Effect of Pulsed Feeding of GIFT Strain of Tilapia in Biofloc System Using Inland Saline Water

,


Introduction
In India, around 8.621 million ha of land has been badly afected with the problem of soil salinity and 1.93 million km 2 areas is under laden with ground saline water [1]. Agricultural farmers abandon these lands as barren felds due to poor agriculture productivity [2], but these lands prove to be a valuable asset for aquaculture [3]. Aquaculture can be the right strategy to reduce the salt content in underground water tables and to generate income through enhanced production of euryhaline and marine fshes with high growth potential [3,4]. However, its expansion is limited due to scarcity of water sources and competition with other water users such as agriculture and urban activities [5]. Also, intensifcation in this sector in ecologically sensitive areas and fragile lands lead to environmental degradation, if efuents are not treated before discharge [6]. So alternate and sustainable use of available water resources has become a necessity for production of food especially quality protein to feed the growing population [7].
Biofoc technology (BFT) would be a remunerative and sustainable means to reclaim salt-afected resources for food production using limited water and land resources [8,9]. BFT minimize water exchange and water usage in aquaculture systems through maintaining adequate water quality within the culture unit, while producing low-cost biofocs rich in protein [10]. Manipulation of Carbon and Nitrogen balance stimulates heterotrophic bacterial growth responsible for transforming nitrogenous waste accumulated as uneaten feed and excreta products in the culture system (70-80%) into microbial protein [11,12] hence minimizing ammonium concentration faster than nitrifcation process [10, 13 14]. In addition, it acts as a potential extra food source (available 24 hours) for the cultured organism supplying protein and other nutrients required for the culture organism [11,15,16]. Considering the long-term sustainability of aquaculture mainly based on feed cost (40-50%), the biofoc system provides an opportunity to reduce this cost [17].
Te fact of continuous development in aquaculture sector places the importance of research on the need for new alternatives in diets and feeding strategies [18]. Reference [19] proposed two ways of reducing feed costs i.e., developing lowcost diets or adopting diferent feeding strategies or good husbandry methods. Pulsed feeding is the short period alteration in feeding strategy. Several studies evaluated diferent feeding strategies in tilapia including mixed feeding schedules of difering protein content, varying percentage and quality of protein, reducing feeding rate, delayed feeding, and alternate feeding to reduce the production cost [20][21][22][23]. BFT too allows implementation of such strategies through adoption of delayed/alternate feeding schedule or through supplementing protein at graded level in diet of aquaculture species reared under BFT [24,25]. When using BFT for tilapia culture the reduction in the use of artifcial feed could be higher and the nutritional demands seem to be modifed based on the biofoc diversity, composition and intake ratio enabling greater fexibility in formulations, and inclusion of nonconventional ingredients [26,27].
Adaptation to crowded condition, higher density, and physiological adaptations to consume biofoc has made GIFT stain of Tilapia as the preferred species for culture under BFT system [28,29]. GIFT strain of tilapia gives better growth compared to normal strains and can be cultured in both freshwater and brackish water upto salinity range of 12-15 ppt [29][30][31] presenting opportunity for diversifcation of its culture in inland saline water. As GIFT strain of tilapia can consume biofoc, hence would be benefted by nutrition through microbial foc consumption during the period of feed deprivation which might reduce the amount of feed for culture of fsh in BFT system. With this in backdrop the current study was planned for 60 days to assess the growth, survival, and immuno-physiological responses of GIFT Tilapia under pulsed feeding in BFT using inland saline water of 10 ppt salinity.

