Green and red macroalgae extracts show antibacterial effects and induce innate immune responses in Nile tilapia and rainbow trout in vitro

Highlights • Marine sulphated polysaccharides (MSP) have direct antibacterial effects in vitro.• The antibacterial effect of MSP is dependent on seaweed species and extraction.• MSP-rich extracts modulate fish immune responses.• Effects of MSP-rich extracts on head kidney leukocytes appear fish species specific.• Differences exist between seaweed species in MSP-driven immune modulation.


Introduction
To keep up with an increasing demand for animal protein, the aquaculture sector has seen an increase in production over the past decades, among others due to intensification of the sector.Intensification of the aquaculture sector has also been accompanied by an increase in disease incidences.Besides prevention of diseases through vaccination, and treatment of diseases by antibiotics, dietary supplementation with immunomodulators may provide an alternative route to maintain animal health in aquaculture.Among the better-known immunomodulatory feed additives are non-starch polysaccharides, with β-glucans probably being the best characterized compounds [as reviewed by: [1,2]].Another group of polysaccharides with potentially immunomodulatory effects are marine sulphated polysaccharides (MSPs) isolated from macroalgae.While MSP-rich extracts containing health-promoting compounds are gaining interest from the aquaculture sector, their immunomodulatory effects are not always clearly defined.
Cell wall matrices of marine algae commonly contain highly complex sulphated polysaccharides.Polysaccharides extracted from green macroalgae are referred to as ulvans, while fucoidans are extracted from brown algae and carrageenans from red algae.Several studies have shown such polysaccharide compounds to exhibit a wide range of biological activities, including immunomodulatory properties, mostly in mammalian studies (Reviewed in: [3]).Desulphation is generally accepted to decrease their bioactivity, indicating sulphate residues are important for their activity [4].Among the better studied are the ulvans; water-soluble sulphated polysaccharides composed of disaccharide repetition moieties made up of sulphated rhamnose linked to either glucuronic acid, iduronic acid or xylose.Stimulation of immune cells with ulvans has been shown to trigger events leading to the activation of the immune system or the control of the inflammation process [5,6].Mammalian literature shows that ulvans are able to enhance macrophage functions such as phagocytosis and induce production of reactive oxygen species (ROS) and nitric oxide (NO), and lead to the secretion of pro-inflammatory cytokines such as tumour necrosis factor (TNFα), interleukins IL-1, IL-6, IL-8, IL-12, and interferon (IFN) [7][8][9][10].Interestingly, incubation of murine (RAW) macrophages with U. rigida polysaccharides induced an increase in nitrite production which could be decreased considerably by desulphation, arguing for an important role of the sulphate groups in the induction of this immune activation by U. rigida derived polysaccharides [11].Further, ulvans, carrageenans and fucoidans have been shown to regulate innate immune responses through Toll-like receptors such as TLR4 and TLR2 expressed on target cells, inducing signalling pathways that lead to the activation of the transcription factor NF-kappa B [12][13][14][15][16]. Taken together, immunomodulatory effects of marine sulphated polysaccharides, or MSP, on mammalian and chicken cells, at least in vitro, are well established.
Previously, the Olmix Group investigated the potential of marine macroalgae harvested on the shore of France to modulate the immune response of porcine intestinal epithelial cells (IPEC-1), or human embryonic kidney cells (HEK293) [12,17].Aqueous MSP-rich extracts prepared from Ulva sp. were evaluated for antibacterial and immunomodulatory activities in vitro.MSP-rich extracts were shown to directly inhibit the growth of pathogenic bacteria and stimulate gene expression of the cytokines IL-1, IL-6, IL-8, TNFα, TGFβ and chemokine CCL20 [17].Further, MSP-rich extracts were used to shed light on the molecular mechanisms underlying their biological activity, evaluating a panel of pattern recognition receptors expressed on HEK cells to identify the target receptor and downstream cytokine expression.MSP-rich extracts prepared from Ulva sp.primarily stimulated TLR4, and Western blot analyses of MSP-treated HEK293-TLR4 cells showed an increase in the phosphorylation of Akt and the p65 subunit of nuclear factor-κB.Inhibition of Akt phosphorylation with a specific inhibitor abrogated MSP-mediated enhancement of IL-8 secretion [12].It was thus concluded that MSP-rich extracts prepared from Ulva sp. could represent TLR ligands with immunomodulatory properties.It is important to take a comparable approach to the understanding of the action and mechanisms of MSP-mediated immunomodulatory activities in fish cells prior to their use as health-improving compounds for future preventive strategies in aquaculture.
For aquaculture, application of fucoidans as immunomodulating compounds has recently been reviewed, including in vivo effects on growth, intestinal health, antioxidant capacity, and immune responses of several finfish and shellfish species [18].However up to date, there are not many in vitro studies in fish that aim to provide detailed information on potential working mechanisms of MSPs.In early studies with turbot (Psetta maxima), protein-free water-soluble extracts and polysaccharide fractions obtained from the green seaweed Ulva rigida and the red seaweed Chondrus crispus, these extracts could induce an increase in the respiratory burst activity of turbot phagocytes [19].Stimulatory capacities of the different extracts varied depending on origin, concentration and time of incubation.In a follow-up study, the authors separated water-soluble U. rigida polysaccharides into uronic acids, rhamnose and xylose and found that especially, the acid sugar fractions with large homogeneous sulphated polysaccharides induced respiratory burst activity [20].Of interest, this capacity was lost after desulphation, while recovered after non-specific resulphation, suggesting an important function of the sulphate groups for immunomodulation [20].More recently, sulphated polysaccharides from Codium fragile (sponge seaweed) were shown to exhibit not only direct anti-bacterial activities against specific fish pathogenic bacteria but also to have immunostimulatory effects in vitro on head kidney and spleen leukocytes from rockfish (Sebastes schlegelii), using induction of gene expression of the cytokines tnfα, il1b, il6, il8 and il10 as read-out [21].Together these studies suggest it could be informative to perform in vitro studies on the effects of MSP-rich extracts on cell-based mechanisms determining innate immune activity, in different fish species.
In the current study, we first investigated the effects of different crude extracts from green macroalgae (Ulva sp.) or red macroalgae (Solieria sp.) rich in MSPs for direct antibacterial effects in vitro.Further, head kidney leukocytes (HKL) isolated from Nile tilapia (Oreochromis niloticus) and from rainbow trout (Oncorhynchus mykiss) were stimulated with different MSP-rich extracts, using induction of reactive oxygen species (ROS) and gene expression as immunomodulatory read-outs.We discuss the results with regard to differences related to the source of MSP (Ulva or Solieria macroalgae), extraction method, suitability of immunomodulatory read-outs (ROS, gene expression), possible cellular signalling pathway (TLR, CLR) and fish species (Nile tilapia and rainbow trout).

