Introduction

With the increase in seafood consumption around the world, the occurrence of seafood safety events presents a rising trend in recent years (Fernandez-Piquer et al. 2011; Xu et al. 2012). Acute gastroenteritis caused by pathogenic Vibrio spp. is one of the major factors limiting the development of the marine economy. Among them, Vibrio parahaemolyticus has become a primary causative factor of food-borne disease outbreaks in coastal countries worldwide (Zarei et al. 2012), including China, Japan, India, Thailand, Australia, and the United States (Wu et al. 2014; Raghunath et al. 2008; Iwamoto et al. 2010; WHO/FAO 2011; Yu et al. 2013): this affects the safety of seafood and the health of consumers. Vibrio parahaemolyticus is a natural bacterium in marine and estuarine environments, the animals carrying V. parahaemolyticus could be the principal vehicle transmitting the pathogenic bacterium to humans, including aquatic farmed animals (Griffitt et al. 2011).

Currently, bacterial diseases in aquatic farmed animals are generally controlled by antibiotics. The long-term use of antibiotics can cause environment pollution and threaten human health. Using probiotics antagonistic to V. parahaemolyticus is an alternative strategy to antibiotics and a common trend to control the food-borne pathogen V. parahaemolyticus (Touraki et al. 2012; Aranda et al. 2012). The concept of probiotics has been constantly enriched from “…a live microbial feed supplement…” (Fuller 1989) to “…any microbial probiotics or the product of the microorganisms (not necessarily live) that are beneficial to the health of the host…” (Salminen et al. 1999). The effect of probiotics on aquatic pathogenic bacteria was mainly manifested in the following ways: competitive exclusion (binding sites, nutrition, energy, and iron carrier), digestion enhancement in its host, production of antimicrobial active substances (such as friendly antibiotics, bacteriolytic enzymes, and bacteriocins), strengthening of aquaculture animal non-specific innate immunity, water quality improvement, etc. (Newaj-Fyzul et al. 2014).

Bacillus species is the oldest type of bacterium known and is widely distributed in the natural world, including marine environments (Nithya et al. 2010). This genus of bacteria has been differentiated in more detail by new classification methods such as 16S rRNA sequence analysis. Many species within this genus have been proven to be safe, even as fermentation strains of food and as probiotics drugs for oral consumption (Bacon and Hinton 2002). Some reports stated that Bacillus could produce biological control agents and has become one of the most important sources of probiotics (Ravi et al. 2007; Barman et al. 2011; Zokaeifar et al. 2012; Donio et al. 2014). However, the application of Bacillus spp. as a probiotic in aquaculture remains limited.

To develop Bacillus spp. into a probiotic to control pathogenic V. parahaemolyticus in aquaculture, strains were isolated and screened from seafood and the coastal sediment from Liaodong Bay of Bohai Sea, China. The active isolates were identified and investigated for hydrolytic exoenzymatic activity and their antagonistic spectra respectively. The bacteriostatic activity of the culture and the supernatant of active isolates were surveyed to uncover novel antibacterial ingredients.

Materials and methods

Sample collection

Five different coastal sediment samples were collected from the exposed intertidal zone of Liaodong Bay, in the north of the Bohai Sea, Jinzhou, China, using a shovel, at a temperature of around 3 °C in March, 2013. Each cuboidal sediment sample measured approximately 8 cm (side length) and they were taken from different sampling sites including: 40°49′29.64″N, 121°03′53.10″E, 40°49′41.07″N, 121°04′26.61″E, 40°49′13.67″N, 121°04′15.62″E, 40°48′53.56″N, 121°04′25.59″E, and 40°48′40.34″N, 121°04′ 36.87″E. Seafood samples (5 kg each) of Litopenaeus vannamei, Oratosquilla oratoria, Mactra veneriformis, and Portunus trituberculatus were obtained from a local aquatic products wholesale market in Jinzhou. The four types of seafood were selected because they were among the favourite local foods and were easily obtained. Each separate sample was placed in a sterile plastic bag without seawater and transported to the laboratory within 2 h and processed immediately to isolate the relevant microorganisms.