Experimental Design and Fish
Stocking. Te experiment was conducted in 15 fber reinforced plastic (FRP) circular tanks (500-L capacity with working volume of 300 L) for 60 days at Central Institute of Fisheries Education, Rohtak Centre, Haryana, India. Genetically improved farmed tilapia (GIFT) fry (n = 1000) for the experiment were procured from Svara biotechnovations, Terku Pethampatti, Madurai, Tamil Nadu, India. Te healthy fsh having an average weight of 0.20 ± 0.01 g were acclimatized and held in a nursery pond at 3 ppt for 3 weeks fed with a commercially available feed containing 36% protein (foating feed, Growel growfn fsh feed). Ten, the fsh were held for a week in 1200 L FRP tank for acclimation to 10 ppt inland saline water. Te experiment followed completely randomized design (CRD), with four biofoc treatments and one control (clear water) viz., T1 (insitu biofoc with daily feeding), T2 (in-situ biofoc with alternate day feeding), T3 (in-situ biofoc with every 3rd-day feeding), T4 (in-situ biofoc with no feeding), and C (clear water with daily feeding), in triplicates. A total of 600 GIFT fry of mean weight (6.15 g ± 0.02) were stocked randomly in 15 tanks. Te carbon-nitrogen ratio (C/N) in BFT was maintained at 20 : 1 as per Avnimelech [11] using jaggery (39.99% carbon, w/w) added twice a week in the treatment units based on quantity and protein content in the feed. Te biofoc inoculum preparation and carbon source requirement calculation were made by following Avnimelech [11]. Te fshes were fed with commercially available feed containing 32% protein (foating feed, Growel growfn fsh feed) @ 4% body weight split into two equal amounts given at 09:00 am and 05:00 pm.

Preparation of Inoculum.
Biofoc inoculum was prepared in 300 L FRP tank flled with 10 ppt inland saline ground water upto 250 L and continuous aeration was provided. Te pond soil was obtained from the dried shrimp pond of CIFE-Rohtak. Te carbon source used for inoculums was molasses. Preparation of inoculum was carried out according to Avnimelech [11] using 10 gm/L pond soil, 10 mg/L ammonium sulphate and 200 mg/L molasses. Te inoculum developed within 10 days (foc volume, >40 ml/L) and was distributed equally (50 L per tank) into the already prepared experimental tanks. Carbon sources were calculated and added twice a week based on quantity and protein content in the feed used. C/N ratio maintained was 20 : 1. Carbon source used for maintenance of biofoc was jaggery. Continuous aeration was provided in all the experimental tanks from a centralized aeration connected to two air pumps of 150 W (Hiblow HP 200) with an output capacity of 200 L·min −1 . Te aeration pipe in each tank was provided with air stone and a regulator to control the air pressure in all the tanks. Tere was no removal of foc or sludge throughout the experiment.

Assessment of Water Quality
Parameters. Water quality parameters like temperature, pH, and salinity were monitored daily. Dissolved oxygen (DO) was monitored weekly using Winkler's titrimetric method [32]. Total ammonia nitrogen (NH 4 -N), Nitrite-N (NO 2 -N), and Nitrate-N (NO 3 -N) were measured spectrophotometrically every 7th-day interval according to standard methods [33]. Floc volume was measured by allowing foc to settle down for 30 min. in Imhof cone without disturbance.

Growth Parameters.
More than 50% of fsh of diferent treatment groups were sampled for length and weight measurement at an interval of 10 days and feeding was adjusted accordingly. Total weight gain, specifc growth rate (SGR), feed conversion ratio (FCR), feed conversion efciency (FCE), protein efciency ratio (PER), and survival percentage were calculated using the following formulae: Total weight gain (g) � final weight (g) -initial weight (g), (1)

Analysis of Immune Parameters.
After completion of the feeding experiment, blood was collected from 9 fsh from each treatment groups (3 from each replicate) with and without anticoagulant. Te collected blood was allowed to clot for 4 h and then centrifuged at 5000 rpm for 5 min followed by the collection of serum. Serum was stored at −80°C for further analysis. Respiratory burst activity (production of superoxide anion O 2 �) was determined by reduction of nitro blue tetrazolium (NBT) to formazan following the method of [34]. Te optical density (OD) was read at 595 nm in an ELISA reader. Serum lysozyme activity was measured using turbidimetric assay utilizing hen egg white lysozyme as standard following Sankaran and Gurnani [35]. Te unit of lysozyme activity was described as the amount of enzyme that caused a decrease in absorbance of 0.001 m −1 . Te total myeloperoxidase of serum was measured following the procedure described by Quade and Roth [36] with some modifcations. Te optical density was read at 450 nm.

