PROBIOTIC POTENTIAL OF Bacillus cereus AGAINST Vibrio spp . IN POST-LARVAE SHRIMPS

Bacillus spp. have been used against diseases caused by bacteria that affect cultured shrimp, providing beneficial effects on the host shrimps by altering their microbial community, and improving zootechnical indexes. The objective of this work was to evaluate the effects of a diet supplemented with Bacillus cereus—a bacterium with probiotic potential—on post-larvae Litopenaeus vannamei shrimps grown in laboratory. The experiment lasted for fifteen days and consisted of six treatments—control (T1), probiotic (T2), Vibrio parahaemolyticus (VP) (T3), probiotic and VP (T4), V. alginolyticus (VA) (T5), and probiotic and VA (T6). The survival rate, weight gain, colonization capacity of the probiotic bacteria, pathogen count, and histopathological lesions were evaluated. There was no significant difference (p≥0.05) in survival between treatments. The groups with pathogens and without probiotics presented lower weight gain. The result of the Bacillus cereus count in the treatments T2, T4 and T6 were significantly different (p<0.05), the probiotic bacteria were more aggressive in competing for space and nutrients when compared to V. parahaemolyticus than when compared to V. alginolyticus. Animals fed with the probiotic presented lower counts of these pathogens than those fed without the probiotic (p<0.05). No histopathological lesions were found in the organs and tissues of the shrimps. Bacillus cereus showed a high colonizing capacity in post-larvae shrimps, causing a significant reduction of pathogens, probably by secreting antimicrobial substances and the competitive exclusion, which justifies their use as probiotic bacteria.


INTRODUCTION
Diseases that affect cultured shrimps are triggered by imbalances in the environmental conditions of the nursery, the health status of the shrimps, and potentially pathogenic agents (MERCIER et al., 2006).Vibrio spp.are among the bacteria that frequently cause diseases in cultured shrimps; they can infect them in all stages of life, and are responsible for mass mortalities in shrimp farms and, consequently, economic losses (AGUIRRE-GUZMÁN et al., 2010).
These pathogens have been controlled with the use of antibiotics, which brought negative consequences, especially bacterial resistance (THUY; NGA; LOAN, 2011).In the search for safe prophylactics, the use of probiotics emerged as an alternative to the use of antibiotics (BALCÁZAR et al., 2006).
Probiotics in aquaculture are cultured products or living microbial supplements that have beneficial effects on the hosts by altering their intestinal microbial community, improving their responses to diseases (VERSCHUERE et al., 2000), and increasing their appetite, promoting a better growth and feed conversion (NEWAJ-FYZUL; HARBI; AUSTIN, 2014).Live probiotics offered as supplements in diets must to be able to survive and pass through the animal's intestinal tract (NAVINCHANDRAN et al., 2014).
Results of pathogen antagonism tests, adhesion capacity and, consequently, colonization in the intestinal tract must be considered in the selection of probiotics (LUIS-VILLASEÑOR et al., 2011), as well as their ability in inhibiting or reducing the colonization of Vibrio bacteria (CHABRILLÓN et al., 2005).Moreover, the microorganism must be non-pathogenic, biochemically, and physiologically well characterized, and preferably isolated from the intestine of juveniles of the species itself, so that the microorganism will be more adaptable to its conditions (BALCÁZAR et al., 2006).
Gram-negative and gram-positive bacteria have been identified as potential probiotics for aquaculture, with effects against various pathogens (BRUNT; NEWAJ-FYZUL; AUSTIN, 2007).Gramnegative, spore-forming bacteria such as Bacillus spp.have shown considerable success as probiotics (HONG;DUC;CUTTING, 2005).The species B. subtilis, B. clausii, B. cereus, B. coagulans and B. licheniformis are the most studied.These species form spores, thus, they are heat stable and can be stored at room temperature as dehydrated products without any deleterious effect affecting their viability.Therefore, these probiotics are widely used in humans as dietary supplements and in aquaculture to increase growth and disease resistance in cultured shrimps (CUTTING, 2011).
The objective of this work was to evaluate the effects of a diet supplemented with Bacillus cereus as a probiotic on the post-larvae shrimps (Litopenaeus vannamei) survival, weight gain, intestinal microbiota, and histological conditions, while challenged with pathogens of the Vibrio genus.

