Dietary tuna hydrolysate modulates growth performance, immune response, intestinal morphology and resistance to Streptococcus iniae in juvenile barramundi, Lates calcarifer

This study investigated the effects of tuna hydrolysate (TH) inclusion in fishmeal (FM) based diets on the growth performance, innate immune response, intestinal health and resistance to Streptococcus iniae infection in juvenile barramundi, Lates calcarifer. Five isonitrogenous and isoenergetic experimental diets were prepared with TH, replacing FM at levels of 0% (control) 5%, 10%, 15% and 20%, and fed fish to apparent satiation three times daily for 8 weeks. The results showed that fish fed diets containing 5% and 10% TH had significantly higher final body weight and specific growth rate than the control. A significant reduction in blood glucose was found in fish fed 10%, 15% and 20% TH compared to those in the control whereas none of the other measured blood and serum indices were influenced by TH inclusion. Histological observation revealed a significant enhancement in goblet cell numbers in distal intestine of fish fed 5 to 10% TH in the diet. Moreover, fish fed 10% TH exhibited the highest resistance against Streptococcus iniae infection during a bacterial challenge trial. These findings therefore demonstrate that the replacement of 5 to 10% FM with TH improves growth, immune response, intestinal health and disease resistance in juvenile barramundi.


Results
Growth performance. All tested diets were readily accepted by the juvenile barramundi during the 8 week feeding trial. Growth performance, feed intake (FI), food conversion ratio (FCR) and survival of barramundi fed the four experimental diets and the control are shown in Table 1. Among the dietary groups, significantly greater final body weight (FBW) and specific growth rate (SGR) were observed in the group fed TH05 and TH10 compared to the control, but they were not significantly different from the other treatments. The optimal levels of TH for FBW and SGR were investigated through the quadratic regression analysis (Fig. 1), and the estimated TH inclusion level was 10.5% for the highest FBW. However, the feed utilization indices such as FI and FCR, and survival of fish were not affected by any dietary treatments.

Biochemical indices.
With the exception of glucose, none of the measured blood and serum biochemical indices including hematocrit, aspartate aminotransferase (AST), glutamate dehydrogenase (GLDH), total protein, albumin, globulin, albumin and globulin ratio (A/G ratio) were influenced by TH inclusion in diets due to large variation between the fish from within the same dietary treatments (Fig. 2). Blood glucose decreased with increasing TH level, with those fish fed TH10, TH15 and TH20 having significantly lower blood glucose than those in the control.
Histopathology and intestinal morphology. Histopathological investigation revealed that those juvenile barramundi fed with TH20 diet had mild to severe alterations in the liver, spleen, and intestinal tissues (Fig. 3). The notable alterations including cytoplasmic vacuolization with an increased amount of lipid accumulation (steatosis) were found in liver of fish fed with TH20 and control diet. However, no histopathological hepatic alterations were observed in fish fed TH10 as indicated by balanced hexagonal hepatocytes with prominent nuclei and rare cytoplasmic vacuolization or granules. Histopathological observation of the spleen revealed higher and bigger melanomacrophage aggregates, increased white pulp and splenic cord in the corpuscles of the spleen of fish fed with TH20 diet than all other diets. The intestinal folds of fish fed TH20 diet were shorter and fewer in number and the lumen was wider, while fish fed all other diets showed histologically normal intestinal folds. No histopathological abnormalities such as muscular dystrophy, injury or necrotic fibres were observed in muscle tissues of fish fed the experimental diets. The histological measurements of the distal intestine of juvenile barramundi fed diets with different levels of TH are presented in Fig. 4. The micromorphology of intestinal parameters such as goblet cell number per fold (GC), fold height (hF), microvillous height (hMV) and external circumference of serosa (ECS) were altered with the inclusion of TH in diets. The significantly increased GC was found in fish fed 5 to 15% TH included diets whereas increased hMV and ECS were found in TH05 and TH10 diets compared to control. The increased hF was found in fish fed TH05 and TH10 while the decreased hF was observed in TH 20 diet compared to control.