Isolation of bacteria

The coastal sediment and the intestines of four types of seafood and the gills of two types (O. oratoria and P. trituberculatus) were selected as experimental samples. Five different sediment samples were individually removed from the outer layer with a sterile scalpel and then mixed evenly. Four kinds of seafood were aseptically shucked to collect the intestines or the gills separately. Some 25 g of each sample (wet mass) were homogenised in 225 ml 0.85 % sterile saline solution following tenfold serial dilution to 104 times. Some 100 μl of each dilution was spread on five specialised media plates respectively, and in triplicate, to get as many bacteria as possible. The five types of separation medium respectively were as follows: (1) all purpose Tween (APT) agar (Qingdao Hope Bio-Technology Co. Ltd, Qingdao, China), (2) thiosulphate citrate bile saccharose (TCBS) agar (Qingdao Hope Bio-Technology Co. Ltd, Qingdao, China), (3) nalidixic acid cetrimide (NAC) agar (Beijing Land Bridge Technology Co. Ltd, Beijing, China), (4) marine agar 2216E (2216E, prepared by the authors), and (5) manganese nutrient agar (Mn2+-NA, prepared by the authors). Their compositions are listed in Table 1. All of the plates were incubated at 28 °C for 24–48 h until the morphology of the colony could be distinguished.

Table 1 Compositions and concentrations of five types of separation medium used in this study
Full size table

Direct antimicrobial activity assay

The antimicrobial activity of all isolates against V. parahaemolyticus was assayed with spot inoculation. The V. parahaemolyticus, with its thermo-stable direct hemolysin related hemolysin (tdh−/trh+) gene (ATCC 17802), was incubated in alkaline peptone water (APW, Qingdao Hope Bio-Technology Co. Ltd, Qingdao, China) and shaken at 180 rpm and 28 °C overnight and diluted to 105 CFU/ml with 0.85 % sterile saline solution for use as an indicator in the direct antimicrobial activity assay. All isolated strains were spot inoculated (Chahad et al. 2012) in triplicate on nutrient agar (Beijing Land Bridge Technology Co. Ltd, Beijing, China) plates coated with V. parahaemolyticus in advance and incubated at 28 °C for 24–48 h. The inhibitory zone diameters were measured with a Vernier caliper.

Extracellular antimicrobial activity assay

The antimicrobial activity of cultures and supernatants from the isolates were determined by the Oxford cup method. In detail, the isolates antagonistic to V. parahaemolyticus in the direct antimicrobial activity assay were respectively inoculated to a 250 ml conical flask filled with 100 ml of nutrient broth (Beijing Land Bridge Technology Co. Ltd, China) and incubated at 28 °C, while being shaken at 180 rpm for 24 h. A part of the culture was collected as one of the samples. Another part of the culture was centrifuged at 8,000 rpm (Thermo Scientific Sorval Biofuge Stratos 75005289, Germany) at 4 °C for 10 min. The supernatant was sterilised through 0.22 μm pore-size filters as for the other sample. Some 180 μl of the culture and the supernatant were added into the Oxford cup in the nutrient agar inoculated with 105 CFU/ml of indicator bacterium V. parahaemolyticus in triplicate respectively and the same amount of liquid medium was used as a blank control. All the plates were incubated at 28 °C for 24 h and the size of the inhibiting zones around the cup was measured thereafter. Those strains exhibiting extracellular antimicrobial activity were stored at −80 °C with 15 % glycerol as potential probiotics.

