Antagonistic impact of Lactobacilli on Escherichia coli in the gut of Labeo rohita fingerlings fed with bakery wastes

The present study was designed to check the feed conversion ratio (FCR) and, changes in gut microbiota of Labeo rohita fingerlings fed with different bakery product diets. The beneficial growth promoting lactobacilli and pathogenic Escherichia coli bacteria were isolated from the gut of Labeo rohita fingerlings and evaluated the antagonistic impact of lactobacilli on E. coli . The study was conducted for three months using five circular tanks of 5 feet diameter in two replicates. The experimental tanks were named as D1, D2, D3, D4 and D5 based on different types of diets. Fingerlings of the D1 tank were fed with control feed of Oryza Organics containing 20% crude protein. The D2 tank was fed with an experimental diet containing 25% cream cake as a basic ingredient. The D3 tank was fed with an experimental diet with 35% cream cake as a basic ingredient, while the D4 tank fingerlings were fed with a diet with 25% biscuits as a basic ingredient, and D5 tank fingerlings with a diet with 35% biscuits as a basic ingredient. The feed was provided according to the body weight of the fish twice a day. By using the staining method and biochemical test, lactobacilli and E. coli were identified from the gut of the fingerlings. To confirm the diagnosis of identified bacteria, 16S rDNA gene sequencing was used. Highest growth was observed (FCR, 1.200 ± 0.028) in D4 followed by that in D1 (1.360 ± 0.014), D2 (1.450 ± 0.014), D5 (1.512 ± 0.002), and D3 (1.595 ± 0.007), respectively. The antagonistic activity of lactobacilli against E. coli was also recorded. In diet D4, the lactobacilli concentration was highly significant, i.e., 9.7 x 10 7 ± 2.4 x 10 7 cfu/g and that of E. coli was found as 1.1 x 10 7 ± 1.4 x 10 6 cfu/g followed by D5 having 6.6 x 10 7 ± 1.6 x 10 7 cfu/g lactobacilli and 1.9 x 10 7 ± 2.5 x 10 6 cfu/g E. coli , D2 with 2.4 x 10 7 ± 1.9 x 10 6 cfu/g lactobacilli and E. coli 1.6 x 10 7 ± 2.1 x 10 6 cfu/g, D1 with 2.4 x 10 7 ± 1.9 x 10 6 cfu/g lactobacilli and E. coli 1.8 x 10 7 ± 1.8 x 10 6 cfu/g, and D3 with 9.1 x 10 6 ± 1.8 x 10 6 cfu/g lactobacilli and E. coli 1.4 x 10 7 ± 1.4 x 10 6 cfu/g.