Lysozyme and complement (ACH50) activity. There was a significant variation observed in the serum lysozyme activities of pre-challenged and post-challenged fish. Fish at 24 hours post-challenge exhibited higher lysozyme activity compared to pre-challenged fish and those 7 days post challenge in all dietary treatments. However, serum lysozyme activity was not influenced by the different inclusion levels of TH in the diets (Fig. 5A). The highest complement activity was registered in fish 7 days post challenge compared to pre-challenge and post-challenge fish at 24 h in all dietary treatments. However, no significant difference was observed between pre-challenge and post-challenge fish at 24 h in all treatments. The complement activity of fish was not influenced by the different inclusion levels of TH in the diets (Fig. 5B). The interactive effects of experimental treatments and sampling period (pre, post-24 and 7 d of challenge) on serum lysozyme activity and complement activity of fish are shown in Table 2.

Discussion
Fish protein hydrolysates (FPH) derived from raw waste materials produced through enzymatic hydrolysis are regarded as promising aqua-feed ingredients due to their favorable functional [22][23][24] and nutritional properties 25,26 .
A number of studies have reported that fish hydrolysates are potent growth promoters in fish [27][28][29] . However, FPH have not been previously studied in barramundi. In this study, tuna hydrolysate (TH) derived from processing by-products was tested in juvenile barramundi and it was found inclusion levels of 5 to 10% enhanced the FBW and SGR. Similar positive growth responses to dietary inclusion of fish hydrolysates have been found in many fish species including olive flounder, Paralichthys olivaceus 19 , yellow croaker, Pseudosciaena crocea 30 and Atlantic salmon, Salmo salar 29 . The improved growth performance in the present study following moderate levels of hydrolysate inclusion may be a result of the improved availability and subsequent uptake of free amino acids and suitable peptide fractions produced during the enzymatic process which may be beneficial for the growth performance of fish 31 . Amino acids are crucial for a wide variety of protein syntheses with major physiological functions, such as carriers of oxygen, carbon dioxide, vitamins, enzymes and structural proteins 25 . FPH containing free amino acids and suitable peptides has a substantial role in maintaining good health of fish 32 . However, the use of FPH in aqua-feeds must be at the appropriate level as higher inclusion of FPH may negatively influence the growth and feed utilization in fish 1,33 . In Japanese flounder, Paralichthys olivaceus 16% or higher inclusion of fish hydrolysate in the diet resulted in significant reduction in growth 34 . Also, an inclusion level of 20% fish hydrolysate in turbot, Scophthalmus maximus resulted in significantly reduced specific growth rate (SGR) and feed utilization 31 . In this study, growth performance was significantly elevated at 5 to 10% FM replaced by TH and at further higher replacements (15 to 20%) growth performance started to decline. The detrimental effects of hydrolysates at high inclusion level on fish physiological functioning could be due to an excessive amount of free amino acids (FAA) and peptides of low molecular weight, which may lead to an imbalance in amino acid absorption and saturation of peptide transportation systems 35,36 . Hematological indices have been considered as valuable biological indicators to assess the health status and physiological condition of fish 37 . The results of the current study showed that dietary inclusion of TH in FM based diets had no significant effect on the hematological indices measured, with the exception of glucose. Likewise, Khosravi et al. 19 found that the addition of protein hydrolysates in low FM diets did not alter most of the hematological indices in juvenile olive flounder, Paralichthys olivaceus while some of the health parameters (lysozyme activity, total immunoglobulin) were improved in hydrolysate supplemented groups. In the current study, the concentration of blood glucose was significantly lower in juvenile barramundi fed with 10 to 20% TH included diets compared to those in the control. This result is in accordance with Khosravi et al. 13 , who reported the same effect in red sea bream, Pagrus major where blood glucose levels were significantly reduced in those fish fed diets containing shrimp hydrolysate. However, another study with the same species found no significant differences in blood glucose levels when fed a diet containing fish hydrolysate 7 . This difference therefore appears to be due to the different types of hydrolysates used between the two studies, but may be due to a number of factors including experimental conditions, fish size and handling methods, as they can strongly affect fish physiological condition 38,39 . The enzymes AST and GLDH are normally measured in fish as the indicators of hepatocellular injury, to determine liver health status. In the present study, the lack of a significant increase in AST and GLDH suggest that the FPH did not cause liver damage). Similarly, Khosravi et al. 13 found no significant difference in serum AST level by the addition of FPH to the diet of red sea bream, Pagrus major. However, Cai et al. 40 observed that yellow croaker, Larimichthys crocea