Production of MSP-rich algae extracts
Green tide algae Ulva sp. were collected on beaches in Brittany, France at Plestin les Grèves (48 The biomass of the bulk samples analysed contained a mixture of two foliose Ulva species, predominantly Ulva lacinulata (Kützing) (previously known as U. laetevirens (Areshoug) or U. armoricana; (Dion, Reviers & Coat)), [22], and a lower amount of Ulva sp.A (previously known as Ulva rigida; [23].Solier's red string weed (Solieria chordalis; [24] was collected on a beach in Vendée, France at Saint-Hilaire de Riez (46 • 42′52''N 1 • 58′46''W).The algae were washed in fresh water, drained, and deep frozen.For extraction, the algae were thawed; wet ground, and liquid and solid phases were separated by band press as part of an industrial process (Patent No. FR 61909).The liquid was fractionated by tangential filtration (Tami Industries).The solid fractions were further processed using different methods (Table 1).
The percentage elemental analysis (C, N, S) elementar Analysensysteme GmbH-vario EL III Element Analyzer (Hanau, Germany) was applied according to the manufacturer's recommendations.Neutral sugars were determined using the phenol sulfuric acid method with anhydrous d-glucose used as standard (0-100 μg mL − 1 ) [25]).Uronic acids were determined using the meta-hydroxydiphenyl method [26].The amount of sulfate groups present on the polysaccharides were determined by the Azure A method (Sigma-Aldrich, France) using 17 % of sulfated dextran as standard in the range of 0 to 100 μg mL − 1 [27].
The molar ratio of monosaccharides was determined after acidic methanolysis of the MSP sample using 3 mL − 1 of MeOH/HCl for 4 h at 100 followed by gas chromatography (GC) analysis of the trimethylsilyl derivatives using the method of Montreuil et al. [28].Myoinositol (1 mg mL − 1 ) was used as the internal reference, nitrogen as a carrier gas and a non-polar column, type HP-5MS (30 m, 0.25 mm internal diameter) for analysis.The temperature profile was: 120 • C for 1 min, followed by a gradient of 1.5 • C min − 1 between 120 and 180 • C and 2 • C min − 1 between 180 and 200 • C. Because of the high content of salt in the sample interfering with the reproducibility of the analyses, salts were eliminated by ultra-filtration using a 1-kDa cut-off Pall minimate membrane (Pall Corporation, France) and freeze-dried.Protein content was quantified by the bicinchoninic acid colorimetric method (BCA) using the Micro BC Assay Kit (Interchim, France) [29].The molecular weight distribution of the sample (0.5 mg mL − 1 ) was analyzed by size exclusion chromatography in 0.1 M sodium nitrate with 0.2 % sodium azide at a flow rate of 0.5 mL min − 1 .The Shodex 802 and 803 columns in series and a multi-angle light scattering refractometer (Wyatt, 18 angles) for detection were used with a dn/dc of 0.150 mL g − 1 .The different algal extracts were dissolved under sterile conditions by gentle agitation for 2 h at 37 • C in D-PBS (Gibco) and pasteurized for 20 min at 80 • C and stored at 4 • C until used.