Introduction
Aquaculture is an important industry and it primarily relies on feed, and only feed accounts for more than 60% of total cost incurred on this field.The provision and consumption of plentiful nutritionally balanced palatable feed by fish is the major module of success of fish culture.Like other terrestrial animals, balanced nutrients are required by fish to grow, reproduce, and maintain itself active.These nutrient requirements can be accomplished through a choice of food items, based on feeding behavior of an animal, and price and availability of food (Das et al., 1991;Alfred et al., 2020).Due to presence of higher protein contents, amino acid profile, vitamins, and mineral and fatty acid contents, fishmeal has been considered the best diet.While on the other hand, as substitutes of fish meal, plant protein sources such as soybean, cereals, oil-seed cakes and maize by-products have been practiced.Alternatively, 2.25 times higher energy contents exist in fat and oil with respect to that in cereals being a regular component of the fish feed (Markovic et al., 2016).
Due to escalating demand of fish meal and its limited supply, soybean meal, and oil-cakes are used extensively in aqua-feed.High market prices of these key ingredients have shifted pressure on cereal grains.Moreover, plant-based feed ingredients in pelleted and dough form are also acquired by fish.Therefore, it is the need of the hour to look at alternatives available in the country.There have been good efforts in research on finding sources of protein in aquaculture diets other than traditional ones (Kalita et al., 2008;Olsen and Hasan, 2012;Imran et al., 2018).
Nutritional requirements of fish are not only accomplished by utilizing waste materials as a fish feed, but it also lowers down the intensity of organic environmental pollutants.Presence of digestible carbohydrate contents (cookies, bread, cakes, rusks, etc.) and their mixture in bakery waste enables them to be used as energy sources.White meal is an essential component of bakery waste; it can be replaced with cereals in a limited amount (Hardy, 2010).In general, 2,981 kcal/kg of net energy is present in the bakery meal, which is almost close to that of maize, i.e., 2,672 kcal/kg (National Research Council, 2012).Some large-scale studies have been conducted on microbiota extracted from the gut of marine and freshwater fish (Cahill, 1990;Ringoe and Birkbeck, 1999;Hansen and Olafsen, 1999;Ghosh et al., 2002;Yukgehnaish et al., 2020).The microbial gut contents can be defined as an ability of indigenous autochthonous or transient allochthonous to stand and colonize the mucus layer of digestive tract (Ringoe and Birkbeck, 1999), and it examines that fish gastrointestinal (GI) tract containing bacterial population at dense level (Austin, 2006).In general, it is observed that diverse enzyme producing bacteria present in the gastrointestinal tract of host animals have association with metabolism (Rowland et al., 2018).So, it has been recommended by the scientists that gut microbiota might be favorable to fish in nutritional point of view (Ghosh et al., 2002;Kar and Ghosh, 2008).
Therefore, to develop fish quality and acceptability thereby improving aquaculture, it is essential to be aware of pathogenic gut microbiota which would allow determining their role in fish health.In some previous reports, pathogenic gut microbiota like fecal coliforms, Streptococci, Vibrio, Pseudomonas, Aeromonas and Enterobacter have been observed from the gastrointestinal tract of Rohu fish (Ghosh et al., 2010;Rahman et al., 2010).
All types of pathogenic bacteria can badly affect the economy of a country and can hinder the development of aquaculture by triggering health concerns to humans and fish.Many fishes are found to be consolidated with infectious pathogens, which can act as vectors of fish borne diseases with outbreak of human beings (Novoslavskij et al., 2016).
The present study was conducted with a primary objective to compare feed conversion ratio (FCR), and isolate the beneficial growth promoting lactobacilli and pathogenic E. coli bacteria from the gut of Labeo rohita fingerlings.The antagonistic impact of lactobacilli was analyzed on E. coli in the gut of Labeo rohita fingerlings fed with bakery wastes, and control feed of Oryza Organics (20% CP).Moreover, for molecular confirmation of the isolates was done by 16S rDNA sequencing.

Study area and experimental plan
The present study was conducted at the Department of Fisheries & Aquaculture UVAS, C-Block, Ravi Campus Pattoki.The experiment was conducted in 5-circular tanks each of 152 cm diameter with two replicates.The tanks were designated based on diets.Five types of feeds were used as experimental diets; fish in D1 was fed with the control diet of Oryza Organics with 20% CP; D2 and D3 were fed with 25% and 35 %, respectively, having cream cake as basic ingredient.While fish in the experimental groups D4 and D5 were fed with 25% and 35%, respectively, having biscuits as a key ingredient (Table 1).The tanks were disinfected with KMnO 4 before stocking.After acclimatization, the fish was stocked at biomass ratio of 1.0 g/gallon of water in each tank and optimum growth conditions were maintained up to 90 days.

Physicochemical parameters
In order to achieve best results, the physicochemical parameters such as dissolved oxygen (DO), pH, total dissolved solids (TDS), salinity, and electrical conductivity (EC) were monitored and maintained at an optimum range by applying appropriate measures.

Growth studies
To check the effect of feeds on growth performance of fingerlings of L. rohita, growth parameters including wet body weight, total length, gain body weight, feed conversion ratio (FCR), and feed conversion efficiency (FCE) were recorded weekly.