Morphological and physiological characteristics of potential probiotic strains

The potential probiotic strains were studied to determine their morphological (Gram staining), cultural (colony), and physiological (biochemical test) characteristics. Physiological identification was undertaken as described by Bergey’s Manual of Systematic Bacteriology using a trace biochemical identification reaction tube (Hangzhou BinHe Microorganism Reagent Co. Ltd, Hangzhou, China). The results of biochemical tests were interpreted according to the identification code book (Hangzhou BinHe Microorganism Reagent Co. Ltd, Hangzhou, China).

16S rRNA gene amplification, sequencing, and analysis of potential probiotic strains

The DNA template of potential probiotic strains was obtained using a bacterial genome DNA rapid extraction kit (Sangon Biotech Co. Ltd, Shanghai, China). Polymerase Chain Reaction was carried out using the 16S universal primers: 16S-27F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 16S-1492R (5′-TACGGCTAC CTTGTTACGACTT-3′). The primers were synthesised by Sangon Biotech Co. Ltd, Shanghai, China. PCR was performed in a 50 μl reaction system consisting of 25 μl 2 × Taq PCR Master Mix (Sangon Biotech Co. Ltd, Shanghai, China), 1 μl DNA template, 2 μl Primer F, 2 μl Primer R, and 20 μl sterile ddH2O. The PCR was performed under the conditions pertaining to: initial denaturation at 94 °C for 2 min, denaturation at 94 °C for 1 min, renaturation at 60 °C for 1 min, extension at 72 °C for 95 s with a total of 30 cycles, and the final extension at 72 °C for 10 min. The purity of PCR products was observed by 1 % agarose gel electrophoresis. The sequencing of the 16S rRNA gene was performed by Sangon Biotech Co. Ltd, Shanghai, China, with 27F and 1492R primers. The RNA sequences were analysed through http://www.ncbi.nlm.nih.gov/ blast by comparing them with bacterial 16S rRNA sequences in Genbank to determine the genus to which each isolate belonged. The 16S rRNA sequences of standard strains within this genus were downloaded and aligned with those of potential probiotic strains using MEGA 5.05 software. The phylogenetic tree was constructed with a neighbour-joining DNA distance algorithm to determine the evolutionary distance between strains.

Enzymatic assay

The hydrolytic exoenzymatic activities of potential probiotics were further detected by using the agar diffusion method. Nutrient agar, as the experimental medium, was added to different substrates according to Nair et al. (2012) with minor modifications, including skimmed milk for proteolytic activity, soluble starch for amylolytic activity, and 1 % Tween 80 for lipolytic activity, respectively. The hydrolytic exoenzymatic activity of the strains inoculated on the plates were observed after incubating at 28 °C for 24 h. Proteolytic activity and lipolytic activity were judged by the transparent circle of milk degradation and the halo formation of precipitated fatty acids around the colony. However, amylolytic activity was tested with 1 % iodine solution.

Antagonistic spectrum of potential probiotic strains

The antagonistic spectrum of the culture and the supernatant of potential probiotics were investigated using other aquaculture or human pathogens, including V. parahaemolyticus with the thermostable direct hemolysin (tdh+/trh−) gene (ATCC 33847), Vibrio vulnificus (environmentally isolated), Vibrio harveyi (environmentally isolated), Salmonella typhimurium (CMCC 50115), Pseudomonas aeruginosa (ATCC 9027), Aeromonas hydrophila (environmentally isolated), Listeria monocytogenes (ATCC 19115), Bacillus subtilis (CMCC 63530), Escherichia coli (CMCC 44102) and Staphylococcus aureus (CMCC 26003). These additional assays were performed with Oxford cup method by using nutrient agar medium and incubated at 28 °C for 24 h. Considering that the majority of isolates were from the natural environment, we selected environmental microbial cultivation temperatures. The pathogens mentioned above were incubated separately in the nutrient broth and shaken at 150 rpm and 37 °C overnight except for Vibrio spp. for which a temperature of 28 °C was used.