Antioxidant Enzyme Parameters.
At the end of the experiment, liver and muscle tissue were extracted from fsh (n � 3) from each tank and were then pooled and stored in 0.25 M chilled sucrose solutions at 1 : 19 (tissue: sucrose) ratio and refrigerated. Te tissues were then homogenized and centrifuged at 6000 rpm for 10 minutes at 4°C in a refrigerated centrifuge Eppendorf (Germany). Te supernatant solution was preserved in the autoclaved tube and stored at −20°C for future use. Protein of diferent tissues was estimated by the Bradford method [37]. Reading taken at 595 nm against the blank was expressed in mg/g wet tissue.
Superoxide dismutase activity in the liver was assayed as per Misra and Fridovich [38] protocol with slight modifcations. Te increase in absorbance was recorded at 480 nm at every 30 s for 3 min in UV spectrophotometer and the values are expressed as 50% inhibition of epinephrine auto-oxidation min −1 mg protein −1 . Catalase activity for tissue liver was carried using H 2 O 2 substrate (0.03 M in phosphate bufer) according to the method followed by [39]. Te decrease in OD was measured at 240 nm at every 30 s for 3 min. and expressed as moles of H 2 O 2 decomposed per min. per mg protein. Glutathione peroxidase activity of the serum was assessed by using Cayman Glutathione Peroxidase Assay kit (Cayman Chemical Company, USA) following the specifed protocols. Te absorbance was measured at 340 nm using ELISA plate reader. Specifc activity was expressed as GPxμmol/min/g protein.

Metabolic Enzymes Assay.
Te lactate dehydrogenase (LDH) activity was assayed from muscle tissue by the method of Wroblewski and Ladue [40]. OD was recorded at 340 nm at 15 seconds interval for 3 minutes. Enzyme activity was expressed as micromoles of NAD released per mg protein per min at 37°C. Malate dehydrogenase (MDH) activity in muscle was assayed in similar to LDH activity except 0.02 M oxaloacetate was used as the substrate as followed by the Ochoa [41]. Liver, Serum, and muscle protein were estimated by Lowry's method [42] using bovine serum albumin as standard.

Statistical
Analysis. All data were statistically analyzed using software SPSS version 16.0 (Chicago, IL, USA). Te signifcance of each parameter among diferent treatments was statistically analyzed using one-way ANOVA and signifcant diferences among treatments were coined using Duncan multiple range tests (P < 0.05). Aquaculture Research 3

Water Quality Parameters.
Te physicochemical parameters of water quality analyzed during the experimental period are shown in Table 1. Although a signifcant difference (P < 0.05) was observed in DO between control and biofoc treatment with the highest average value in control, no signifcant diference (P > 0.05) was observed in temperature and pH among the diferent treatment groups. Salinity was maintained at 10 ppt throughout the experimental period. During the production cycle, nitrogenous compound i.e., ammonia (NH 4 + -N: mg/L), nitrite (NO 2 -N: mg/L), and nitrate (NO 3 -N: mg/L) showed a signifcant diference (P > 0.05) between the treatments. Te lowest value was recorded in control and highest in T1. Ammonia (NH 4 + -N) was in the range of 0.11-0.64 mg/L, nitrite-N was 0.08-0.37 mg/L and nitrate-N was 0.54-9.10 mg/L. All these parameters were signifcantly higher in biofoc based units as compared to control. During the culture period, foc volume in biofoc tanks ranged from 6.13 to 40.67 ml/L. T4 recorded the lowest foc volume throughout the experimental period.

Growth Performance.
Growth performance of the GIFT strain of tilapia over the experimental period is represented in Table 2. After 60 days of trial, a signifcant diference (P < 0.05) was observed between the treatments in terms of average body weight, body weight gain, length gain, specifc growth rate (SGR), feed conversion efciency (FCE), feed conversion ratio (FCR), protein efciency ratio (PER), and survival rate as shown in Table 3. Among the treatment groups highest average body weight, length gain, weight gain, and SGR were observed in T1 compared to control. However, FCR of the group, T1 did not difer signifcantly (P > 0.05) with control. Highest FCE (1.27 ± 0.06) was noticed in T3. T1 and control did not show any signifcant diference (P > 0.05) in FCE. Signifcantly lowest survival rate was observed in T4 compared to other treatments and control.

