all 40 to 42 reinoculations, corresponding to 220 to 280 generations, the lineages and clones. These libraries were subjected to 100-bp paired-end maximum concentrations of TDA allowing growth were 1.5ϫ Ϫ sequencing, using the Illumina HiSeq2000 platform, at Beijing Genomic MIC to 1.75ϫ MIC (18.8 to 21.9 ␮gml 1); it was not possible, Institute. The 40 isolates were mapped onto the V. anguillarum 90-11-286 despite repeated attempts, to grow the lineages at higher concen- reference genome. Subsequently, single-nucleotide polymorphisms trations. Four of the seven TDA-exposed lineages (lineages 2, 3, 4, (SNPs) and deletion-insertion polymorphisms (DIPs) with frequencies and 5) reached 1.75ϫ MIC, and three (lineages 1, 6, and 7) above 70% in the reads were detected using CLC Genomic Workbench reached 1.5ϫ MIC (Fig. 1). The growth rates clearly decreased at 8.0.2 (Qiagen). TDA concentrations higher than the wild-type MIC, and periods Motility observations. Several isolates had mutations in the fliM gene, of 2 to 3 days were required to reach visible turbidity, compared to which is important for flagellate motility in V. anguillarum. Therefore, motility was assessed by microscopy (ϫ1,000 magnification, using an 1 day at the MIC. Control lineages treated only with DMSO were Olympus BX51 microscope) of broth cultures grown in ½YTSS medium reinoculated 30 times, corresponding to approximately 210 gen- for 24 h at 25°C. erations. The MIC values of clones from the control lineages were Ϫ1 Fish cell line infection assay. The effects of the wild-type strain, eight 12.5 ␮gml TDA, corresponding to the wild-type MIC. TDA-exposed isolates, and four control isolates on the fish cell line CHSE- Stability of TDA tolerance. To determine whether the slight 214 (catalogue no. 91041114; Sigma) (46) were assessed in three indepen- TDA tolerance of the lineages was stable and whether clonal vari- dent cell infection assays. CHSE-214 cells were grown for 7 days at 20°C in ation occurred, the MICs of the lineages and individual clones Leibovitz L-15 medium with glutamine (Sigma-Aldrich), with 10% heat- were determined. Twenty-four-hour MIC assays with TDA con- inactivated (56°C for 30 min) fetal bovine serum (FBS) (Sigma-Aldrich) 2 centrations ranging from 2 times the wild-type MIC to 0.375 times and 1% antibiotic-antimycotic solution (Sigma-Aldrich), in a 75-cm the wild-type MIC revealed that all clones showed the same MIC, flask until cells reached an 80% confluent cell layer. After removal of the i.e., 12.5 ␮gmlϪ1, corresponding to the wild-type MIC. After an growth medium, the cells were washed with 5 ml Hanks’ balanced salt solution (HBSS) (catalogue no. H4641; Sigma). HBSS was removed, and additional 48 h, a total of 26 clones in addition to the wild-type the cells were trypsinized with 1 ml trypsin-EDTA (catalogue no. 59428C; strain were able to grow at slightly higher concentrations than the Sigma) and incubated for 10 min. Cells were dislodged and suspended in wild-type MIC (Table 1). The MIC was based on two separate 2.5 ml fresh medium. Two hundred microliters of CHSE-214 cell suspen- experiments, with three replicates per clone or lineage. sion (containing 1,000 to 10,000 cells mlϪ1) was transferred into each well Whole-genome sequencing and SNP/DIP analyses. The of a six-well test plate with 3 ml fresh medium. After 7 to 9 days of incu- closed genome of wild-type V. anguillarum strain 90-11-286 con-

4804 aem.asm.org Applied and Environmental Microbiology August 2016 Volume 82 Number 15 TDA Exposure of Vibrio anguillarum

Effects of fliM gene mutations on motility. The fliM gene is important for flagellate motility in V. anguillarum and potentially its virulence. Three different SNP mutations were found in the fliM gene and, since all lineages and clones harbored one of the mutations, the motility of the lineages and clones was assessed via phase-contrast microscopy. All clones and lineages were equally motile and similar to the wild-type strain (data not shown). Fish cell infection assays. Since antimicrobial compounds can Downloaded from affect the virulence of pathogenic bacteria, the ability of 12 clones (8 TDA-exposed clones and 4 control clones) to kill fish cells was tested in CHSE-214 infection assays. The 12 clones were selected to cover all seven lineages and the different mutational patterns. None of the clones resulted in significantly higher mortality rates for the fish cells than did the wild-type strain; in fact, most clones FIG 1 Attempt to select for TDA-tolerant or TDA-resistant V. anguillarum

strains in an adaptive laboratory evolution experiment. Solid lines, individual http://aem.asm.org/ lineages, of which 3 of 7 reached 1.5 times the wild-type MIC (18.75 ␮g/ml) and the rest reached 1.75 times the wild-type MIC (21.9 ␮g/ml). Dashed line, wild-type MIC (12.5 ␮g/ml TDA). TABLE 1 MICs of TDA after 24 h and 72 h for wild-type 90-11-286 strain and TDA-exposed (T) and DMSO-exposed (C) lineages and clones ␮ sists of two chromosomes, of 3.0 Mbp and 1.3 Mbp, with a GϩC TDA MIC ( g/ml) content of 44.4%, and it has 3,817 coding sequences (CDSs), Clone After 24 h After 72 h 4,003 genes, 31 rRNAs, and 106 tRNAs. This genome served as 90-11-286 12.5 15.6 a reference for mapping and identiﬁcation of single-nucleotide