Sample collection for microbial analysis
From each treatment, 25 Labeo rohita fingerlings were dissected and gut content of each fish was homogenized in a saline solution prepared by dissolving 1.0 g of NaCl in 100 mL of distilled water and was serially diluted up to 6-fold.One µl of each dilution was poured on the Macconkey and MRS agar and incubated for 24 h at 37 °C.Colonies of E. coli formed on the Macconkey agar and lactobacilli on the MRS agar were enumerated.The colonies developed on cultured plates were identified morphologically using the gram staining technique (Bergey and Holt, 1994).Biochemical tests such as Gram staining, endospore staining, motility test, capsule staining, glucose fermentation test, oxidase test, catalase test, indole test, urease test and Voges-Proskauer test were performed for confirmation of bacteria at species level following the protocol described elsewhere (Sivasubramanian et al., 2012).

Antagonistic activity of gut isolates
The antimicrobial properties of the isolated bacterial strains of Lactobacillus and E. coli from Labeo rohita fingerlings fed with bakery wastes of different percentage were assessed using the well diffusion techniques previously described by Sivasubramanian et al. (2012).

DNA extraction and molecular study
DNA extraction of the isolates was carried out using kits (ZymoBIOMICS DNA KITS, USA) following the protocol of the manufacturer.Molecular identification of bacterial isolates was done using PCR primers targeting the eae gene specific for pathogenic E. coli, F (5'-TCAATGCAGTTCCGTTATCAGTT-3') and R (5'-GTAAAGTCCGTTACCCCAACCTG-3') (Miri et al., 2017).For PCR amplification of Lactobacillus group specific gene, F (5'-AGCAGTAGGGAATCTTCCA-3') R (5'-ATTYCACCGCTACACATG-3') primers were used (Rauta et al., 2013;Aslam et al., 2016).Agarose gel electrophoresis was used to separate mixtures of DNA according to molecular size by applying electric field and a gel documentation system (Michl et al., 2017).

Statistical analysis
The SAS statistical computer software version 9.1 was used for working out Pearson's Correlation matrix between the physicochemical parameters and fish morphometrics.Analysis of variance (ANOVA) for comparing the growth among the control and treated tanks was worked out, and the t-test was performed for comparing bacterial species (Eaves et al., 2004).

Physico-chemical properties
The parameters of all experimental groups were recorded twice a day throughout the trial.There was a significant (P < 0.05) difference in the values of physico-chemical parameters as shown in Figure 1.

Growth parameters
The results obtained from the growth parameters after analysis are given in Table 2.The (FCR) feed conversion ratio of the D4 (25% Biscuit) group was found the best, while other experimental groups were rated as sufficient.Comparison of all growth parameters is shown in Figure 2.

Proximate records
A significant difference (P < 0.05) was found between the treatments regarding growth parameters.The protein was recorded as 62% in D3, which was highly significant compared with those of D2 and D1 (Table 3).

Bacterial enumeration
The colonies of E. coli and lactobacilli were enumerated as cfu/g of the sample.The data in cfu/g of all samples are given below in Table 4.

Antagonistic activity
The antagonistic activity of lactobacilli against E. coli is shown in Table 5.The antagonistic effect was observed to be highest in D4 followed by that of D5.While D1, D2 and D3 showed a decreasing trend.

Isolation of E. coli and lactobacilli
From serially diluted plates used for enumeration, the putative colonies of E. coli and lactobacilli were isolated and pure cultured on TSA, EMB and Casein agar, incubated for 24 h at 37 °C, and then colonies of E. coli and lactobacilli observed.The bacteria produced colonies on a specific agar which are shown in Figure 3.

Biochemical test identification of E. coli and lactobacilli
The pure culture of each bacterium was used for the biochemical test identification and following results were observed as shown in Table 6 and Figure 4.