Immune Parameters.
Respiratory burst activity (OD at 595 nm) and serum myeloperoxidase activity (OD at 450 nm) difered signifcantly (P < 0.05) (Figures 1(a) and  1(b)). Higher activities were observed in biofoc treatment groups compared to control. Fish reared in T1 showed signifcantly higher NBT activity while fsh in T4 showed higher myeloperoxidase activity. Serum lysozyme activity did not exhibit any signifcant diference (P > 0.05) among the treatments (Figure 1(c)).

Antioxidant Enzymes.
A signifcantly (P < 0.05) higher level of liver SOD activity was observed in biofoc groups than control but the highest was in T4 (Figure 2(a)). Liver catalase activity difered signifcantly with the highest being in T4 and lowest in control (Figure 2(b)). Tere was no signifcant diference (P > 0.05) in serum GPx among the treatments (Figure 2(c)).

Carbohydrate Metabolic Enzymes.
Te LDH activity in the muscle of GIFT tilapia showed the highest values in T4 and lowest in T1 group while no signifcant diference (P > 0.05) was observed in T2, T3, and control (Figure 3(a)). However, liver MDH activity did not difer signifcantly (P > 0.05) among the treatments (Figure 3(b)).

Discussion
During the experimental period, DO concentrations and pH values were within the acceptable range as reported [43] for the Nile tilapia under biofoc except for the low temperature as the seasonal change directed towards low temperature (Table 1). Te study was conducted in the month of October-November that recorded lower temperature. Martinez et al. [44] recognized the growth of fsh as a complex process afected by many abiotic factors and temperature is one of the most important factors. Hossain et al. [45] recommended 25°-35°C suitable for raising tilapia. Lower pH and DO in BFT treatment compared to control is likely due to increased C : N ratio which stimulates the growth of heterotrophic bacteria which in turn require oxygen for their growth [46][47][48][49]. Salinity was maintained at 10 ppt but a slight increase was noticed in BFT treatment which could possibly due to evaporation [47].
Increasing C : N ratio of 20 : 1 reduces NH 3 -N, NO 2 -N, and NO 3 -N [50] through uptake by the microbial community [51] and maintaining DO enables bacteria to convert ammonium into bacterial biomass [52]. In the present study, inorganic nitrogen concentration was within the range for raising tilapia under BFT [29,53]. Te foc volume in BFT treatments of the present study was recorded as 7.4 ml/L in no feeding treatment to 40.67 ml/L in daily feeding treatment. Te volume was sufcient to support growth in all other treatment [54] except no feeding treatment which may be due to feeding on foc which was the only source of food in the system.
Signifcant variation in average body weight was observed in BFT treatments and control with BFT treatments showing higher ABW than control until 30 days of the culture period (Table 2). Later a similar trend followed till last. Variation in temperature observed in a range of 20°-26.3°C could be the reason for such variation in growth. Weight gain percentage of SGR was signifcantly higher in BFT with daily feeding compared to other BFT treatments and control (Table 3). Similar enhanced weight gain and SGR of GIFT tilapia were reported in BFT system than the control group [43,55,56]. Other biofoc treatments also had enhanced growth parameters indicating that GIFT tilapia fngerlings harvested and assimilated the biofoc efectively, but not sufcient growth was obtained as in T1 and control as they received daily feeding. Eroldogan et al. [57] observed partial compensation in growth when Gilthead Sea Bream, Sparus aurata juveniles were subjected to moderate levels of restriction (50%) and short-term restriction (2d) compared    to fsh fed to satiation. A similar report on reduced growth on skip feeding was reported by Cuvin-Aralar et al. [58] in Lake Bato indicating only partial compensation by natural food available in cages. Tese results are distinctly diferent from reports of Bolivar et al. [23] on an alternate day feeding strategy for Nile Tilapia in which the mean growth performance of fsh in the daily and alternate day feeding groups lacked any signifcant diference in a pond ecosystem. Qiang et al. [59] observed SGR and FE in the range 2.01-2.38%/day and 0.78-0.86, at water temperature 27.5°C and 10 ppt salinity. In the present study, the weight gain and SGR decreased with the increased gap in feeding days, this may be due to competition for food. In T4 lowest average body weight, SGR and weight gain were attained in GIFT fngerlings as artifcial feed was not used and fngerlings were fed purely on biofocs. Sharma et al. [60] in the Labeo rohita fngerlings reared in biofoc suspension supporting that the role of artifcial feed in intensive