C1 10.9–12.5 12.5 on July 30, 2016 by TECH KNOWLEDGE CTR OF DENMARK polymorphisms (SNPs) and deletion-insertion polymor- C1-1 12.5 15.6 phisms (DIPs) in the 40 genomes. All genomes had between C1-2 12.5 15.6 7.2 ϫ 106 and 8.4 ϫ 106 reads. The SNP and DIP analyses of the 40 C1-3 12.5 15.6 genomes revealed that 11 different genes were affected by 15 dif- C2 12.5 15.6 ferent mutations with frequencies above 70% in the reads (Table C2-1 12.5 15.6 2). The 15 mutations included 11 SNPs and 4 DIPs. Each genome C2-2 9.4–12.5 15.6 C2-3 12.5 15.6 had between two and eight amino acid-changing mutations, with C3 12.5 12.5 no distinct differences between TDA-exposed clones (0.007 to C3-1 12.5 12.5 Ϫ1 0.033 mutations generation ) and DMSO-exposed (control) C3-2 12.5 12.5–15.6 Ϫ1 clones (0.010 to 0.028 mutations generation ) being observed. C3-3 12.5 12.5 Clonal variations in mutations were seen within the populations T1 10.9–12.5 15.6 of control lineages C2 and C3 and the TDA-exposed lineage T7 T1-1 10.9–12.5 15.6 but not in the other lineages. Six of the 15 different mutations were T1-2 10.9–12.5 15.6 found only in the genomes of TDA-exposed clones and lineages, T1-3 10.9–12.5 15.6 whereas two were found only in the control genomes (Table 3). T2 10.9–12.5 15.6 Three different mutations were found in the competence gene T2-1 10.9–12.5 15.6 T2-2 10.9–12.5 15.6 comM, which encodes a magnesium chelatase-related protein. All T2-3 10.9–12.5 15.6 three comM mutations were found in genomes of 20 TDA-ex- T3 12.5 12.5 posed clones and lineages and six control clones and lineages. A T3-1 12.5 12.5–15.6 SNP in the kinE gene, encoding a two-component sensor histidine T3-2 12.5 12.5 kinase, was found in two TDA-exposed clones and eight control T3-3 12.5 12.5 clones and lineages. In lineage 1, a SNP was found in the dmlR T4 12.5 12.5–15.6 gene, which codes for a helix-turn-helix (HTH)-type transcrip- T4-1 12.5 12.5 tional regulator. In addition, a SNP was found in the transport T4-2 12.5 12.5 permease protein gene yadH. This mutation was present in lineage T4-3 12.5 12.5 2 and two clones from lineage 7. A SNP mutation was also found T5 10.9–12.5 15.6 T5-1 12.5 15.6 in all isolates from lineage 5 in 90-11-286_01842, which encodes a T5-2 12.5 12.5 2-oxoglutarate/Fe(II)-dependent oxygenase. The isolates from T5-3 12.5 12.5 lineage 6 had a SNP in the ompD gene, which encodes an outer T6 12.5 12.5 membrane porin protein. Furthermore, a DIP resulting in a T6-1 12.5 12.5–15.6 frameshift in the ftsI gene, which is involved in cell division, was T6-2 12.5 12.5 found in two clones from control lineage 3. Finally, the fliM gene, T6-3 10.9–12.5 15.6 which encodes a flagellar motor switch protein and thus is impor- T7 12.5 12.5 tant for motility and potentially virulence, was affected. In the fliM T7-1 12.5 15.6 gene, three different SNPs were found, with all clones and lineages T7-2 12.5 15.6 harboring one of the three different SNPs. T7-3 12.5 15.6

August 2016 Volume 82 Number 15 Applied and Environmental Microbiology aem.asm.org 4805 Rasmussen et al.