fed a diet with 40% fish hydrolysates had higher AST levels than fed a control diet. The intestine, a primary immune organ of the body, plays a major role in the ingestion and absorption of nutrients, and participates in the protection of the host body through a strong defence against pathogens, allergens and toxins 41 . Some earlier studies have stated that the distal intestine of carnivorous fish is more sensitive in relation to diets and have larger absorptive surface area including villi, microvilli, and higher densities of goblet cells (GC) in the epithelium [42][43][44] . Furthermore, this part of the intestine has shown the highest variations when alternative protein sources are incorporated in the diets of fish 45 . In the present study, the GC in the intestine were found scattered in order to protect the mucus membranes by secreting mucus 46 . Fish fed TH05, TH10 and TH15 had higher numbers of mucus-secreting GC in the intestine compared to the control. A number of previous studies have reported that GC are positively correlated with the absorption of digestible substances and higher GC results in higher mucosal membrane protection 47,48 . The increment of GC in fish fed the TH05 and TH10 diets might be due to the improved innate immune function against invading microorganisms. These observations are in agreement with an earlier study on red sea bream, Pagrus major where dietary inclusion of shrimp hydrolysate in a low fishmeal diet resulted in an increased GC 13 . It is well known that dietary intake of fish has a marked effect on intestinal health, development and function. The longer fold and villus height of intestine are associated with the good health and high absorptive efficiency, whereas shorter fold and villus height are correlated with higher number of pathogenic bacteria in the digestive tract. Moreover, a shortening of the microvillus height can lead to poor nutrient utilization and absorption, reduced immune functions, thereby lower growth performance of fish 49 . According to Dimitroglou et al. 50 good intestinal health in fish is of great importance not only to achieve target growth rates and feed efficiency but also improved the health status of the mucosal epithelium by providing an effective immune barrier against potential intestinal pathogens. In the current study, the histological evaluations in terms of hF, hMV and ECS were increased in fish fed TH05 and TH10 diets might be due to the greater nutrient absorption and utilization results in more surface area for nutrient uptake which was demonstrated by enhanced growth performance of fish. Novriadi et al. 51 reported that the inclusion of 4% squid hydrolysate in the plant based diet partially restore the intestinal inflammation caused by the high inclusion of plant proteins in the diet of Florida pompano, Trachinotus carolinus. When a fish is challenged with pathogens, it is the task of the innate defense system to protect or fight against the pathogens. In order to compensate for a deficiency in the adaptive immune system, fish lysozyme, in the absence of complement has substantial antibacterial activity compared with mammalian lysozymes, not only against Gram-positive bacteria but also against Gram-negative bacteria. Neutrophils and macrophages are the major sources for producing lysozyme 52 . The alternative pathway of complement activity is also an innate component of the immune system protecting fish from invasive pathogens 53 . Multiple studies have suggested that inclusion of FPH in fish diets may stimulate the non-specific immune responses, and this stimulant is strongly influenced by the amount of hydrolysate in the diet 28,54 . However, if the inclusion level of the hydrolysate is too high (>30%), it may have a negative effect in fish 33,54 . The higher lysozyme activity in infected fish demonstrates the defense response to the S. iniaie infection in 24 h post challenge and decline at 7 d post-challenge may be  56 . In the current study, fish fed the TH05 and TH10 diets showed higher resistance against infection, while control fish showed the lowest resilience during the 14 days of bacterial challenge. Similarly, dietary administration of FPH increased the disease resistance of various fish, such as red sea bream, Pagrus major and juvenile olive flounder, Paralichthys olivaceus against Edwardsiella tarda 7 and European sea bass larvae, Dicentrarchus labrax to Vibrio anguillarum 28 .
In summary, based on the quadratic regression analysis of FBW level, the optimum TH for juvenile barramundi was estimated to be 10.5%. Although the immune parameters (lysozyme, ACTH50) were not affected by TH inclusion in the diets, the increased growth performance and intestinal micro-morphological parameters (GC, hF, hMV and ECS), and decreases in blood glucose level in fish fed TH included diets at moderation might be associated with the improved resistance of juvenile barramundi against S. iniae infection, resulting in higher survival during post-challenge. However, further studies on this subject are needed to connecting the linkage between FPH utilization and disease resistance of fish.

Materials and Methods
Ethic statements. This study was conducted in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of Australia. The protocol was approved by the Ethics Committee in Animal Experimentation of the Curtin University (Approval number AEC_2015_41).