MATERIAL AND METHODS
Ten-day-old post-larvae shrimps were obtained from a commercial laboratory in the state of Rio Grande do Norte, Brazil.The animals were kept for 14 days in a nursery, in a 1.0-m 3 water box, at a density of 24 post-larvae per liter (PL L -1 ), constant aeration, temperature of 28.0±0.3°C, and salinity of 28 mg L -1 .They were fed six times a day with a commercial feed consisted of 56% crude protein, according to the observed consumption.They were treated with a medicated feed with oxytetracycline (purity of 50%) at dose of 0.4 g kg -1 to ensure that the animals were free of pathogens during this period.The drug was incorporated into the diet following the method described by Brock and Main (1994).The medicated feed was offered in two of the daily meals for seven days.Subsequently, the post-larvae shrimps remained in the nursery for another five days, which was the time required for the starting of the experiment due to the antibiotic application.
The experimental plots consisted of 5-liter plastic containers, with 4.3 L of water containing 130 shrimps-density of 30 post-larvae per liter-with a mean weight of 9 mg and age of 24 days.The water used in the containers was previously filtered, sterilized in ultraviolet light, chlorinated at 15 ppm for 24 hours, and dechlorinated with ascorbic acid (5 ppm).A constantly mild aeration with porous stone was maintained in each experimental plot, and part of the water (10%) of each container was exchanged daily.A feed with 40% crude protein was offered six times a day in amounts depending on the observed consumption.
The treatments T 2 (PB), T 4 (PB-VP) and T 6 (PB-VA) received feed supplemented with B. cereus, isolated from the intestine of juvenile shrimps (L.vannamei).The B. cereus strain was added to the feed at a concentration of 1.0×10 8 CFU g -1 , as suggested by Vieira et al. (2007) and recommended in commercial products (CUTTING, 2011).
On the 7 th day of the experiment, the T 3 (VP), T 4 (PB-VP), T 5 (VA) and T 6 (PB-VA) treatments were challenged with the pathogenic bacteria, which was added to culture water at a concentration of 1.0×10 8 CFU mL -1 , as recommended by Buglione et al. (2008).The experiment was completed seven days after this addition.

Survival and weight gain
Survival (%) was determined at the end of the experiment by counting the final post-larvae and divided by the final storage density-difference between the number of post-larvae stored and the number of post-larvae collected during the experiment for analysis (Eq. 1).
Weight was calculated using data from seven biometrics performed throughout the experiment, on the 1 st , 4 th , 7 th , 9 th , 11 th , 13 th and 15 th days of culture.
The excess moisture of the shrimps was removed with an absorbent paper, the animals were weighed in a digital scale with an accuracy of 0.0001 g, and the number of individuals in a biomass of 0.1 g was counted (Eq.2), (1) (2) wherein Sur (%) is the survival rate, FNPL is the final number of post-larvae, FPLD is the final post-larvae density, PLW (g) is the mean post-larvae weight per gram, NPL is the number of post-larvae.

Microbiological analyzes
Microbiological analyzes were performed using the standard plate count method.The colonization capacity of the B. cereus in post-larvae shrimps was evaluated by quantitative determination of the Bacillus at the beginning and throughout the experiment-1 st , 4 th , 9 th and 15 th days of culture.
The health status of the animals was evaluated through quantitative determination of the pathogenic bacteria (Vibrio spp.) in the animals in the first day of experiment, and in the 9 th and 15 th days after adding the pathogenic bacteria.The bacteria of the Vibrio genus were counted in the culture water at the beginning of the experiment to ensure that it was free of pathogens.
The concentration of bacteria in the animals was estimated using a shrimp biomass sample of 0.1 g that was washed in 70% ethyl alcohol, then in sterile distilled water, and in a saline solution (3.5% NaCl) to remove possible external bacteria.The samples were macerated and diluted in 0.9 mL of a sterile saline solution (dilution 10 -1 ), and then successive dilutions were performed up to 10 -4 .The samples were plated according to Buller (2004) in a Man-Rogosa-Sharpe agar medium to count the Bacillus, and in a Thiosulfate Citrate Bile Sucrose agar medium to count the Vibrio spp., and incubated at 35 °C for 24 to 36 hours and at 37 °C for 18 to 24 hours, respectively.The bacteria grew, the colonies were counted, and the results were expressed in Colony Forming Units per gram (CFU g -1 ).
The Vibrio bacteria in the culture water were counted by performing successive decimal dilutions of the water sample followed by plating, according to the same methodology described for the analysis in the animals.