TABLE 2 Distribution of SNPs and DIPs in TDA-exposed lineages and clones of V. anguillaruma Control 1 Control 2 Control 3 Lineage 1b Lineage 2c Gene name and type of mutation Gene product C1 C1-1 C1-2 C1-3 C2 C2-1 C2-2 C2-3 C3 C3-1 C3-2 C3-3 T1 T1-1 T1-2 T1-3 T2 T2-1 T2-2 T2-3 fliM, SNP1 Flagellar motor switch protein ϩϩϩϩϩϩϩϩ fliM, SNP2 Flagellar motor switch protein ϩϩϩϩ fliM, SNP3 Flagellar motor switch protein ϩϩϩϩϩϩϩϩ —,d DIP Hypothetical protein ϩϩϩϩϩϩϩϩϩX ϩϩX ϩϩϩϩX ϩϩ comM, SNP1 Mg chelatase activity X ϩϩϩ ϩ ϩ ϩϩϩϩϩϩϩϩ Downloaded from comM, DIP1 Mg chelatase activity X ϩϩϩ ϩ ϩ ϩϩϩϩϩϩϩϩ comM, DIP2 Mg chelatase activity X ϩϩϩ ϩ ϩ ϩϩϩϩϩϩϩϩ kinE, SNP Two-component sensor histidine ϩϩϩϩϩϩϩϩ kinase —, SNP1 Hypothetical protein ϩϩϩϩϩϩϩ ϩ dmlR, SNP HTH-type transcriptional regulator ϩϩϩϩ yadH, SNP Transport permease protein ϩϩϩϩ 90-11-286_01842, 2-Oxoglutarate/Fe(II)-dependent SNP oxygenase ompD, SNP Outer membrane porin protein http://aem.asm.org/ ftsI, DIP Cell division protein ϩϩ —, SNP2 Hypothetical protein a Nonsynonymous SNPs and DIPs with frequencies of occurrence above 70% in the sequencing reads are included. X, frequency between 70% and Ͻ80%; ϩ, frequency between 80% and 100%. b Lineage that reached 1.5 times the wild-type MIC in the TDA exposure experiment. c Lineage that reached 1.75 times the wild-type MIC in the TDA exposure experiment. d —, hypothetical protein. appeared less infectious than the nonexposed wild-type strain antiporter, upsetting the proton motive force; also, TDA resis- on July 30, 2016 by TECH KNOWLEDGE CTR OF DENMARK (Fig. 2). The wild-type strain caused killing rates between 26% Ϯ tance in P. inhibens could be genetically linked and proved to be 8% and 39% Ϯ 10%, based on three independent experiments controlled by the tdaR genes (26), which conferred TDA resis- (Fig. 2). Four of the clones (C2-1, C1-1, T1-1, and T6-3) resulted tance in E. coli upon horizontal transfer. Tolerance and resistance in the same mortality rates as the wild-type strain, killing 15% Ϯ to TDA have also been found in environmental marine bacteria; 6%, 51% Ϯ 9%, 20% Ϯ 9%, and 26% Ϯ 5% of the fish cells, the majority (126 of 136 isolates) of non-TDA-producing bacteria respectively (P values of 0.106, 0.294, 0.086, and 0.268, respec- isolated along with a TDA-producing Pseudovibrio sp. displayed tively). The remaining eight clones caused lower mortality rates. tolerance to TDA (23). It is not known, however, whether this Two clones, C2-2 and T5-1, killed 13% Ϯ 4% and 20% Ϯ 8% of finding is related to the tdaR genes and/or whether these genes can the fish cells, respectively; these findings were just below the sig- be shared through horizontal gene transfer. In the present study, nificance level of 0.05, with both exhibiting P values of 0.044. The we were not able to adapt V. anguillarum to increasing TDA con- other six clones (C3-2, T2-2, T3-2, T4-3, T7-1, and T7-2) were centrations. The slight increase in tolerance, corresponding to 1.5 significantly less virulent than the wild-type strain (P values of to 1.75 times the wild-type MIC, was likely caused by the extended 0.002, 0.003, 0.022, 0.029, 0.001, and 0.002, respectively) and incubation time (2 to 3 days) and disappeared when clones were killed 4% Ϯ 2%, 3% Ϯ 6%, 17% Ϯ 1%, 18% Ϯ 4%, 3% Ϯ 1%, and tested in 24-h MIC assays. 5% Ϯ 1% of the fish cells, respectively. Very few variations were As mentioned, the proton motive force is the target of TDA seen between TDA-exposed and DMSO-exposed (control) (26). This function is of course essential and highly conserved, clones, and TDA-exposed clones did not show higher mortality which likely explains why resistance or tolerance does not develop, rates than DMSO-exposed clones. since mutational changes in the PMF apparatus would be lethal to the bacteria. It could be speculated that multiple cooccurring mu- DISCUSSION tations could result in resistance, as has been shown for antimi- The successful application of TDA-producing roseobacters as crobial peptides (30, 31). It was assumed that resistance to AMPs probiotics in aquaculture requires detailed knowledge about could not develop; however, several adaptive laboratory evolution the long-term effects of the roseobacters and TDA on the (ALE) experiments have demonstrated that long-term exposure pathogens and on the general microbiota. Here we focused on does indeed result in the evolution of resistant strains (27, 30). TDA exposure and demonstrated that continuous exposure of Using this method, the MIC of the AMP pexiganan for Pseudomo- V. anguillarum to TDA did not have any impact on the devel- nas fluorescens increased to 128 times the wild-type MIC after opment of TDA resistance or tolerance. Furthermore, pro- approximately 700 generations (27), and the MICs of the synthetic longed TDA exposure did not seem to increase the mutation AMP analogues ␣-peptide/␤-peptoid peptidomimetics against E. rates of exposed strains or the infectivity of the bacteria toward coli increased to 16 to 32 times the wild-type MIC after approxi- CHSE-214 fish cell lines. mately 500 generations (30). However, our ALE experiments did It has been known for some time that resistance to TDA does not result in TDA-resistant strains, and the experiment was ter- not arise easily in target pathogens, and single-point mutations do minated after two failed attempts to obtain growth in the presence not seem to confer resistance (22). However, since the mechanism of increased TDA concentrations. The experiment covered ap- of action was not known, assessing the risk of resistance was diffi- proximately 230 generations per lineage. Several antimicrobial cult. Wilson et al. recently demonstrated that TDA acts as a neutral compounds target the proton motive force, including food pre-

4806 aem.asm.org Applied and Environmental Microbiology August 2016 Volume 82 Number 15 TDA Exposure of Vibrio anguillarum

TABLE 2 (Continued) Lineage 3c Lineage 4c Lineage 5c Lineage 6b Lineage 7b

T3 T3-1 T3-2 T3-3 T4 T4-1 T4-2 T4-3 T5 T5-1 T5-2 T5-3 T6 T6-1 T6-2 T6-3 T7 T7-1 T7-2 T7-3

ϩϩϩϩ ϩϩ ϩϩϩϩϩϩϩϩ ϩϩϩϩ ϩϩ ϩϩϩϩϩϩϩϩX ϩϩϩXX ϩϩϩϩϩ

ϩϩϩϩ ϩϩϩϩ ϩϩϩϩ Downloaded from ϩϩϩϩ ϩϩϩϩ ϩϩϩϩ ϩϩϩϩ ϩϩϩϩ ϩϩϩϩ ϩϩ

ϩϩ ϩ X ϩϩϩϩϩϩ

ϩϩ ϩϩϩϩ

ϩϩϩϩ http://aem.asm.org/

ϩϩ

servatives such as sorbic acid and benzoic acid (47–49). These antiporter (57). The proposed model for TDA resistance in P. weak acids, when nondissociated, diffuse across the cell mem- inhibens relies on a GDAR-like system (26), and low pH com- brane and the cell envelope and, to maintain homeostasis and a bined with high glutamate availability could potentially induce