fsh farming cannot be ignored as nutritional requirements of fsh depend upon the feed supplied and natural productivity of the system that was in the form of biofoc in the present experiment. In the absence of external feed source, they shifted to biofoc that resulted in good growth but feeding on biofoc alone was not suitable for efcient growth as mixed biofoc and artifcial feeding also resulted in comparatively less growth. As expected, FCR was lowest in T4. Te decrease in FCR from 1.29 ± 0.01 to 0.79 ± 0.4 indicates that in biofoc treatments with alternate day and every 3rd-day feeding, fngerlings grazed on the biofoc. According to Bolivar et al. [23], grazing undoubtedly contributes to low FCR in Nile tilapia grazing on plankton in ponds which is an important component of the diet of the fsh. Wasielesky et al. [61]    observed a decrease in FCR from 1.39 : 1 to 1.03 : 1 when culture subsisted on natural productivity, indicating the potential to reduce the amount of feed in the presence of biofoc. PER was lower for control group as compared to biofoc treatment but was comparable with biofoc with daily feeding, indicating microbial protein utilization as an alternative food source apart from the artifcial feed by the fsh [29].
Te total absence of artifcial feed afected the survival rate (85%) in biofoc with no feeding treatment. A period of 7.86 days was sufcient to obtain 50% mortality in Fenneropeneaus chinensis juveniles under starvation [62]. Lara et al. [24] observed 37.50% survival in Litopenaeus vannamei under starvation for 21 days in a biofoc culture system. Intermittent feeding did not afect the survival rate in juvenile L.vannamei as reported by Zhu et al. [63]. Te only source of feed to fsh in T4 group of the present study was biofoc, refecting the contribution of heterotrophic bacteria to survivability upto 85% of the starved fsh during 60 days of experiment period. Also, lower survival can be related to cannibalism, indicating fsh did not survive when fed solely on biofoc and require food to maintain basic metabolic activities.
Te general feature of biofoc is immunostimulation efect and considering the immunological factor lower weight and survival is expected in fsh under food restriction than fsh that did not experience any feeding stress [24] but the immune-stimulation extent is afected by feed restriction in in-situ biofoc. Te nitroblue tetrazolium (NBT) assay is indicative of oxidative radical production from neutrophils and monocytes for use in defense against pathogens [64]. In the present study, NBT activity was higher in biofoc based treatments compared to control (Figure 1(a)). Xu and Pan [65] and Ekasari et al. [66] reported increased respiratory burst activity of shrimp in biofoc based culture system. Te earlier fnding reported enhanced NBT activity in L. rohita and MPO values in GIFT tilapia cultured in BFT, supporting the immunostimulatory potential of microbial foc [29,53]. Among the biofoc based treatments, the NBT activity decreases with the increased days of pulsed feeding indicating stressful condition and requirement of feed by the fsh. Lysozyme is an important defense molecule of the innate immune system, which is important in mediating protection against microbial invasion [67] and breaks down β-1, 4 glycosidic acids and N-acetyl-glucosamine in the peptidoglycan of bacterial cell walls. No signifcant diference was observed in lysozyme activity due to pulsed feeding in biofoc based system (Figure 1(c)). Caruso et al. [68] observed no efect of lysozyme in plasma in 58 days starved European eel. Luo et al. [52] reported no diference in serum lysozyme activity in BFT and RAS in GIFT. MPO is an important enzyme having antimicrobial activity. It utilizes hydrogen peroxide during the respiratory burst to produce hypochlorous acid [69]. Reduced activity may indicate the presence of contaminants or stress [64]. In the present study highest MPO activity in biofoc based treatments in GIFT, tilapia signifes well-developed immune status compared to control (Figure 1(b)). Te results of the present study are in agreement with Ahmad et al. [53]. who reported the highest MPO activity in biofoc based treatments in L. rohita. Wu et al. [70] observed increased myeloperoxidase content after oral administration of Sophora favescens in GIFT tilapia.