Antibacterial effects of MSP-rich algae extracts
Antimicrobial activity of the MSP-rich extracts was investigated against 13 fish bacterial strains (Table 2).Bacterial strains were identified by and obtained from the National Reference Laboratory for Antimicrobial Resistance and for Fish Diseases, Lelystad, the Netherlands (Wageningen Bioveterinary Research, WBVR).Bacteria were cultured overnight at 22 • C on heart infusion agar with 5 % sheep blood (HIS)-plates (Aeromonas hydrophila, Edwardsiella tarda, Photobactrium damselae, Aeromonas salmonicida salmonicida, Edwardsiella ictaluri, Vibrio anguillarum, Yersina ruckeri, Aeromonas veronii, Shewanella putrefaciens and Vibrio vulnificus), for four days at 22 • C on HIS-plates (Streptococcus iniae and Streptococcus agalactiae) or for four days at 15 • C on Marine agar plates (Aliivibrio salmonicida).Subsequently, individual colonies of each strain were suspended in sterile physiological saline solution (sodium chloride 0.85 %) and bacterial suspension density was adjusted to equal to 0.5 McFarland standard.Next, concentration of suspension equal to 0.5 McFarland was determined per strain by performing a ten-times dilution series (10 − 1 -10 − 10 ) in triplicate.Dilution was performed on HIS-plates and cultured overnight at 22 • C (A. hydrophila, E. tarda, P. damselae, A. salmonicida salmonicida, E. ictaluri, V. anguillarum, Y. ruckeri, A. veronii, S. putrefaciens and V. vulnificus), on HIS-plates and cultured for four days at 22 • C (S. iniae and S. agalactiae) or on Marine agar plates and cultured for four days at 15 • C (Aliivibrio salmonicida).Based on the CFU mL -1 as determined by titration, 0.5 MacFarland was prepared in 9 ml sterile physiological saline solution and subsequently diluted to 10 5 CFU mL − 1 in Mueller Hinton Broth (MH broth).All eight different MSP-rich extracts (see Table 1) were first dissolved in ultra-pure water and sterilized by autoclaving, subsequently a two-times serial dilution was made from 25 mg mL − 1 -0.012 mg mL − 1 in MH broth.In a sterile culture plate, 100 µL bacterial suspension (equal to 10 4 CFU) was combined with 100 µL MSP solution dissolved in MH broth, resulting in a two-times serial dilution series from 12.5 mg mL − 1 -0.006 mg mL − 1 .Each plate included a positive control with bacterial strain alone in MH broth, and two negative controls; only MSP rich extract at the different dilutions, and only MH broth.For each bacterial strain, each concentration of MSP-rich extract was tested in three independent culture plates.For both Streptococcus strains, MH broth was supplemented with 5 % defibrinated sheep blood.Bacteria were incubated for 144 h at 22 • C, except Aliivibrio salmonicida which was incubated for 144 h at 15 • C. Following 24 h, 48 h, 72 h and 144 h incubation, or 72 h, 96, 120 h and 144 h for slow growing bacteria (i.e.Aeromonas salmonicida, Aliivibrio salmonicida, Streptococcus agalactiae, Streptococcus iniae) optical density was measured at O.D. 600 nm and plates were photographed.Antimicrobial effects were determined as the lowest MSP extract concentration required to inhibit growth (O.D. 600 nm with visually confirmation from photographs) of tested microorganisms.

Animals
Nile tilapia (Oreochromis niloticus) were reared at 28 ± 2 • C temperature with a 12-12 h light-dark cycle in the Aquatic Research Facilities of Wageningen University and Research (Carus-ARF, Wageningen, The Netherlands).Fish were fed a commercial diet twice per day (Sparos, Portugal).Rainbow trout (Oncorhynchus mykiss) were reared at 14 ± 0.5 • C temperature with a 12-12 h light-dark cycle, also in Carus-ARF.Fish were fed a trout specific research diet twice per day (Research Diet Services B.V., The Netherlands).Both Nile tilapia and rainbow trout were housed under comparable conditions, n = 30 fish per 200 L rectangular tank.Fish were 6 -9 months old during the period that the animals were used.Tanks were connected to a recirculation system, including a trickling filter, sump, drum filter (Hydrotech 500®) and oxygenation unit.The flow-through for each tank was set at 7 L/min using a hand held liquid rotameter.Water quality parameters were checked regularly to ensure that the water quality remained within preset ranges.During the experiment with tilapia, ranges were for temperature 27.8-28.2• C, pH 7.0-7.8,dissolved oxygen 4.7 and 7.5 mg/L, total ammonia nitrogen < 5 mg/L, nitrite < 1 mg/L and nitrate < mg/L, all within the pre-set ranges.During the experiment with rainbow trout, ranges were for temperature 14.6-16.2• C, pH 6.9-8.0,dissolved oxygen 4.7 and 7.5 mg/L, total ammonia nitrogen < 5 mg/L, nitrite < mg/L and nitrate < 500 mg/L, all within the pre-set ranges.All experiments were performed with a permit (DEC number 2017.W-0034) approved by the Central Committee for Animal Testing, and supervised by the Animal Welfare Authority of Wageningen University and Research.