Experimental fish. Following a stringent size grading, a total of 300 healthy juvenile barramundi were Experimental diet. All ingredients, except TH were purchased from Specialty Feeds Pty. Ltd, Great Eastern Highway Western Australia. Liquid TH was provided by SAMPI, Port Lincoln, Australia. The dried TH contains 58.4% protein, 1.05% lipid and 11.3% ash. Five isonitrogenous and isocaloric diets were prepared for barramundi having 47% crude protein (CP) and 20 MJ.kg −1 gross energy (GE). These diets were labelled as TH0, TH5, TH10, TH15 and TH20 to replace FM at 0%, 5%, 10%, 15% and 20%, respectively by TH. TH0 diet with no replacement was considered as the control. The formulation and proximate composition of the experimental diets are presented in Table 3. The experimental diets were prepared based on the standard method of CARL 57 . All test diets were processed with the addition of water to about 35% mash dry weight of mixed ingredients to form a dough. This dough was then passed through a mincer to create pellets of the desired size (3 mm). The moist pellets were then oven dried at 60 °C for 48 hours and then cooled at room temperature, sealed in plastic bags and stored at −15 °C until further use.
Experimental conditions and feeding. Following the aforementioned 14 day acclimation period, 300 uniformly sized juvenile barramundi (pool weight of 12.23 ± 0.11 g fish −1 ) were randomly distributed into fifteen independent tanks (300-L water capacity) at a stocking density of 20 fish per tank. Each tank was supplied with constant aeration and water was recirculated from an external bio-filter (Fluval 406, Hagen, Italy) at a rate of 10 L min −1 . The water quality parameters such as temperature (27.90-29.20 °C), salinity (32-36 ppt), dissolved oxygen (5.92-7.42 mgL −1 ), ammonia nitrogen (<0.50 mgL −1 ) and nitrite (<0.50 mgL −1 ) were monitored daily and were always within the suitable range of fish culture in recirculating aquaculture systems. Fish were kept at 14:10 hr light: dark cycle using automatic indoor light switches (Clipsal, Australia). During the experimental period of 8 weeks, fish were fed the treatment diets to satiety three times a day at 0800, 1200 and 1700 h. Fish were starved for 24 h prior to being anaesthetised (AQUI-S ® , 8 mgL −1 ), weighed and taking blood samples.

Biochemical indices of blood and serum.
At the end of the feeding trial, duplicate blood samples from two anaesthetized fish per tank (six fish per dietary treatment) were withdrawn by caudal vein puncture with a 1 mL non-heparinized syringe. The first set of extracted blood was transferred to heparinised tubes for the determination of haematocrit and blood glucose level. The second set of blood samples were transferred to non-heparinized tubes and allowed to clot overnight. The following day clotted blood samples were centrifuged at 3000 rpm for 15 min at 4 °C, serum was separated and then stored at −80 °C for later measurement of the serum biochemical parameters and immunological indices described below. Hematocrit (Ht %) was determined by centrifugation of whole blood in glass capillary tubes at 2000 rpm for 5 min following the method of McLeay and Gordon 58 and expressed as a percentage. A blood glucose meter kit (Accu-Chek, Australia) was used to measure the blood glucose level. Serum biochemical parameters, including aspartate aminotransferase (AST), glutamate dehydrogenase (GLDH), total protein and albumin were measured using an automated blood analyzer (SLIM; SEAC Inc, Florence, Italy) following the methods from Blanc et al. 59 . The total globulin content was determined by subtracting the albumin values from the total serum protein values. The albumin and globulin ratio (A/G ratio) was obtained by dividing albumin values by globulin values.

Histology and intestinal micromorphology analysis.
In order to analyze the histopathological condition of liver, spleen, muscle, distal intestine, and histomorphological condition of the intestine, two fish from each replicate were examined (i.e. six juvenile barramundi per dietary treatment) from which blood had previously been extracted. Samples of all tissues were fixed in 10% buffered formalin, dehydrated in ethanol before equilibration in xylene and embedding in paraffin wax. Sections of approximately 5 µm were cut and stained   with haematoxylin and eosin (H&E) for histological examination under a light microscope (BX40F4, Olympus, Tokyo, Japan). Digitalized histology images were analyzed using Image J software at different magnification for assessing the height of folds, enterocytes and microvilli according to the procedures described by Escaffre et al. 60 with minor modifications. The number of goblet cells were counted in the highest 10 mucosal folds with the numbers expressed as average number of goblet cells per fold as described by Ramos et al. 61 . For gut sample, three cross-sections were quantified for GC, hF, hMV and ECS of the distal intestinal samples.