neutral intracellular pH, the bacteria use energy to pump out the a response in E. coli or other target bacteria, causing TDA tol- on July 30, 2016 by TECH KNOWLEDGE CTR OF DENMARK hydrogen ions, thus depleting the PMF (50, 51). The antimicro- erance. Furthermore, a similar response could possibly be bial effects of both compounds are dependent on environmen- achieved by a mutation causing continuous expression of the tal factors, especially pH, and the compounds are primarily LuxR-like regulator GadE, the main regulator of GDAR (58, effective under acidic conditions (52, 53). Similar to our obser- 59). Since access to glutamate would be a limiting factor, how- vations with TDA, bacterial resistance to sorbic acid and ben- ever, we consider it unlikely that TDA resistance would occur zoic acid does not seem to evolve easily (54, 55). To the best of via simple mutations or deletions. Even so, horizontal gene our knowledge, mutational resistance to sorbic acid and ben- transfer of TDA resistance genes could still be a risk. zoic acid in bacteria has not been reported, although several The bacteriocin nisin is a cationic peptide that interferes with cases have been described for yeast (54, 55). However, adaptive the PMF by forming pores in the cytoplasmic membrane, causing resistance to benzoic acid was observed in E. coli strain leakage of intracellular contents, loss of membrane potential, and O157:H7 after induction of the glutamate-dependent acid re- disruption of ⌬pH (60, 61). Four to 5 days of exposure to sub-MIC sistance (GDAR) response at pH 2.0 (56). The GDAR response levels of nisin resulted in a 32-fold MIC increase for S. aureus decarboxylates glutamate, replacing the ␣-carboxyl group with SH1000 subcultures; for one mutant, the increase was attributed a proton, and reduces glutamate to ␥-aminobutyric acid to two amino acid changes, in the purine operon repressor and a (GABA), which is then exchanged for glutamate via the GadC putative sensor histidine kinase (61). However, when the muta-

TABLE 3 Mutational changes of SNPs and DIPs in TDA-exposed lineages and clones of V. anguillarum No. of affected clones Mutation Amino acid Amino acid Gene name Gene product type position change Control Exposed fliM Flagellar motor switch protein SNP 48 Gly to Arg 8 0 fliM Flagellar motor switch protein SNP 54 Glu to Asp 0 10 fliM Flagellar motor switch protein SNP 230 Asp to Glu 4 18 —a Hypothetical protein DIP Frameshift 12 27 comM Mg chelatase activity SNP 1 Leu to Met 6 20 comM Mg chelatase activity DIP 2 Pro to frameshift 6 20 comM Mg chelatase activity DIP 306 Val to frameshift 6 20 kinE Two-component sensor histidine kinase SNP 66 Ile to Val 8 2 — Hypothetical protein SNP 3 15 dmlR HTH-type transcriptional regulator SNP 228 Lys to Arg 0 4 yadH Transport permease protein SNP 157 Gly to Val 0 6 90-11-286_01842 2-Oxoglutarate/Fe(II)-dependent oxygenase SNP 32 Ala to Glu 0 4 ompD Outer membrane porin protein SNP 272 Leu to Val 0 4 ftsI Cell division protein DIP 4 Asn to frameshift 2 0 — Hypothetical protein SNP 0 2 a —, hypothetical protein.

August 2016 Volume 82 Number 15 Applied and Environmental Microbiology aem.asm.org 4807 Rasmussen et al.

105 times after 4 days of incubation on selective plates (66), and ciprofloxacin and streptomycin increased the frequency of mutations causing rifampin resistance in Streptococcus pneumoniae (33). In the present study, no differences in mutation rates were seen between TDA-exposed clones and control strains; therefore, an increase in the mutation rate, resulting in potentially faster resistance development or changes in virulence genes, does not seem to be a concern in the application of TDA- Downloaded from producing probionts. Antimicrobial compounds can affect the virulence of bacteria after prolonged exposure to the compound. When different fish pathogens were exposed to the antimicrobial peptide cecropin B, V. anguillarum strain 775 displayed increased adhesion to CHSE- 214 fish cells and an increased mortality rate in challenged medaka (Oryzias latipes)(67). Induction of virulence genes by antimicrobial compounds has also been reported for human pathogens such http://aem.asm.org/ as Listeria monocytogenes (37, 68) and Staphylococcus aureus (35, 38). In the present study, none of the TDA-exposed clones was more infective toward CHSE-214 fish cells than were nonexposed clones and the wild-type strain. In contrast, 8 of 12 clones, including 6 TDA-exposed clones and 2 control clones, were significantly less infective than the wild-type strain. Three different mutations were found in the virulence-related flagellar motor switch protein gene fliM. Since TDA affects the PMF and thereby motility (26), on July 30, 2016 by TECH KNOWLEDGE CTR OF DENMARK the mutations in fliM are likely a result of TDA exposure. How- ever, no changes in motility could be observed in the mutants, and no clear patterns were seen for the mutations and the mortality rates in the fish cell trials. Since no differences in motility were observed and representatives from both the TDA-exposed and DMSO-exposed lineages resulted in fewer dead fish cells than did the wild type, we cannot conclude which of the factors (culturing or TDA exposure) had the strongest influence on the decreased infectivity. Somerville et al. (69) showed, for example, that the activity of aconitase, affecting the synthesis of several virulence factors and the expression of the global regulators RNA III and sarA, decreased 38% after 6 weeks of in vitro serial passage of a S. aureus strain isolated from a patient with toxic shock syndrome. Thus, it seems unlikely that TDA increases virulence postexpo- sure. Our study addressed the possible mutational changes caused by TDA; however, antimicrobials may also affect gene expression, FIG 2 Killing of CHSE-214 fish cell lines by TDA-exposed V. anguillarum clones representing different mutational patterns in three independent exper- and antibiotics at subinhibitory concentrations can activate “si- iments. All three experiments (A, B, and C) included the wild type and control. lent” gene clusters (34). Thus, studies addressing gene expression Each bar represents independent triplicates, and error bars are the standard in TDA-exposed pathogenic bacteria will be relevant in assess- deviation of the mean. ments of risks and applications. In conclusion, the present study demonstrates that prolonged TDA exposure does not result in TDA-resistant or TDA-tolerant tions’ contributions to resistance were evaluated via allelic re- strains. TDA exposure did not result in a changed mutational placements, it was possible to reach only a 16-fold MIC increase, pattern and did not enhance virulence, as measured in fish cell indicating that one or more additional unknown mutations con- lines. In fact, virulence appeared lower in TDA-exposed clones. tributed to the resistance (61). Thus, the development of resis- Thus, based on the present study, the use of TDA-producing ro- tance to a PMF-interfering compound is known; however, here seobacters would not likely pose risks of unwanted changes in the modification of the membrane binding target is a potential resis- pathogen. tance mechanism. Mutation rates in bacteria are influenced by several factors, ACKNOWLEDGMENTS such as nutritional availability and environmental stress (62). This study was funded by grants from the Danish Council for Strategic Several antibiotics have been reported to increase mutation Research Programme Commission on Health, Food, and Welfare (grant rates in exposed microorganisms (63–65), thus reducing the 12-132390; ProAqua), the European Union Seventh Framework Pro- time it takes for resistance, or other genetic changes, to occur. gramme (FP7/2007–2013; grant agreement 311975), and the Villum Tetracycline increased Pseudomonas aeruginosa mutation rates Kahn-Rasmussen Foundation (grant VKR023285).