Head kidney leukocyte (HKL) isolation
Fish were euthanized with 0.3 g L − 1 tricaine methanesulphonate in aquarium water buffered with 0.6 g L − 1 sodium bicarbonate (Nile tilapia) or 2-phenoxyethanol (1 mL L − 1 ) in aquarium water (rainbow trout), and then bled via the caudal vein.Head kidneys were removed aseptically and total head kidney leukocytes (HKLs) were immediately separated on a Percoll (GE Healthcare, Thermo Fisher Scientific) density gradient (51 %), as previously described for Nile tilapia [30], and for rainbow trout [31].Directly following isolation, HKLs were resuspended in RPMI-1640 (Gibco) supplemented with 100 IU penicillin/streptomycin (p/s) without serum (in case of ROS analysis) or supplemented with 1.5 % pooled tilapia or trout serum (depending on species).Cells were directly used after isolation in reactive oxygen or gene expression experiments.

Induction of reactive oxygen species (ROS) production by MSP extracts
Production of ROS was determined by a real-time luminol-based luminescence assay, as previously described [32], with minor modifications.Briefly, HKLs were seeded in RPMI-1640 (Gibco) supplemented with 100 IU p/s at a density of 1 × 10 6 per well and incubated for 60 min at 27 • C (Nile tilapia) or at 19 • C (rainbow trout) in white 96-well plates (CLS3912; Corning).Subsequently, cells were stimulated with one of the following: RPMI-1640 cell culture medium (negative control), zymosan (positive control; tlrl-zyd, 50 µg mL − 1 , InvivoGen) or one of the MSP-rich extracts (Table 1) at concentrations of 250, 500, 750, 1000 and 1500 µg mL − 1 .Per independent experiment, all stimuli were performed on HKLs isolated from the same animal.For each independent experiment, each condition was tested in triplicate.For ROS analysis, n = 4 independent experiments (n = 4 fish) per species was performed.Chemiluminescence emission was measured in real time (every 2 min for 120 min) with a FilterMax F5 Multi-Mode Microplate Reader at 27 • C (Nile tilapia) or at 19 • C (rainbow trout), and expressed as fold changes based on areas under the curve [33].Fold changes were based on data from stimulated relative to unstimulated HKLs (treated with RPMI).

Head kidney leukocyte stimulation for gene expression analysis
For gene expression analysis, HKLs were seeded in RPMI-1640 (Gibco), supplemented with 100IU p/s and 1.5 % pooled tilapia or trout serum, at a density of 4.5 × 10 6 per well in 24-well plates (CORN3524; Corning) and stimulated with RPMI (negative control), zymosan (tlrl-zyd, 50 mg ml − 1 , InvivoGen) or one of the extracts (Table 1) at a concentration of 500 µg mL − 1 and incubated at 27 • C (Nile tilapia) or at 19 • C (rainbow trout).At 3 and 6 h post stimulation, cells were lysed in RLT lysis buffer (Qiagen, 79,216) and stored at − 80 • C until RNA isolation.Per independent experiment, all stimuli were performed on HKLs isolated from the same animal.For each independent experiment, each condition and each timepoint was tested in triplicate.Time points were based on preliminary gene expression studies (see supplemental figure S3).For gene expression analysis, n = 4 independent experiments (n = 4 fish) per species was performed.

Total RNA isolation and cDNA synthesis
Total RNA from cells in RLT lysis buffer (Qiagen, 79216) was isolated using the RNeasy mini Kit (Qiagen, 74106), including on-column Dnase treatment, according to the manufacturer's instructions, and stored at − 80 • C. Prior to cDNA synthesis, 750 ng RNA was treated with DNase I, Amplification Grade (Invitrogen), and cDNA was synthesized using random primers (300 ng, Invitrogen) and Superscript III First-Strand Synthesis for RT-PCR (Invitrogen).cDNA samples were diluted in nuclease-free water prior to real-time quantitative PCR (RT-qPCR) analysis.