Bacterial challenge trial. S. iniae, a bacterium pathogenic for barramundi was obtained from the Bacteriology Laboratory, Department of Agriculture & Food, Perth, Australia. The bacteria were grown in trypticase soy broth (Oxoid, Basingstoke, UK) at 24 °C for 24 h and the broth containing the culture was centrifuged at 5000 g for 15 min. The supernatant was discarded and the pellets were washed twice in phosphate-buffered saline (pH 7.2). At the end of growth trial, 10 average sized fish from each replicate tank were moved to each of 20 × 100 L capacity glass aquaria in separate room in CARL for 14 days bacterial challenge. Of the 20 aquaria, 15 were used for survival assessment counting and 5 were utilised for blood sampling after challenge. The experimental conditions were as follows: water temperature 28.2 °C, salinity 35 g L −1 , pH 7.6 and photoperiod 14:10 hr light: dark. Following the acclimation, fish were subjected to a bacterial bath challenge with S. iniae by removing the fish from the tank and adding them to a bath containing 1.8 × 10 3 CFU mL −1 of the bacteria for 1 minute according to Bromage and Owens 62 . After bathing, fish were returned to their respective aquaria and feeding continued on the treatment diets once per day and fish were closely monitored for bacterial infection. During the challenge period, fish were monitored for signs of infection counted twice daily at 0800 and 1700 h. Infected fish were counted and then removed. Blood sampling for the immune parameters lysozyme and complement activity were conducted before challenge and then again 24 h and 7 d post challenge, and the fish were returned to the respective aquaria after bleeding. To avoid the repeated blood sampling from same fish, fish were tagged individually during stocking.
Immunological indices of serum. Lysozyme activity assay. Serum lysozyme activity was assessed by a turbidimetric assay described by Le et al. 63 with slight modifications. Briefly, Micrococcus luteus (0.6 mg mL −1 ) (Sigma) suspension at 0.2 mg mL −1 was suspended in sodium phosphate citrate buffer (pH 7.2, 0.05 M) and 30 µL of serum samples were placed into wells of a 96-well plate in triplicate. The mixture was incubated at 25 °C and its absorbance was monitored every 5 min for a total of 30 min at 450 nm with a plate reader. The results are presented as Unit mL −1 .
Alternative complement activity assay. The alternative complement activity was measured using a method modified from Yadav et al. 64 using rabbit red blood cells. Briefly, the rabbit red blood cells (RaRBC) were washed 3 times in 10 mM EGTA-GVB buffer (ethylene glycol tetra-acetic acidmagnesium-gelatin veronal buffer) and then diluted to give 1% suspension containing 2 × 10 8 cells mL −1 in the same buffer. The RaRBC suspension was standardized by adding 100 µL of the 1% suspension to 3.4 mL of distilled water as a blank and the OD of the hemolysate was measured at 405 nm against distilled water to obtain the 100% lysis value. For the blank, red blood cells were similarly mixed with the EGTA -GVB working buffer. A quantity of 100 μL aliquots of serially diluted serum in EGTA -GVB buffer were mixed with 20 μL of red blood cells in a 96 round bottom well plate. The plate was incubated for 90 min at room temperature with gentle shaking every 15 minutes to suspend the RaRBC. After incubation, the plate was centrifuged for 10 min at 800 g at 4 °C. The optical density (OD) of the supernatant was measured at 405 nm using a plate reader. The reciprocal of the serum dilution inducing 50% lysis of RBCs was determined as the ACH50 expressed as unit mL −1 .
Statistical analysis. The data were analysed using SPSS for Windows version 25, IBM Curtin University, Australia. Except lysozyme and complement activity all data was subjected to one-way analysis of variance (ANOVA) followed by Turkey multiple range tests to compare the control diet against each test diet containing tuna hydrolysate (TH). Lysozyme and complement activity were analysed by multifactorial analysis of variance (ANOVA). All results are expressed as means and standard errors (S.E.) with p-values less than 0.05 were considered statistically significant. The FBW was subjected to quadratic regression analysis with TH inclusion levels. The survival graph was constructed using the Kaplan-Meier method and the differences among different dietary groups were performed using log-rank test.

Data Availability
The datasets generated and analysed during the current study are available from the corresponding author on reasonable request.