Histopathological analysis
Samples were collected from all treatments for histopathological analysis.Seven collections were carried out throughout the experiment on the 1 st , 4 th , 7 th , 9 th , 11 th , 13 th and 15 th days of culture.The animals were fixed in a Davidson's AFA solution (HUMASON, 1972) for 24 to 48 hours, and then transferred to a 70% alcohol solution.The material was subjected to dehydration and diaphanization, embedded in paraffin, cut with a microtome at a thickness of 5 microns, and stained with hematoxylin-eosin (BEHMER et al., 2003).

Statistical analysis
The survival rates of the post-larvae shrimps in all the treatments were subjected to analysis of variance (ANOVA) (α=0.05), after homogeneity evaluation by the K test (Cochran), and the means were subjected to the Tukey's test to assess significant differences (p≤0.05) between treatments.
The following mathematical model was used to evaluate the relationship of post-larvae weight gain with density, time of culture, and treatments: (3) wherein Weight is the average weight per post-larva sample (0.1 g), β 0,1,2,3, ... 9 are the model parameters, TC is the time of culture; SD is the storage density, Trat n is the treatment-T1 (CT), T 2 (PB), T 3 (VP), T 4 (PB-VP), T 5 (VA) and T 6 (PB-VA)-, and ε is the error with normal distribution and parameters (0, σ 2 ).The Stepwise process (selection of variables) was used to verify the effect (p<0.05) of each variable of the model, and the robustness of the model was evaluated at the end of the process based on the Snedecor's F-statistic, probability value of F, R 2 , normality of errors (Shapiro-Wilk), and number of discrepant points (outlier).
The results of Bacillus bacterial counts in post-larvae shrimp (L.vannamei) in T 2 (PB), T 4 (PB-VP) and T 6 (PB-VA), at the 4 th , 9 th and 15 th days of culture, and the results of counts of Vibrio spp. in T 3 (VP), T 4 (PB-VP), T 5 (VA) and T 6 (PB-VA), at the 9 th and 15 th days of culture, were subjected to analysis of variance (ANOVA) after homogeneity evaluation through the K test (Cochran).Means with significant differences between treatments and between times of culture were compared by the t-Student test at 5% probability.The values of the counts were transformed to Ln (x +1) for the analyses.All statistical calculations were performed using the SysEapro (v.2) software.

RESULTS AND DISCUSSION
The means of temperature, pH, salinity, dissolved oxygen, alkalinity and concentrations of nitrogen compounds-total ammonia, nitrite and nitrate-of the treatments (Table 1) were adequate for the culture of Penaeidae shrimps, according to Boyd (1990).Thus, the water quality did not compromise the development of the shrimps during the experiment.