Gene expression analysis
Gene expression was measured with RT-qPCR using ABsolute qPCR SYBR Green Mix (Thermo Scientific) in a Rotor-Gene Q (Qiagen), and fluorescence data were analysed using Rotor-Gene Analysis software version 2.3.5.Briefly, 7 µL ABsolute qPCR SYBR Green Mix was combined with 2 µL primer mix (2.1 µM) and 5 uL cDNA and RT-qPCR was initiated by 15 min at 95 • C, subsequently 40 cycles of 20-second denaturation at 95 • C, 20 s annealing at 60 • C and 20 s extension at 72 • C followed by a stepwise melting curve with 0.25 • C per 5 s steps.The relative expression ratio (R) of each sample was calculated according to the Pfaffl method [34] based on the take-off deviation of each sample versus each of the PBS controls and normalized relative to elongation factor 1α (elf1α) as previously verified reference gene (see Tables 3  and 4 for the primers used).

Statistical analyses
Statistical analysis was performed using IBM SPSS statistical data editor version 26.Data presented as fold changes were transformed with natural logarithm, prior to statistical analyses.Subsequently, transformed data were tested for normality by using a Q-Q plot and performing a Shapiro-Wilk test.Data was then analysed using a repeated measures linear mixed model followed by a Bonferroni post hoc test.Analysis of gene expression differences was performed without testing for the factor 'time post stimulation' (differences between 3 h or 6 h), however this factor was taken into account as repeated measure.All values are means expressed with standard deviation (SD), and data were considered significant for p<0.05.

Chemical analysis of MSP rich algae extracts reveals distinct differences between two species of algae and processing techniques
Algae were processed and, in total, eight different MSP-rich extracts were produced.Composition (proximal analysis) of the crude extracts differed (Table 5), not only based on the species that the extracts were derived from, green Ulva (UC) versus red Solieria (SC), but were also different based on the extraction methods.Differences were noted in the mineral and carbon content between the two crude processed extracts, most-likely attributable to the difference between the algae species.The osidic composition is not shown here but provided in Supplementary Table S1.

MSP-rich extracts from Solieria chordalis have more potent direct antimicrobial effects on pathogenic fish bacteria than MSP-rich extracts from Ulva sp
In order to analyse antimicrobial effects of the MSP-rich algae

Table 3
Overview of Nile tilapia RT-qPCR primers used in the current study.In case of (genome), sequences from related species were used as BLAST query against the GCA_001858045.3 genome build and primers were designed using the recovered rainbow trout sequences.extracts, we cultured several common pathogenic fish bacteria and titrated the eight different MSP-rich extracts into the bacterial cultures.
In general for each of the analysed MSP-rich extracts antibacterial effects were most pronounced at the earliest timepoint (24 h incubation or 72 h incubation for slow growing bacteria (i.e.Aeromonas salmonicida, Aliivibrio salmonicida, Streptococcus agalactiae, Streptococcus iniae)) and as such only those timepoints are presented in Table 6.Overall, extracts derived from red (Solieria) algae (SC, SC-T and SC-E) showed more pronounced antimicrobial effects than MSP-rich extracts isolated from green (Ulva) algae (Table 6).To extracts stood out in their antibacterial activity; SC-T and SC-E.For these two extracts, we observed clear antibacterial effects, at relatively low concentrations, on most of the tested bacterial species.Overall, we observed a certain degree of antibacterial effects for each of the MSP-rich extracts for at least one of the tested bacterial species.

MSP-rich extracts induce ROS production in leukocytes from Nile tilapia and rainbow trout
In HKLs isolated from Nile tilapia, MSP-rich extracts from green algae (Ulva sp.) significantly induced ROS production in a dosedependent manner, albeit to different degrees, whereas none of the red algae Solieria-derived extracts induced ROS production significantly higher than the base line (RPMI) control (Fig. 1).Also in HKLs isolated

Table 4
Overview of rainbow trout RT-qPCR primers used in the current study.Primers were designed to discriminate between duplicated genes.All duplicated genes were tested individually, but only genes found regulated are presented.The last letter or number of the primer designates the duplicated gene.Further, * indicates the use of common primers (detecting duplicated genes).In case of (genome), sequences from related species were used as BLAST query against the GCA_013265735.3 genome build and primers were designed using the recovered rainbow trout sequences.For abbreviations see also Table 1.

Table 6
Analysis of antimicrobial effects of MSP-rich extracts against fish pathogenic bacteria.Values represent the dose (mg mL − 1 ) at which there was inhibition of growth, as determined by OD600 and visual determination.A value of >12.5 mg mL Note: inhibition of bacterial growth after 24 h for 'fast' growing bacteria, or after 72 h for 'slow' growing bacteria.Data from later timepoints not shown.
For abbreviations see also Table