Discussion
In this study, we used bakery wastes, i.e., cream cake and biscuit each as 25% and 35% of feed ingredient and compared their growth effect with control (commercial) diet.The aim of the study was to utilize bakery wastes as value-added ingredients.After a successful feeding trial of 90 days, it was found that diet D4 (25% Biscuit) had the highest growth performance according to the growth data compared with the control and other treatment groups, followed by the control group which had the highest performance after D4.Whereas the remaining treatments had growth significantly lower than that of the control group, but still it was satisfactory.The final weight of D4 was highest among the treatment groups, i.e., 38.65 ± 0.070 (g).The other variables also showed a similar trend.However, FCR showed the results other way around (Table 2).
The effect of bakery wastes was also evaluated on intestinal microbiota of the Labeo rohita fingerlings.The bakery wastes enhanced the intestinal Lactobacillus production as previously studied by Jini et al. (2011).E. coli occurs as antagonistic to lactobacilli.Lactobacillus is a nonpathogenic naturally occurring microbe, while E. coli is a pathogenic and a naturally occurring bacterium in the intestine.The maximum lactobacilli were recorded in the D4 diet (biscuit 25%) 9.7 x 10 7 ± 2.4 x 10 7 cfu/g, followed by those in D5 (35% Biscuit) 6.6 x 10 7 cfu/g, D1 (Control diet) and D2 (Cream cake 25%) 2.4 x 10 7 cfu/g, and minimum were observed in D3 (Cream cake 35%) 9.1 x 10 6 cfu/g.Whereas E. coli were present maximum in D5 (35% Biscuit) 1.9 x 10 7 cfu/g followed by those in D1 (Control Diet) 1.8 x 10 7 cfu/g, D2 (25% Cream cake) 1.6 x 10 7 cfu/g, D3 (35 % Cream cake) 1.4 x 10 7 cfu/g (Table 4).These results were like those of Punom et al. (2017) who isolated and cultured E. coli along with other pathogenic bacteria.The total microbial count recorded by them was 5.24 ± 2.02 × 10 7 cfu/g from the gastrointestinal tract.
The antagonistic activity of lactobacilli against E. coli was also calculated and it was observed that the lactobacilli against E. coli were present in maximum number in the samples of fishes fed with diet D4 with 25% biscuits as basic ingredients with P value of 0.001165, followed by 0.008261 in D5 (Biscuit 35%), 0.010248 in D2 (Cream cake 25%), 0.024162 in D1 (Control diet) and 0.093606 in D3 (Cream cake 35%) as shown in Table 5.In the same way, Buntin et al. (2008) isolated 160 lactic acid bacteria from fishes, shrimps and mollusks and observed inhibitory activity against E. coli.The antibacterial activity of culture supernatants was low while treating with catalase between pH 6.5-7.0.The 16S rDNA had homology of 98% (492/501 bp) in Enterococcus faecium SF, 98% (655/668 bp) in Pediococcus pentosaceus SL4 and 97% (691/712 bp) in Pediococcus pentosaceus LM2.
Furthermore, for molecular identification of bacterial isolates, DNA extraction was done using a kit (ZymoBIOMICS DNA KITS, USA), and gene specific PCR was carried out for identification of E. coli and lactobacilli.The identified E. coli was the same as isolated previously (Miri et al., 2017).However, the identified isolates of lactobacilli were similar to the isolates extracted earlier (Rauta et al., 2013).

Conclusion
The overall findings of the current study showed that the bakery wastes, i.e., biscuits and cream cake are valuable feed ingredients and can be utilized in fish feed, especially with 25% concentration.These ingredients promoted a significant increase in the population of beneficial bacterial colony ensuring better feed intake and better digestibility, which in return, gave better FCR and FCE.

Figure 4 :
Figure 4: Biochemical tests for E. coli and lactobacilli Molecular identification of E. coli and Lactobacilli isolates DNA isolation DNA was isolated by the organic method with a slight modification by adding NaCl.After DNA isolation, isolated DNA was run on 1% agarose gel for DNA confirmation as shown in Figure 5.

Figure 6 :
Figure 6: PCR results of Lactobacillus group specific