4808 aem.asm.org Applied and Environmental Microbiology August 2016 Volume 82 Number 15 TDA Exposure of Vibrio anguillarum

FUNDING INFORMATION Bergh Ø, Gram L. 2012. Phaeobacter gallaeciensis reduces Vibrio anguil- This work, including the efforts of Bastian Barker Rasmussen, Torben larum in cultures of microalgae and rotifers, and prevents vibriosis in Grotkjær, Paul W. D’Alvise, Guangliang Yin, Faxing Zhang, and Lone cod larvae. PLoS One 7:e43996. http://dx.doi.org/10.1371/journal .pone.0043996. Gram, was funded by The Danish Council for Strategic Research | Pro- 17. Liang L. 2003. Investigation of secondary metabolites of North Sea bac- gramme Commission on Health, Food and Welfare (12-132390). This teria: fermentation, isolation, structure elucidation and bioactivity. Ph.D. work, including the efforts of Boyke Bunk, Cathrin Spröer, and Lone thesis. Georg August University, Göttingen, Germany. Gram, was funded by European Union Seventh Framework Programme 18. D’Alvise PW, Melchiorsen J, Porsby CH, Nielsen KF, Gram L. 2010.

(FP7) (311975). This work, including the efforts of Mikkel Bentzon-Tilia Inactivation of Vibrio anguillarum by attached and planktonic Roseobacter Downloaded from and Lone Gram, was funded by Villum Kahn-Rasmussen Foundation cells. Appl Environ Microbiol 76:2366–2370. http://dx.doi.org/10.1128 (VKR023285). /AEM.02717-09. 19. Prol García MJ, D’Alvise PW, Gram L. 2013. Disruption of cell-to-cell The funders had no role in study design, data collection and interpreta- signaling does not abolish the antagonism of Phaeobacter gallaeciensis to- tion, or the decision to submit the work for publication. ward the ﬁsh pathogen Vibrio anguillarum in algal systems. Appl Environ Microbiol 79:5414–5417. http://dx.doi.org/10.1128/AEM.01436-13. REFERENCES 20. Grotkjær T, Bentzon-Tilia M, D’Alvise P, Dourala N, Nielsen KF, Gram 1. Food and Agriculture Organization, Fisheries and Aquaculture Depart- L. 2016. Isolation of TDA-producing Phaeobacter strains from sea bass ment. 2014. The state of world ﬁsheries and aquaculture: opportunities larval rearing units and their probiotic effect against pathogenic Vibrio