MSP-rich extracts induce gene expression responses after stimulation in vitro, differing between Nile tilapia and rainbow trout HKLs
We employed the fact that HKLs from common carp, more than HKLs isolated from Nile tilapia and rainbow trout, easily produce NO in vitro, to titrate MSP-rich extracts prior to subsequent studies on gene expression analysis in Nile tilapia and rainbow trout.Thus, based on the titration of ROS induction in Nile tilapia and rainbow trout fish species, and based on the ROS and NO production in common carp (Supplementary Figures S1 and S2), we chose to proceed our studies with a fixed concentration of 500 µg mL − 1 MSP-rich algae extract for gene expression analysis after stimulation.Genes studied were either cytokines, or genes associated with signalling pathways downstream of receptors sensing pathogen-associated molecular patterns (PAMPs), with a specific focus on the pathways associated with Toll-Like Receptors (TLR) and C-type Lectin Receptors (CLR).Gene expression was studied at two time points (3 and 6 h) after stimulation of HKLs.
Analysis of cytokine gene expression revealed a clear induction of most of the studied cytokines (interleukin-1β (il1b); interleukin-10 (il10); tumour necrosis factor-α (tnfa); interferon-γ (ifny); interleukin-12 p40 (p40)).Induced cytokine responses generally were more prominent in Nile tilapia than in rainbow trout (cf.Figs.3A and 4A).Although regulation of cytokine expression appeared ubiquitous, differences in gene expression of il10 and ifny were less informative in rainbow trout.Despite the differences between the two fish species, cytokine read-outs suggested immunomodulatory effects of most if not all red algae Solieriaderived and green algae Ulva-derived extracts.
Analysis of gene expression associated with TLR signalling pathways (irak1, irak4 and myd88) only revealed significant regulation of irak1 and only in Nile tilapia (cf.Figs.3B and 4B).Analysis of gene expression associated with CLR signalling pathways (card9 and bcl10) revealed regulation of both, card9 and bcl10, again only in Nile tilapia (Figs.3C  and 4C).Gene expression analysis suggested the immunomodulatory effects of MSP-rich extracts in Nile tilapia could possibly regulate at least CLR-mediated signalling, independent of algae source.Comparable to the use of ROS production as read-out for suggested immunomodulatory effects of red algae Solieria-derived and green algae Ulva-derived extracts, gene expression as read-out was dependent on the fish species investigated.Overall, expression of cytokine genes in particular, appeared most informative, especially for measuring immunomodulatory effects in Nile tilapia.