Results and discussion

Effect of samples and separation media on the recoverability of bacteria

A total of 249 bacteria were isolated from five samples of coastal sediment and the intestines or gills of L. vannamei, O. oratoria, M. veneriformis, and P. trituberculatus by using five types of separation media of APT, TCBS, NAC, 2216E, and Mn2+-NA respectively. According to the sample source and separation media, all isolates were named A to P in sequence. The number of recovered bacterial isolates was different between each of the five samples and five types of separation medium (Fig. 1) based on colony morphology. The coastal sediment harboured the most isolates with 90 bacteria (accounting for 36.14 % of all the separation microbes), followed by P. trituberculatus with 53 isolates (21.29 %), O. oratoria with 44 isolates (17.67 %), M. veneriformis with 35 isolates (14.06 %), and L. vannamei with 27 isolates (10.84 %). In comparison with the samples of four types seafood, more colonies forming on the plates were obtained from coastal sediment samples, which was consistent with the results of Nair et al. (2012), so that the sediment sample was selected for the separation of marine microorganisms in many studies (Annamalai et al. 2014).

Fig. 1
figure 1

Number of bacteria isolated from five samples using five separation media. Note: samples 1–4 were from the intestines or the gills of L. vannamei, O. oratoria, M.veneriformis, P. trituberculatus, respectively, sample 5 was from the coastal sediment

Full size image

The separation media used has a direct impact on the recoverability of microorganisms (Zhang et al. 2013). In this study, the separation medium of 2216E produced the most isolates (136) while NAC produced the fewest (16) accounting for 54.62 and 6.43 % of all separated isolates (249) respectively. APT, Mn2+-NA, and TCBS produced 38, 32, and 27 isolates accounting for 15.26, 12.85, and 10.84 % respectively. 2216E was the best separation medium for bacterial recovery from coastal sediment and seafood samples, which may have been because 2216E is an exclusive medium for marine bacteria isolation and, when prepared with seawater containing a variety of metal ions and an appropriate salt, as opposed to distilled water, meant that 2216E was the closest match to the original growth environment of these marine microorganisms (Nair et al. 2012). The majority of microorganisms from the experimental samples came from a marine environment except for the sample from the gills of O. oratoria.

Direct antimicrobial activity of 249 isolates

All of the 249 bacterial isolates were screened for their direct antimicrobial activity against V. parahaemolyticus with spot inoculation. A total of 24 strains displayed antagonistic activity (Table 2) accounting for 9.64 % of the separation isolates. The direct antimicrobial activity assay for screening potential probiotic strains was performed on nutrient agar: if the strains needed special nutrients for producing antimicrobial metabolites, antagonism could not be observed.

Table 2 The origin, separation medium, and inhibitory zone diameter (mm) of direct antimicrobial activity assay with spot inoculation and extracellular antimicrobial activity assay by Oxford cup method for 24 active isolates against V. parahaemolyticus in vitro
Full size table

Extracellular antimicrobial activity of the 16 selected isolates

Sixteen of 24 active isolates were selected for extracellular antimicrobial activity against V. parahaemolyticus using the Oxford cup method. The other eight isolates (A2, A4, A8, A16, B15, F15, F16, and J9) were not studied any further for their poor inheritance stability (data not shown). The results of extracellular antimicrobial activity assays of 16 selected isolates are listed in Table 2. Among 16 selected active isolates, two strains (B16 and J7) displayed extracellular antimicrobial activity for the supernatant of B16 and J7 and the formation of inhibiting zones (Fig. 2). So strains of B16 and J7 were the most suitable candidate probiotics antagonistic to V. parahaemolyticus for sustainable aquaculture.