Parameters
Mean±standard

Survival and weight gain
The treatment with the highest survival of post-larvae shrimps at the 15 th day of experiment was T 2 (PB), with a rate of 60.52±23.71%,whereas the treatments with pathogens showed the lowest rates.However, no significant difference (p≥0.05) was found between the treatments (Table 2).Thus, the use of B. cereus as a probiotic did not contribute significantly to the survival of the animals during the experiment.Although, the treatments with the probiotic showed higher survival rates than the control groups without probiotic.These results confirm those found by Guo et al. (2009) for larvae of Pennaeus monodon fed with the probiotic Bacillus fusiformis; they found no significant difference in the survival rate when compared to the control group (without probiotic).
However, increased survival rates of shrimps were observed by other authors using probiotics.The results for survival rate in the present study are probably due to the control of water quality in all treatments, and the adequate food conditions during the experiment, which caused no physiological stress to the animals.
Table 2. Effect of treatments without probiotic and pathogen (CT), with probiotic (PB), with V. parahaemolyticus (VP), with probiotic and V. parahaemolyticus (PB-VP), with V. alginolyticus (VA), and with probiotic and V. alginolyticus (PB-VA) on the average survival rate of post-larvae shrimps (L.vannamei) grown for 15 days in laboratory.The time of culture positively affected (p<0.05) the weight gain of the post larvae shrimps (L.vannamei), while the storage density and the T 3 (VP) and T 5 (VA) negatively affected (p<0.05)their weight gain, based on the Eq. 3 that resulted in the model: wherein Weight (g) is the weight of post-larvae in grams, TC is the time of culture, SD is the storage density, T 3 is the treatment 3-without use of probiotic and challenged with V. parahaemolyticus (VP)-, T 5 is the treatment 5-without use of probiotic and challenged with V. alginolyticus (VA).
This model was considered robust, since it presented the following results: R 2 = 73.65%,Probability of the Snedecor's F-statistic <0.00001, six outliers, and normality of errors within the acceptable range by the Shapiro-Wilk, D´Agostino, and D´Agostino-Pearson tests.Thus, the estimates and comparisons of the results showed that, for a time of culture of 15 days and a mean of 78 individuals per experimental unit, the animals from the treatments with pathogens, but with probiotics and the controls did not differ significantly (p≥0.05)presenting the best weight gains.Contrastingly, T 3 (VP) and T 5 (VA)-challenged with V. parahaemolyticus and V. alginolyticus, without using probiotics, respectively-were the ones that presented the lowest weight gain, and presented significant differences (p<0.05) from the other treatments (Table 3).
Table 3.Effect of treatments without probiotic and pathogen (CT), with probiotic (PB), with V. parahaemolyticus (VP), with probiotic and V. parahaemolyticus (PB-VP), with V. alginolyticus (VA), and with probiotic and V. alginolyticus (PB-VA) on the weight gain of post-larvae shrimps (L.vannamei) grown for 15 days in laboratory.
The presence of pathogens in treatments without the probiotic negatively affected the shrimp weight gain.Balcázar, Rojas-Luna and Cunninghan (2007) also observed beneficial effects with the use of feed with probiotics isolated from the gastrointestinal tract of L. vannamei, against experimental infection with V. parahaemolyticus.Zokaeifar et al. (2014) found a greater weight gain with the use of probiotics in the culture water of L. vannamei challenged with V. harveyi.According to Navinchandran et al. (2014), probiotic bacteria produce digestive enzymes, and necessary nutrients for growth, such as vitamins and amino acids, and improve food absorption, resulting in a better growth rate of the host.This was observed in the present work, since treatments with Vibrios spp.and without probiotics presented the lowest weight gain.

Microbiological analyzes
The pathogenic bacteria of the Vibrio genus did not develop in the culture water and animals, and ℎ () = 0.00930 + 0.00252 − 0.00002 *  − 0.00286 3 − 0.00395 5 the bacteria of the Bacillus genus did not develop in the post-larvae of L. vannamei, in the first day of experiment.The Vibrio bacteria did not grow in T 1 (CT), T 2 (PB), and the Bacillus bacteria did not grow in the T 1 (CT), T 3 (VP) and T 5 (VA), throughout the experiment.
The colonization of a probiotic in the gastrointestinal tract of an animal is a very important factor for its intestinal balance (ZOKAEIFAR et al., 2014).The adhesion capacity and consequent colonization of bacteria in the gastrointestinal tract, and the ability of inhibiting or reduce colonization of Vibrio spp.must be considered when choosing a probiotic (CHABRILLÓN et al., 2005).
The administration of feed supplemented with B. cereus was successful regarding its colonization, since it was possible to recover it from the animal samples from all groups treated with the probiotic.
The counts of Bacillus bacteria in post-larvae shrimps from treatments with supplementation of Bacillus cereus are presented in Table 4.The count of bacterial probiotics remained stable in T 2 (PB)treatment with no pathogens-, with no significant difference (p≥0.05)throughout the experiment.In T 4 (PB-VP) and T 6 (PB-VA)-treatments with V. parahaemolyticus and V. alginolyticus added on the 7 th day of experiment, respectively-, the count of Bacillus reduced on the 9 th day; this was probably due to the presence of pathogens, and occurrence of competitive exclusion.
The three treatments were significantly different (p<0.05) on the 15 th day of experiment; the Bacillus bacterial count in T 4 (PB-VP) (5.83±1.18CFU mg -1 ) was higher than in T 6 (PB-VA) (4.90±0.42CFU mg -1 ).This indicates that the probiotic bacteria were more aggressive in competing for space and nutrients when compared to V. parahaemolyticus than when compared to V. alginolyticus in post-larvae shrimps (L.vannamei), and promoted a positive substitution of harmful bacteria to the animals, such as V. parahaemolyticus, by beneficial bacteria, such as B. cereus.The results of the counts of pathogenic bacteria in the animals are presented in Table 5.In the first bacteriological analysis for Vibrio in the 9 th day of culture, no significant difference was found between treatments.However, differences (p<0.05) were found between the treatments T 3 (VP) and T 4 (PB-VP), and between T 5 (VA) and T 6 (PB-VA) at the end of the experiment (15 th day).
Regarding the time of culture, the counts of bacteria in the 9 th and 15 th days of culture showed a significant difference (p<0.05) between the treatmentsT 4 (PB-VP) and T 6 (PB-VA) .Thus, the groups of post-larvae fed with feed supplemented with probiotic had lower pathogen counts than those fed without the probiotic.This is probably connected to the antimicrobial action of the Bacillus cereus against V. parahaemolyticus and V. alginolyticus, confirming the findings of Vieira et al. (2007).
According to Abriouel et al. (2011), bacteria of the Bacillus genus produce many antimicrobial substances, including peptidic, lipopeptide antibiotics, and bacteriocins, the latter is very important, due to the broad spectrum of bacteriagram-negative and gram-positive-, yeasts and fungi that it can inhibit.Moreover, according to Verschuere et al. (2000), Bacillus spp.are able to compete with pathogenic bacteria for nutrients and space, increasing their proportion in the intestinal microbiota of shrimps.