Discussion
Increased interest in the use of dietary supplementation with immunomodulators such as marine sulphated polysaccharides (MSP), to maintain the health of animals in intensive aquaculture systems, has triggered a need for research on the underlying cell-based mechanisms of their modulatory capacities.MSP are highly complex sulphated polysaccharides extracted from cell wall matrices of marine algae, with their immunomodulatory mechanisms mostly detailed by studies in higher vertebrate species.So far, effects of MSP on fish cells, and also effects of MSP on fish pathogenic bacteria, had not been investigated.Here, we investigated the effects of (eight) different crude extracts rich in MSP from green macroalgae (Ulva sp.) or red macroalgae (Solieria sp.), first for direct antibacterial effects, and then for in vitro immunomodulatory action on primary head kidney leukocytes (HKLs) from Nile tilapia, and rainbow trout.
Direct antimicrobial effects of MSP-rich algae extracts were investigated by titration on cultures of 13 different bacterial strains representing both, Gram-positive and Gram-negative fish pathogens.Although to some degree we observed antibacterial effects for each of the MSP-rich extracts, extracts derived from red (Solieria) algae showed more pronounced antibacterial effects than MSP-rich extracts isolated from green (Ulva) algae.Inhibitory values were defined as concentrations at which there was notable inhibition of growth after a predetermined time period of 24-72 h, and could be as low as 0.20 mg mL− 1.These values are comparable to the lowest minimum inhibitory concentration (MIC) of 0.16 mg mL− 1 for MSP extracts prepared from green (Ulva) algae, when tested against 42 bacterial strains and isolates from livestock animals [12,17].Although in the latter studies only green (Ulva) algae were included, our data suggest that antibacterial activities of MSP could be host species specific to some extent.Our study further suggests that algal species per se (here, Solieria versus Ulva) could be an important determinant, and that either thermal processing or protease assisted extraction appears beneficial for antibacterial properties of the MSP-rich extracts.Overall, MSP-rich extracts appear to have direct antibacterial effects, at least in vitro, and might be of use to combat bacterial infections or aid in maintaining microbiota homeostasis of both, livestock and aquaculture animals.
MSP-rich extracts derived from green Ulva algae stimulated ROS production by HKLs from both, Nile tilapia and rainbow trout.While the  from either red (Solieria) algae, or green (Ulva) algae.
Pathogen-associated molecular patterns such as likely present in MSP-rich extracts could be sensed by pattern recognition receptors (PRR) from the TLR family, but could also be sensed by other PRR such as those from the C-type lectin receptors (CLR) family.In human embryonic kidney (HEK293) cells, MSP-rich extracts activated cells primarily through Toll-Like Receptor 4, with subsequent activation of signalling via the transcription factor NF-κB, leading to the secretion of several cytokines [12].In the absence of reporter assays for PRR-driven activation of fish cells, analysis of induced expression of genes associated with particular PRRs and their downstream signalling pathways might provide useful information.Here, in Nile tilapia, gene expression analysis of irak1, irak4 and myd88 after stimulation of HKLs with MSP-rich extracts showed regulation of irak1 only.At the same time, gene expression analysis of card9 and bcl10 showed regulation of both genes, suggesting that the recognition of MSP-rich extracts in HKLs might be mediated by a member of the CLR family or that MSP-rich extracts might induce the CLR signalling pathway.Of course, cellular reporter assays should be useful to confirm these preliminary observations on the sensing of MSP by particular cellular receptors on fish cells.Analysis of cytokine gene expression in MSP-stimulated HKLs revealed a clear induction of most of the cytokines studied (il1b, il10, tnfa, ifny, il-12 p40).Since cytokine gene expression can be the result of many different signalling pathways, this provides no further information on a PRR responsible for sensing MSP-rich extracts.Regardless of the exact receptor and downstream signalling pathway involved, the increased gene expression of several cytokines shows the immunomodulatory properties of MSP-rich extracts in Nile tilapia and rainbow trout, at least in vitro.
Gene expression data from Nile tilapia showed, in general, more obvious results, compared to data from rainbow trout.Further, data from Nile tilapia showed, in general, more pronounced effects, compared to data from rainbow trout when it came to studying ROS production by HKLs.In addition, data from common carp were again different from Nile tilapia and rainbow trout; HKLs isolated from carp reacted to almost all MSP-rich extracts with both, enhanced ROS production and nitric oxide (NO) response (supplementary data).Given that Nile Tilapia are Cichlidae, common carp are Cyprinidae, and rainbow trout are Salmonidae, biological diversity between members of such different fish families should not come as a surprise.It may be evident that investigations into the immunomodulatory effects of MSP-rich extracts based on effects found in a single species cannot always be simply extrapolated to several target fish species with different physiology.
Chemical analysis of the different MSP-rich extracts showed clear differences in the composition related to the different algal species but also in the effects of the different treatments.Of interest, sugar content of extracts increased due to either thermal extraction or proteaseassisted extraction compared to crude cold-press extraction.For Ulvaderived extracts a similar treatment effect was observed for sulphate groups content, however, this was not as evident for Solieria-derived extracts.Revisiting the ROS and gene expression analyses with the chemical composition of the different extracts in mind, we found no clear relation between the potency of an induced ROS or gene expression response and the chemical composition of the extracts.This implies that observed immunomodulatory effects of the tested Ulva-and Solieriaderived extracts might not solely depend on the presence or absence of certain sulphate groups or polysaccharide content, but may also depend on other properties, for instance of structural nature.
Finally, for future experiments on immunomodulatory effects of MSP-rich extracts from either red (Solieria) or green (Ulva) algae, one could have a small preference for using green (Ulva) algae extracts.For Nile tilapia, none of the red (Solieria) algae-derived extracts induced ROS production by HKL at levels significantly higher than the base line control, while MSP-rich extracts derived from green (Ulva) algae induced ROS in a dose-dependent manner.Cytokine gene expression read-outs suggested immunomodulatory effects of most if not all MSPrich extracts from either red (Solieria)-or green (Ulva) algae, so provide no information on preference for which algae extract to use, not for Nile tilapia nor for rainbow trout.For common carp, MSP-rich extracts derived from green (Ulva) algae induced ROS production more so than MSP-rich extracts from red (Solieria) algae, although products from both, red (Solieria)-and green (Ulva) algae triggered NO production.
With regard to processing, it could be of interest to study in vivo, differences, or the lack of differences, induced by either cold extraction or thermal processing, and investigate correlations with higher concentrations of active material, or differences in composition.For future experiments with immunostimulatory MSP-rich extracts from green (Ulva) algae, one could have a small preference for using UC-T treatments; i.e.2 h of thermal processing at 80 • C to extract immunostimulatory MSP-rich extracts.In our hands, treatment effects differed between assays and between fish species and thus were not unidirectional, but thermal processing is a relatively reliable process that could be preferred by suppliers.Preferably, future experiments on the immunomodulatory effects of MSP-rich extracts could focus on thermallyprocessed extracts from green (Ulva) algae and address in vivo feeding, respecting biological differences between target species.
1. UC = Ulva concentrate untreated; UC-T = Ulva thermal extraction; UC-E = Ulva protease extraction; SC = Solieria concentrate untreated; SC-T = Solieria thermal extraction; SC-E = Solieria protease extraction; UC-F15 = Ulva filtrated at 15 kDa; UC-F50 = Ulva filtrated at 50 kDa.from rainbow trout, MSP-rich extracts significantly induced ROS production.Responses in rainbow trout were different from responses in Nile tilapia, indicating fish species specific responsiveness, already clear from much higher relative ROS responses induced by the positive control zymosan in rainbow trout (Fig. 2).Different from Nile tilapia, red algae Solieria-derived extracts did induce significant ROS production in HKLs isolated from rainbow trout, albeit without a clear dose-dependent responses.Similar to Nile tilapia, green algae Ulva-derived extracts also stimulated ROS production in HKLs of rainbow trout, however in a less consistent manner and without a clear dose-dependent response.ROS induction by MSP-rich extracts could be most reliably evaluated in Nile tilapia, with clear doseresponses for several green algae Ulva-derived extracts.