Fig. 2
figure 2

The antagonistic activity of the supernatant of potential probiotics B16 (A) and J7 (B) against V. parahaemolyticus in vitro. Note The left two holes were the supernatant of isolate B16 (A) and J2 (B), the right two holes were the supernatant of isolate B15 (A) and J7 (B). Each isolate of the third replica was in another plate

Full size image

The other 14 strains (O8, O9, O12, O15, M1, L1, L5, L6, J2, J8, J11, J13, J18, and J21) with direct antimicrobial activity had no extracellular antimicrobial activity. Their supernatant could not form inhibiting zones. Interestingly, the culture of eight strains of O12, O15, M1, L1, L5, L6, J18, and J21 showed antagonism to V. parahaemolyticus. The reason for this result might be that the antimicrobial active ingredients of these eight strains were not extracellular secretions, so the cell-free supernatants had no antibacterial activity. In other relevant studies, the extracellular secretion of microorganisms was generally a source for novel antimicrobial substance extraction (Snook et al. 2009; Bacon et al. 2012). However, from the results of this study, the effective antibacterial ingredients of these eight isolates should be in their own cellular material. Those with the potential to become probiotics in aquaculture would probably be missed if the extracellular secretions of microorganisms was the research target. We do not know anything about the chemical properties of the antimicrobial compound(s) of any isolated strains in this study, so further research should focus on the chemical nature and the biosecurity of antimicrobial ingredients produced by such isolates to provide a theoretical basis for further fundamental research and application in aquaculture.

Identification of potential probiotics B16 and J7

Preliminary morphological and physiological identifications

The colony morphology of strain B16 was a light yellow, flat, glossy surface; while strain J7 was white, opaque, with a rough surface, and irregular edges around the colony on nutrient agar. Two potential probiotics were both Gram-positive, Bacillus spp. The spores of strains B16 and J7 were located in the middle of the cell, and underwent no swelling. Differences between biochemical tests of the two strains included the ability to use xylose, lysine, amylum, and nitrate. Biochemical test results are listed in Table 3. Based on the morphological, cultural, and physiological characteristics, strains B16 and J7 should be members of the Bacillus subtilis group.

Table 3 The results of biochemical tests of two potential probiotic strains (B16 and J7) isolated in this study
Full size table

Molecular biological identifications based on 16S rRNA gene sequences

The length of the 16S rRNA sequences of the strains of B16 and J7 was around 1450 bp as detected by 1 % agarose gel electrophoresis (data not shown). The sequences of two potential probiotics and 16S rRNA sequences obtained from other species within the same genus, and downloaded from GenBank at NCBI, were aligned to construct the phylogenetic tree. The results indicated that two the potential probiotic strains of B16 and J7 were closely related to Bacillus spp. which is well known for having the widely antimicrobial activity (Aunpad and Na-Bangchang 2007; Aunpad and Panbangred 2012; Li et al. 2012; Xu et al. 2014).

The identities of 16S rRNA between strains of B16 and Bacillus pumilus ATCC 7061 and strains of J7 and Bacillus mojavensis SWFU16 were 100 and 99 % respectively (Fig. 3). Two strains with a sequence identity of greater than 99 % should be classified as the same species according to Wu and Ahn (2011). It is difficult for B. pumilus and B. mojavensis, having a B. subtilis-like phenotype, to be distinguished from B. subtilis. Therefore, the 16S rRNA sequences analysis was used to determine the phylogenetic status for these potential probiotic strains of B16 and J7, which is the basis for identifying this kind of species (Roberts et al. 1994) and classifying bacteria (Woese 1987). Based on the results of physiological tests and 16S rRNA sequence analysis, strains B16 and J7 were identified as B. pumilus and B. mojavensis, respectively. As far as we know, this is the first report of B. mojavensis being isolated from Liaodong Bay.

Fig. 3
figure 3

Neighbour-joining tree exhibiting the phylogenetic positions of isolate B16, J7, and representatives of other related strains within Bacillus genus based on 16S rDNA gene sequences

Full size image

Enzymatic activities of B. pumilus B16 and B. mojavensis J7

Both B. pumilus B16 and B. mojavensis J7 exhibited proteinase activity and lipase activity. As for amylase activity B. mojavensis J7 was positive while B. pumilus B16 was negative. B. pumilus B16 and B. mojavensis J7 had the production capacity of extracellular hydrolytic enzymes, which made them potential probiotic candidates for the enhancement of digestion and nutrient uptake by the host (Nimrat et al. 2011). Members of Bacillus, compared with other species of probiotics, have achieved considerable success by the use of endospore-forming agents (Hong et al. 2005), the screened B. pumilus B16 and B. mojavensis J7 in this study provided new probiotic candidates.