and challenges. Food and Agriculture Organization of the United Nations, spp. in Artemia cultures. Syst Appl Microbiol 39:180–188. http://dx.doi http://aem.asm.org/ Rome, Italy. .org/10.1016/j.syapm.2016.01.005. 2. Kibenge FSB, Godoy MG, Fast M, Workenhe S, Kibenge MJT. 2012. 21. Brinkhoff T, Bach G, Heidorn T, Liang L, Schlingloff A, Simon M. 2004. Countermeasures against viral diseases of farmed fish. Antiviral Res 95: Antibiotic production by a Roseobacter clade-affiliated species from the 257–281. http://dx.doi.org/10.1016/j.antiviral.2012.06.003. German Wadden Sea and its antagonistic effects on indigenous isolates. 3. Thompson FL, Iida T, Swings J. 2004. Biodiversity of vibrios. Microbiol Appl Environ Microbiol 70:2560–2565. http://dx.doi.org/10.1128/AEM Mol Biol Rev 68:403–431. http://dx.doi.org/10.1128/MMBR.68.3.403 .70.4.2560-2565.2003. -431.2004. 22. Porsby CH, Webber MA, Nielsen KF, Piddock LJV, Gram L. 2011. 4. Frans I, Michiels CW, Bossier P, Willems KA, Lievens B, Rediers H. Resistance and tolerance to tropodithietic acid, an antimicrobial in aqua- 2011. Vibrio anguillarum as a fish pathogen: virulence factors, diagnosis culture, is hard to select. Antimicrob Agents Chemother 55:1332–1337. and prevention. J Fish Dis 34:643–661. http://dx.doi.org/10.1111/j.1365 http://dx.doi.org/10.1128/AAC.01222-10. -2761.2011.01279.x. 23. Harrington C, Reen FJ, Mooij MJ, Stewart FA, Chabot J-B, Guerra AF, on July 30, 2016 by TECH KNOWLEDGE CTR OF DENMARK 5. Cabello FC. 2006. Heavy use of prophylactic antibiotics in aquaculture: a Glöckner FO, Nielsen KF, Gram L, Dobson ADW, Adams C, O’Gara F. growing problem for human and animal health and for the environment. 2014. Characterisation of non-autoinducing tropodithietic acid (TDA) Environ Microbiol 8:1137–1144. http://dx.doi.org/10.1111/j.1462-2920 production from marine sponge Pseudovibrio species. Mar Drugs 12: .2006.01054.x. 5960–5978. http://dx.doi.org/10.3390/md12125960. 6. Defoirdt T, Sorgeloos P, Bossier P. 2011. Alternatives to antibiotics for 24. Neu AK, Månsson M, Gram L, Prol-García MJ. 2014. Toxicity of bio- the control of bacterial disease in aquaculture. Curr Opin Microbiol 14: active and probiotic marine bacteria and their secondary metabolites in 251–258. http://dx.doi.org/10.1016/j.mib.2011.03.004. Artemia sp. and Caenorhabditis elegans as eukaryotic model organisms. 7. Ringø E, Olsen RE, Jensen I, Romero J, Lauzon HL. 2014. Application Appl Environ Microbiol 80:146–153. http://dx.doi.org/10.1128/AEM of vaccines and dietary supplements in aquaculture: possibilities and chal- .02717-13. lenges. Rev Fish Biol Fish 24:1005–1032. http://dx.doi.org/10.1007 25. Wichmann H, Vocke F, Brinkhoff T, Simon M, Richter-Landsberg C. /s11160-014-9361-y. 2015. Cytotoxic effects of tropodithietic acid on mammalian clonal cell 8. Sommerset I, Krossøy B, Biering E, Frost P. 2005. Vaccines for fish in lines of neuronal and glial origin. Mar Drugs 13:7113–7123. http://dx.doi aquaculture. Expert Rev Vaccines 4:89–101. http://dx.doi.org/10.1586 .org/10.3390/md13127058. /14760584.4.1.89. 26. Wilson MZ, Wang R, Gitai Z, Seyedsayamdost MR. 2016. Mode of 9. Skjermo J, Vadstein O. 1999. Techniques for microbial control in the action and resistance studies unveil new roles for tropodithietic acid as an intensive rearing of marine larvae. Aquaculture 177:333–343. http://dx anticancer agent and the ␥-glutamyl cycle as a proton sink. Proc Natl Acad .doi.org/10.1016/S0044-8486(99)00096-4. Sci U S A 113:1630–1635. http://dx.doi.org/10.1073/pnas.1518034113. 10. Gatesoupe FJ. 1999. The use of probiotics in aquaculture. Aquaculture 27. Perron GG, Zasloff M, Bell G. 2006. Experimental evolution of resistance 180:147–165. http://dx.doi.org/10.1016/S0044-8486(99)00187-8. to an antimicrobial peptide. Proc R Soc B Biol Sci 273:251–256. http://dx 11. Kesarcodi-Watson A, Kaspar H, Lategan MJ, Gibson L. 2008. Probiotics in .doi.org/10.1098/rspb.2005.3301. aquaculture: the need, principles and mechanisms of action and screening 28. Friedman L, Alder JD, Silverman JA. 2006. Genetic changes that corre- processes. Aquaculture 274:1–14. http://dx.doi.org/10.1016/j.aquaculture late with reduced susceptibility to daptomycin in Staphylococcus aureus. .2007.11.019. Antimicrob Agents Chemother 50:2137–2145. http://dx.doi.org/10.1128 12. Ruiz-Ponte C, Cilia V, Lambert C, Nicolas JL. 1998. Roseobacter gallae- /AAC.00039-06. ciensis sp. nov., a new marine bacterium isolated from rearings and collec- 29. Toprak E, Veres A, Michel J, Chait R, Hartl DL, Kishony R. 2012. tors of the scallop Pecten maximus. Int J Syst Bacteriol 48:537–542. http: Evolutionary paths to antibiotic resistance under dynamically sustained //dx.doi.org/10.1099/00207713-48-2-537. drug selection. Nat Genet 44:101–105. http://dx.doi.org/10.1038/ng.1034. 13. Hjelm M, Bergh O, Riaza A, Nielsen J, Melchiorsen J, Jensen S, Duncan 30. Hein-Kristensen L, Franzyk H, Holch A, Gram L. 2013. Adaptive evo- H, Ahrens P, Birkbeck H, Gram L. 2004. Selection and identification of lution of Escherichia coli to an ␣-peptide/␤-peptoid peptidomimetic in- autochthonous potential probiotic bacteria from turbot larvae (Scophthal- duces stable resistance. PLoS One 8:e73620. http://dx.doi.org/10.1371 mus maximus) rearing units. Syst Appl Microbiol 27:360–371. http://dx /journal.pone.0073620. .doi.org/10.1078/0723-2020-00256. 31. Jochumsen N. 2013. The causes and consequences of antibiotic resistance 14. Hjelm M, Riaza A, Formoso F, Gram L, Melchiorsen J. 2004. Seasonal evolution in microbial pathogens. Ph.D. thesis. Technical University of incidence of autochthonous antagonistic Roseobacter spp. and Vibrion- Denmark, Kongens Lyngby, Denmark. aceae strains in a turbot larva (Scophthalmus maximus) rearing system. 32. Romero D, Traxler MF, López D, Kolter R. 2011. Antibiotics as signal Appl Environ Microbiol 70:7288–7294. http://dx.doi.org/10.1128/AEM molecules. Chem Rev 111:5492–5505. http://dx.doi.org/10.1021 .70.12.7288-7294.2004. /cr2000509. 15. Planas M, Pérez-Lorenzo M, Hjelm M, Gram L, Uglenes Fiksdal I, 33. Henderson-Begg SK. 2006. Effect of subinhibitory concentrations of an- Bergh Ø, Pintado J. 2006. Probiotic effect in vivo of Roseobacter strain tibiotics on mutation frequency in Streptococcus pneumoniae. J Antimi- 27-4 against Vibrio (Listonella) anguillarum infections in turbot (Scoph- crob Chemother 57:849–854. http://dx.doi.org/10.1093/jac/dkl064. thalmus maximus L.) larvae. Aquaculture 255:323–333. http://dx.doi.org 34. Seyedsayamdost MR. 2014. High-throughput platform for the discovery /10.1016/j.aquaculture.2005.11.039. of elicitors of silent bacterial gene clusters. Proc Natl Acad Sci U S A 16. D’Alvise PW, Lillebø S, Prol-Garcia MJ, Wergeland HI, Nielsen KF, 111:7266–7271. http://dx.doi.org/10.1073/pnas.1400019111.