Histopathological analysis
No relevant morphological lesions were found in organs or tissues of the post-larvae shrimps in all treatments throughout the experiment when using 108 CFU mL -1 of pathogenic bacteria.Some specimens presented mild peripheral hemocytic infiltrations in the hepatopancreas, which could not be connected to the groups challenged with pathogens.Contrastingly, Soto-Rodriguez et al. (2015) found lesions, such as elongation and atrophy of the tubular epithelial cells, hemocytic infiltration, and necrosis in experimental infection with V. parahaemolyticus.
Despite the high counts of Vibrio spp. in the animals from the treatments challenged with V. parahaemolyticus and V. alginolyticus, no clinical signs or lesions could be connected to these Vibrios spp.Several factors may have caused the absence of alterations, such as the short period of exposure to the pathogen, and the maintenance of environmental conditions-pH, salinity, dissolved oxygen, nitrite, nitrate, and ammonia.According to Mugnier et al. (2013), sufficiently stressful factors external to the animal could anticipate the incubation period of the agent, showing earlier clinical signs.

CONCLUSION
B. cereus isolate from the intestine of shrimps Litopenaeus vannamei can colonize the intestine of post-larvae shrimps of the same species and promote a significant reduction of pathogens, with great effectiveness in reducing the pathogenic bacteria V. parahaemolyticus and V. alginolyticus in shrimps grown in laboratory.Washington, v. 64, n. 4, p. 655-671, 2000.VIEIRA, F. N. et al.Lactic-acid bacteria increase the survival of marine shrimp, Litopenaeus vannamei, after infection with Vibrio harveyi.Brazilian Journal of Oceanography, São Paulo, v. 55, n. 4, p. 251-255, 2007.ZOKAEIFAR, H. et al.Administration of Bacillus subtilis strains in the rearing water enhances the water quality, growth performance, immune response, and resistance against Vibrio harveyi infection in juvenile white shrimp, Litopenaeus vannamei.Fish & Shellfish Immunology, Amsterdam, v. 36, n. 1, p. 68-74, 2014.This work is licensed under a Creative Commons Attribution-CC-BY https://creativecommons.org/licenses/by/4.0/

Table 1 .
Means±standard deviations, and minimum and maximum variations of the physical-chemical parameters of the culture water of post-larvae shrimps (L.vannamei).

Table 4 .
Probiotic bacteria count in post-larvae shrimp samples grown for 15 days in laboratory, under diet supplemented with B. cereus, and challenged with Vibrio spp.bacteria.Ln CFU mg -1 = log of Colony Forming Units (CFU) counts per milligram.Means followed by different lowercase letters in the same column and different uppercase letters on the same row differ significantly (p<0.05) by the t-Student test at 5% probability; the data were transformed to Ln (x+1).

Table 5 .
Pathogenic bacteria count in post-larvae shrimp samples grown for 15 days in laboratory, under diet with feed supplemented and non-supplemented with B. cereus, and challenged with Vibrio spp.bacteria.T (PB-VP) with probiotic and V. parahaemolyticus, T 5 (VA) with V. alginolyticus, and T 6 (PB-VA) with probiotic and V. alginolyticus.Ln CFU mg -1 = log of Colony Forming Units (CFU) counts per milligram.Means followed by different lowercase letters in the same column and different uppercase letters on the same row differ significantly (p<0.05) by the t-Student test at 5% probability; the data were transformed to Ln (x+1).