Reduction in feeding depletes organ antioxidant storage which is the defense system against oxidative stress [71]. Te present study recorded higher antioxidant enzymes activity in the biofoc based treatment units as compared to the control (Figure 2). However, within the biofoc based treatment groups, the enzyme's activity decreases with increase in nonfeeding intervals in pulsed feeding and the highest value was observed in no feeding treatment. Te results are in agreement with Kumar et al. [72] who noticed higher SOD and catalase activities in muscle and serum of shrimp fed with periphyton-incorporated diets. According to Luo et al. [52], SOD and CAT activity in GIFT tilapia was observed higher in fsh cultured in the BFT than RAS. Te antioxidative enzymes, SOD and CAT increased in the 2 and 3 days per week feeding groups of L. rohita in an adaptive response to cope with the oxidative stress caused due to feed deprivation [73] which is in contradiction with the present study which may be due to the biofoc being consumed in the absence of artifcial feed. Biofoc serves as a potential source of antioxidant containing an appropriate amount of carotenoids [65] and fat-soluble vitamins [74]. Tis agrees with the earlier reports where the higher immune response in terms of SOD and catalase activities were recorded in shrimps fed ß-glucan, carotene [75], microalgae [76], and macroalgal supplements [77]. Tese suggest that in situ biofoc with pulsed feeding can elicit the antioxidative enzymes in fsh and enhance the defense potential against oxidative stress. Prolonged starvation leads to enhanced oxidation and oxidative stress in the liver of D. dentex despite activation of some antioxidant defense mechanisms [78]. Hence, biofoc with no feeding can be the result of oxidative stress which is also concerned with the decreased survival rate observed in the treatment. Tough there was no signifcant diference in GPx among the treatments but followed the same trend as that of SOD and CAT activity (Figure 2(c)).
Metabolism is a physiological process refecting the energy expenditure of living organisms. Lactate dehydrogenase (LDH) assay serves as a useful stress indicator [79]. In fsh, there is a negative relationship between growth rate and LDH and MDH activity [80]. Te lower level of these enzymes was recorded when fsh and shrimps were fed with dietary supplements like tryptophan [81], pyridoxine [82], or periphyton [83]. Kumar et al. [72] observed lower LDH activity in L. rohita fed with high protein feed compared to low protein fed groups suggesting that higher dietary protein helps in reducing stress. Generally, the LDH activity increase during temperature stress [84], starvation stress [85], and confnement stress [79]. In the present study, signifcantly higher LDH activity was noticed in T4 indicating the starvation stress (Figure 3(a)). Lower LDH activity in the biofoc receiving pulsed feeding groups Aquaculture Research suggested that the supplementation of biofoc with feed helps in reducing stress in GIFT tilapia.
MDH is an enzyme which catalyzes the NAD/NADHdependent interconversion of the substrates malate and oxaloacetate. Te MDH enzyme activity decreases in proportion to the feeding rate. Hence, MDH plays an important role in the generation of NADPH for fatty acid synthesis [86]. Shikata and Shimeno [87] observed that the fatty acid synthesis in the hepatopancreas is markedly depressed by feed restriction including starvation due to the low reproduction rate of NADPH which may cause the decreased fatty acid synthesis. In the present study, the muscle lipid content was not afected and hence the MDH activity was not afected by the pulsed activity in GIFT tilapia under biofoc culture (Figure 3(b)). Kumar et al. [72] noticed no signifcant diference in MDH activity in the hepatopancreas of treatment groups compared with control indicating that dietary supplementation of biofoc helps to maintain shrimps in a less stressed condition and reduces energy demand in shrimps. Shimeno et al. [88] observed that MDH signifcantly decreases with decreasing feeding rates in metabolic response in common carp, Cyprinus carpio. Te fsh deprived of feed for 3 DPW had a lower activity of the metabolic enzymes MDH suggesting reduced metabolic activity in these groups, which might be a metabolic adaptation to cope with longer periods of feed deprivation [76].

Conclusion
Te present study demonstrates that biofoc reared GIFT tilapia under pulsed feeding enhances growth and improves immunity, hence this culture method will cut the feed cost without diminishing production and has obvious potential for production without compromising the survival. But the need for artifcial feed cannot be ignored in the intensive culture for higher fllet, growth performance and fnal production. Further research can be carried out on the identifcation of gut microbial content, higher salinity culture in inland saline waters and biofoc composition in inland saline water so as to utilize the resources in the arid zones.

Data Availability
Te data that support the fndings of the study are available from the corresponding author upon request.

Conflicts of Interest
Te authors declare that there are no conficts of interest.