Antagonistic spectrum of B. pumilus B16 and B. mojavensis J7

The culture and supernatant of B. pumilus B16 and B. mojavensis J7 were further investigated against the other ten pathogens (three Gram-positive bacteria and seven Gram-negative bacteria) by Oxford cup method. Five among the 10 indicator bacteria were common pathogens in aquaculture, others were human pathogens. The supernatants of B. pumilus B16 and B. mojavensis J7 showed extracellular antimicrobial activity to more than one indicator bacteria in this research. Three indicator bacteria were inhibited by the supernatant of B. pumilus B16 including S. typhimurium, L. monocytogenes and S. aureus. Moreover, four indicator bacteria were inhibited by the supernatant of B. mojavensis J7 including V. vulnificus, V. harveyi, E. coli, and S. aureus (Table 4). Staphylococcus aureus was the most sensitive indicator, while the indicator bacteria of V. parahaemolyticus ATCC 33847, P. aeruginosa, A. hydrophila, and B. subtilis were not inhibited by the supernatants of B. pumilus B16 and B. mojavensis J7 in this study.

Table 4 The antagonistic spectra of potential probiotics (B. pumilus B16 and B. mojavensis J7) isolated in this study with Oxford cup method (All values are in mm.)
Full size table

There is no particular trait as to antagonism of B. pumilus B16 or B. mojavensis J7 to Gram-positive, or Gram-negative, indicator bacteria. Combined with the result of this extracellular antimicrobial activity assay, the culture and/or the isolated cell itself may contain some kind of antimicrobial compounds which did not exist in the supernatant. It has been reported that even dead, or inactivated, probiotic cells had the ability to stimulate the innate immune system of hydrocoles (Irianto and Austin 2003). So the culture and/or isolated cells with direct antimicrobial activity should be used as the source for novel antimicrobial substance extraction (Lai and Wong 2013). Other research confirmed that members of genus Bacillus can produce more than one type of antimicrobial metabolite (Aunpad and Panbangred 2012; Bacon et al. 2012; Li et al. 2012; Xu et al. 2014). In addition, the strain of B. pumilus B16 or B. mojavensis J7 may produce different types, or amounts, of antimicrobial agents according to the variation in the potency of antagonism to different indicator bacteria (Table 4). This was in accordance with the result of Vynne et al. (2011) that the strain could change secondary metabolite production to adjust inhibitory activity within the species level itself. To our knowledge, this is the first time that reported that B. mojavensis can inhibit pathogenic Vibrio spp., including V. parahaemolyticus (ATCC 17802), V. vulnificus (environmentally isolated) and V. harveyi (environmentally isolated).

In recent years, some studies of the antagonistic activity of the isolated bacteria have been undertaken, and a variety of organisms antagonistic to pathogens in aquaculture have been found, including B. pumilus B16 and B. mojavensis J7 in this study. Due to Bacillus having the ability to produce heat-resistant spores, they have good prospects for application as probiotics.

Conclusion

In this study, the isolation, screening, and identification of Bacillus spp. antagonistic to V. parahaemolyticus were reported. The Bacillus spp. respectively isolated from the coastal sediment and the gills of O. oratoria showed biological control potential in aquaculture. The culture and/or the isolated cell, with its steady, direct, antimicrobial activity, could be used as a source for the exploration of novel bio-control agents. Further research will continue on the purification, identification, and biosecurity of antibacterial metabolites produced by those potential probiotics B. pumilus B16 and B. mojavensis J7 obtained here.