Fig. 1 .
Fig. 1.Induction of reactive oxygen species (ROS) in head kidney leukocytes (HKLs) of Nile tilapia after stimulation with MSP-rich algae extracts.HKLs were isolated and stimulated with zymosan (Z; 50 µg mL − 1 , open bars) as positive control, or with different concentrations of Ulva derived (1A), Solieria derived (1B) or filtered Ulva derived (1C) MSP extracts (250 -1500 µg mL − 1 ).Total ROS production was measured immediately following stimulation of cells for 2 h.Bars indicate mean + SD of n = 4 independent experiments.Asterisk (*) indicates significant difference (p < 0.05) relative to the corresponding control (dotted line, cells stimulated with culture medium only) as assessed by a linear mixed model, followed by an LSD post hoc test.Significant differences between concentrations are not indicated.

Fig. 2 .
Fig. 2. Induction of reactive oxygen species (ROS) in head kidney leukocytes (HKLs) of Rainbow trout after stimulation with algae extracts.HKLs were isolated and stimulated with zymosan (50 µg mL − 1 , open bars) as positive control, or with different concentrations of Ulva derived (1A), Solieria derived (1B) or filtered Ulva derived (1C) MSP extracts (250 -1500 µg mL − 1 ).Total ROS production was measured immediately following stimulation of cells for 2 h.Bars indicate mean + SD of n = 4 independent experiments.Asterisk (*) indicates significant difference (p < 0.05) relative to the corresponding control sample (dotted line, cells stimulated with RPMI only) as assessed by a linear mixed model, followed by an LSD post hoc test.Significant differences between concentrations are not indicated.

Fig. 3 .
Fig. 3. Effects of Ulva derived and Solieria derived MSP-rich extracts on gene expression of cytokines and selected receptor signalling molecules in Nile tilapia head kidney leukocytes (HKLs).Gene expression of selected cytokines (3A), TLR pathway associated signalling molecules (3B) and CLR associated signalling molecules (3C) after stimulation with zymosan (50 µg mL − 1 , white bars), Ulva sp.derived extracts (500 µg mL − 1 , green bars) or Solieria derived extracts (500 µg mL − 1 , red bars), at 3 h post stimulation (striped bars) or 6 h post stimulation (open bars).Gene expression is expressed relative to RPMI-stimulated HKLs (negative control, dotted line) and normalized for elf1α expression.All values are means expressed with standard deviation (SD) of n = 4 independent experiments, and data were considered significant for p<0.05.Significant differences between groups (overall regulation per treatment) were assessed by linear mixed model followed by Bonferroni post hoc test.Differences between time points post stimulation were not tested for significance.Groups with different letters are statistically different from one another, with 'a' indicating the lowest value.Groups with asterisk (*) are statistically different from the negative control.

Fig. 4 .
Fig. 4. Effects of Ulva derived and Soleiria derived MSP-rich extracts on gene expression of cytokines and selected receptor signalling molecules in rainbow trout head kidney leukocytes (HKLs).Gene expression of selected cytokines (4A), TLR pathway associated signalling molecules (4B) and CLR associated signalling molecules (4C) after stimulation with zymosan (50 µg mL − 1 , white bars), Ulva sp.derived extracts (500 µg mL − 1 , green bars) or Soleiria derived extracts (500 µg mL − 1 , red bars), at 3 h post stimulation (striped bars) or 6 h post stimulation (solid bars).Gene expression is expressed relative to RPMI stimulated HKLs (negative control, dotted line) and normalized for elf1α expression.All values are means expressed with standard deviation (SD) of n = 4 independent experiments, and data were considered significant for p<0.05.Significant differences between groups (overall regulation per treatment) were assessed by linear mixed model followed by Bonferroni post hoc test.Differences between time points post stimulation were not tested for significance.Groups with different letters are statistically different from one another, with 'a' indicating the lowest value.Groups with asterisk (*) are statistically different from the negative control.Note: for rainbow trout gene expression paralogous genes showed comparable patterns and only one paralog is shown.

Table 1
Description of different processing methods for green and red algae extracts.MSP-rich extracts are divided by species and by processing method.

Table 2
Bacterial strains used for analysis of direct antimicrobial effects of MSP-rich extracts.

Table 5
Proximal analyses of the different MSP-rich extracts (% dry matter).
− 1 indicates absence of growth inhibition at this highest dose tested.Lower values indicate growth inhibition at the dose indicated.Bold and underlined values indicate growth inhibition at a dose <1.0 mg mL − 1 .