August 2016 Volume 82 Number 15 Applied and Environmental Microbiology aem.asm.org 4809 Rasmussen et al.

35. Subrt N, Mesak LR, Davies J. 2011. Modulation of virulence gene ex- 54. Russell AD. 1991. Mechanisms of bacterial resistance to non- pression by cell wall active antibiotics in Staphylococcus aureus. J Antimi- antibiotics: food additives and food and pharmaceutical preservatives. crob Chemother 66:979–984. http://dx.doi.org/10.1093/jac/dkr043. J Appl Bacteriol 71:191–201. http://dx.doi.org/10.1111/j.1365-2672 36. Yim G, De La Cruz F, Spiegelman GB, Davies J. 2006. Transcription .1991.tb04447.x. modulation of Salmonella enterica serovar Typhimurium promoters by 55. Beales N. 2004. Adaptation of microorganisms to cold temperatures, sub-MIC levels of rifampin. J Bacteriol 188:7988–7991. http://dx.doi.org weak acid preservatives, low pH, and osmotic stress: a review. Compr Rev /10.1128/JB.00791-06. Food Sci Food Saf 3:1–20. http://dx.doi.org/10.1111/j.1541-4337.2004 37. Knudsen GM, Holch A, Gram L. 2012. Subinhibitory concentrations of .tb00057.x.

antibiotics affect stress and virulence gene expression in Listeria monocy- 56. Lin J, Smith MP, Chapin KC, Baik HS, Bennett GN, Foster JW. 1996. Downloaded from togenes and cause enhanced stress sensitivity but do not affect Caco-2 cell Mechanisms of acid resistance in enterohemorrhagic Escherichia coli. Appl invasion. J Appl Microbiol 113:1273–1286. http://dx.doi.org/10.1111/j Environ Microbiol 62:3094–3100. .1365-2672.2012.05435.x. 57. Foster JW. 2004. Escherichia coli acid resistance: tales of an amateur aci- 38. Stevens DL, Ma Y, Salmi DB, McIndoo E, Wallace RJ, Bryant AE. 2007. dophile. Nat Rev Microbiol 2:898–907. http://dx.doi.org/10.1038 Impact of antibiotics on expression of virulence-associated exotoxin genes /nrmicro1021. in methicillin-sensitive and methicillin-resistant Staphylococcus aureus.J 58. Ma Z, Gong S, Richard H, Tucker DL, Conway T, Foster JW. 2003. Infect Dis 195:202–211. http://dx.doi.org/10.1086/510396. GadE (YhiE) activates glutamate decarboxylase-dependent acid resistance 39. Pedersen K, Larsen J. 1993. rRNA gene restriction patterns of Vibrio in Escherichia coli K-12. Mol Microbiol 49:1309–1320. http://dx.doi.org anguillarum serogroup O1. Dis Aquat Organ 16:121–126. http://dx.doi /10.1046/j.1365-2958.2003.03633.x. .org/10.3354/dao016121. 59. Hommais F, Krin E, Coppée JY, Lacroix C, Yeramian E, Danchin A, http://aem.asm.org/ 40. Skov MN, Pedersen K, Larsen JL. 1995. Comparison of pulsed-field gel Bertin P. 2004. GadE (YhiE): a novel activator involved in the response to electrophoresis, ribotyping, and plasmid profiling for typing of Vibrio an- acid environment in Escherichia coli. Microbiology 150:61–72. http://dx guillarum serovar O1. Appl Environ Microbiol 61:1540–1545. .doi.org/10.1099/mic.0.26659-0. 41. Sobecky PA, Mincer TJ, Chang MC, Helinski DR. 1997. Plasmids 60. Bruno MEC, Kaiser A, Montville TJ. 1992. Depletion of proton motive isolated from marine sediment microbial communities contain replica- force by nisin in Listeria monocytogenes cells. Appl Environ Microbiol tion and incompatibility regions unrelated to those of known plasmid 58:2255–2259. groups. Appl Environ Microbiol 63:888–895. 61. Blake KL, Randall CP, Neill AJO. 2011. In vitro studies indicate a high 42. Gibson LF, Khoury JT. 1986. Storage and survival of bacteria by ultra- resistance potential for the lantibiotic nisin in Staphylococcus aureus and freeze. Lett Appl Microbiol 3:127–129. http://dx.doi.org/10.1111/j.1472 define a genetic basis for nisin resistance. Antimicrob Agents Chemother -765X.1986.tb01565.x.

55:2362–2368. http://dx.doi.org/10.1128/AAC.01077-10. on July 30, 2016 by TECH KNOWLEDGE CTR OF DENMARK 43. Clinical and Laboratory Standards Institute. 2006. Methods for dilution 62. Martinez JL, Baquero F. 2000. Mutation frequencies and antibiotic re- antimicrobial susceptibility tests for bacteria that grow aerobically; ap- sistance. Antimicrob Agents Chemother 44:1771–1777. http://dx.doi.org proved standard—7th ed. Publication M07-A7. Clinical and Laboratory /10.1128/AAC.44.7.1771-1777.2000. Standards Institute, Wayne, PA. 44. Li H, Durbin R. 2009. Fast and accurate short read alignment with Bur- 63. Kohanski MA, DePristo MA, Collins JJ. 2010. Sublethal antibiotic treat- rows-Wheeler transform. Bioinformatics 25:1754–1760. http://dx.doi ment leads to multidrug resistance via radical-induced mutagenesis. Mol .org/10.1093/bioinformatics/btp324. Cell 37:311–320. http://dx.doi.org/10.1016/j.molcel.2010.01.003. 45. Seemann T. 2014. Prokka: rapid prokaryotic genome annotation. Bioinfor- 64. Baharoglu Z, Mazel D. 2011. Vibrio cholerae triggers SOS and mutagen- matics 30:2068–2069. http://dx.doi.org/10.1093/bioinformatics/btu153. esis in response to a wide range of antibiotics: a route towards multiresis- 46. Lannan CN, Winton JR, Fryer JL. 1984. Fish cell lines: establishment and tance. Antimicrob Agents Chemother 55:2438–2441. http://dx.doi.org/10 characterization of nine cell lines from salmonids. In Vitro 20:671–676. .1128/AAC.01549-10. http://dx.doi.org/10.1007/BF02618871. 65. Gutierrez A, Laureti L, Crussard S, Abida H, Rodríguez-Rojas A, 47. Eklund T. 1985. The effect of sorbic acid and esters of p-hydroxybenzoic Blázquez J, Baharoglu Z, Mazel D, Darfeuille F, Vogel J, Matic I. 2013. ␤ acid on the protonmotive force in Escherichia coli membrane vesicles. J -Lactam antibiotics promote bacterial mutagenesis via an RpoS- Gen Microbiol 131:73–76. mediated reduction in replication fidelity. Nat Commun 4:1610. http://dx 48. Sofos JN, Pierson MD, Blocher JC, Busta FF. 1986. Mode of action of .doi.org/10.1038/ncomms2607. sorbic acid on bacterial cells and spores. Int J Food Microbiol 3:1–17. 66. Alonso A, Campanario E, Martínez JL. 1999. Emergence of multidrug- http://dx.doi.org/10.1016/0168-1605(86)90036-X. resistant mutants is increased under antibiotic selective pressure in Pseu- 49. Krebs HA, Wiggins D, Stubbs M, Sols A, Bedoya F. 1983. Studies on the domonas aeruginosa. Microbiology 145:2857–2862. http://dx.doi.org/10 mechanism of the antifungal action of benzoate. Biochem J 214:657–663. .1099/00221287-145-10-2857. http://dx.doi.org/10.1042/bj2140657. 67. Sallum UW, Chen TT. 2008. Inducible resistance of fish bacterial patho- 50. Salmond CV, Kroll RG, Booth IR. 1984. The effect of food preservatives gens to the antimicrobial peptide cecropin B. Antimicrob Agents Che- on pH homeostasis in Escherichia coli. J Gen Microbiol 130:2845–2850. mother 52:3006–3012. http://dx.doi.org/10.1128/AAC.00023-08. 51. Warth AD. 1985. Resistance of yeast species to benzoic and sorbic acids 68. Kastbjerg VG, Larsen MH, Gram L, Ingmer H. 2010. Influence of and to sulfur dioxide. J Food Prot 48:564–569. sublethal concentrations of common disinfectants on expression of viru- 52. Eklund T. 1983. The antimicrobial effect of dissociated and undissociated lence genes in Listeria monocytogenes. Appl Environ Microbiol 76:303– sorbic acid at different pH levels. J Appl Bacteriol 54:383–389. http://dx 309. http://dx.doi.org/10.1128/AEM.00925-09. .doi.org/10.1111/j.1365-2672.1983.tb02632.x. 69. Somerville GA, Beres SB, Fitzgerald JR, DeLeo FR, Cole RL, Hoff JS, 53. Eklund T. 1985. Inhibition of microbial growth at different pH levels Musser JM. 2002. In vitro serial passage of Staphylococcus aureus: changes by benzoic and propionic acids and esters of p-hydroxybenzoic acid. in physiology, virulence factor production, and agr nucleotide sequence. J Int J Food Microbiol 2:159–167. http://dx.doi.org/10.1016/0168 Bacteriol 184:1430–1437. http://dx.doi.org/10.1128/JB.184.5.1430-1437 -1605(85)90035-2. .2002.

4810 aem.asm.org Applied and Environmental Microbiology August 2016 Volume 82 Number 15