Antimicrobial Properties of Bacillus Probiotics as Animal Growth Promoters

Antibiotic growth promoters (AGPs) suppress the growth of infectious pathogens. These pathogens negatively impact agricultural production worldwide and often cause health problems if left untreated. Here, we evaluate six Bacillus strains (BPR-11, BPR-12, BPR-13, BPR-14, BPR-16 and BPR-17), which are known for their ability to survive harsh environmental conditions, as AGP replacements in animal feed. Four of these Bacillus strains (BPR-11, BPR-14, BPR-16 and BPR-17) showed antimicrobial activity against the pathogenic strains Clostridium perfringens, Escherichia coli and Staphylococcus aureus at 25 μg/mL, with BPR-16 and BPR-17 also able to inhibit Pseudomonas aeruginosa and Salmonella enterica at 100 μg/mL. Further chemical investigation of BPR-17 led to the identification of eight metabolites, namely C16, C15, C14 and C13 surfactin C (1–4), maculosin (5), maculosine 2 (6), genistein (7) and daidzein (8). Purified compounds (1–4) were able to inhibit all the tested pathogens with MIC values ranging from 6.25 to 50 μg/mL. Maculosin (5) and maculosine 2 (6) inhibited C. perfringens, E. coli and S. aureus with an MIC of 25 μg/mL while genistein (7) and daidzein (8) showed no activity. An animal trial involving feeding BPR-11, BPR-16 and BPR-17 to a laboratory poultry model led to an increase in animal growth, and a decrease in feed conversion ratio and mortality. The presence of surfactin C analogues (3–4) in the gut following feeding with probiotics was confirmed using an LC–MS analysis. The investigation of these Bacillus probiotics, their metabolites, their impacts on animal performance indicators and their presence in the gastrointestinal system illustrates that these probiotics are effective alternatives to AGPs.


Introduction
Antimicrobial growth promoters (AGPs) are currently utilised in the agricultural industry to improve livestock production, feed-energy conversion and to prevent the spread of infectious diseases [1][2][3]. These AGPs eliminate the microbiota in the gut and allow the host to access more nutrients [4,5]. They also halt the production of toxins produced by pathogens and improve livestock production [4,5]. However, there has been a steady rise of antimicrobial resistance due to the misuse of these AGPs. This has driven restrictions and bans of AGPs in countries throughout the EU, U.S. and Indonesia [6]. These regulations have been linked to decreases in livestock production and higher rates of food-borne infections, highlighting a need for safer and more effective alternatives to AGPs, such as probiotics [6][7][8][9].
Probiotics are live microorganisms that provide a range of benefits to their hosts once consumed [10,11]. Their benefits include the production of enzymes that assist in breaking down indigestible material, providing competition for nutrients in the gut to inhibit pathogenic bacterial growth and the production of antimicrobial metabolites [12][13][14][15][16]. These mechanisms are linked to an observed increase in animal growth and a reduction in

The Antimicrobial Activity of Selected Bacillus Strains
Six Bacillus strains (BPR-11, BPR-12, BPR-13, BPR-14, BPR-16 and BPR-17) were selected (Table 1) for the current study. Each strain was cultured in a tryptone soy broth (TSB) medium. The bacterial cultures were then lysed, centrifuged and the cell pellet removed. The supernatant broth was split into two halves: the first half was extracted with ethyl acetate (EtOAc) to give the EtOAc extract while the second half was freeze dried and resuspended in water to provide an aqueous crude extract. The antimicrobial activity of the EtOAc extracts was evaluated against C. perfringens, E. coli, P. aeruginosa, S. aureus and S. enterica at the concentrations of 200, 100, 50 and 25 µg/mL (Supplementary Materials Table S9). The results showed that the EtOAc extracts of strains BPR-11, BPR-14, BPR-16 and BPR-17 inhibited C. perfringens, E. coli and S. aureus growth at 25 µg/mL, whilst the EtOAc extracts of BPR-16 and BPR-17 were also active against P. aeruginosa and S. enterica at 100 µg/mL. The EtOAc extracts of BPR-12 and BPR-14 showed no activity. The crude extracts of the Bacillus strains were tested for antimicrobial activity at 800, 400, 200 and 100 µg/mL. However, no activity was detected for any of the crude extracts at the concentrations tested.

Isolation and Structure Elucidation of Antimicrobial Compounds
The chemical compositions of the active EtOAc extracts from BPR-11, BPR-14, BPR-16 and BPR-17 were analysed by HPLC ( Figure 1B-E) and 1 H NMR ( Figure S42). They were also compared with the EtOAc extract of the culture media TSB ( Figure 1A, Figure S42). The chromatograms showed similar profiles among BPR-11, BPR-14, BPR-16 and BPR-17, comprising an initial peak inherent from TSB at 0-2 min and a major peak at 4.5 min indicative of secondary metabolites. BPR-17 ( Figure 1E) also contained several minor compounds at the retention time between 4-6 min. As BPR-17 was the most active and most abundant in secondary metabolites, it was selected for further chemical investigation to identify the antimicrobial secondary metabolites.
The antimicrobial activity of the EtOAc extracts was evaluated against C. perfringens E. coli, P. aeruginosa, S. aureus and S. enterica at the concentrations of 200, 100, 50 and 25 μg/mL (Supplementary Materials Table S9). The results showed that the EtOAc extracts of strains BPR-11, BPR-14, BPR-16 and BPR-17 inhibited C. perfringens, E. coli and S. aureus growth at 25 μg/mL, whilst the EtOAc extracts of BPR-16 and BPR-17 were also active against P. aeruginosa and S. enterica at 100 μg/mL. The EtOAc extracts of BPR-12 and BPR-14 showed no activity. The crude extracts of the Bacillus strains were tested for antimicrobial activity at 800, 400, 200 and 100 μg/mL. However, no activity was detected for any of the crude extracts at the concentrations tested.

Animal Trial
The animal trial was conducted on 256 broiler hatchlings of mixed sex for 28 days. The hatchlings were individually weighed and birds that had a similar weight (±2 g) were randomly assigned to either a control or probiotic group. The control group was fed a basal, medication free wheat-soybean diet, whilst the probiotic group's feed was supplemented with a probiotic mixture F1, consisting of BPR-11, BPR-16 and BPR-17 (Table 3). The broiler chickens were reared in floor pens and had ad libitum access to feed and water for 28 days [36]. Table 3. Bacillus strains and specifications in F1 probiotics.
The body weight (BW), average daily gain (ADG), average daily feed intake (ADFI) and feed conversion ratio (FCR) were recorded on a weekly basis. The data collection after 28 days is shown in Table 4.  The Bacillus probiotic-treated groups showed an overall increase in the measured indicators. Specifically, an average increase in mean BW (1573 g to 1606 g; p = 0.12), ADG (56.19 g to 57.36 g; p = 0.12), ADFI (71.95 g to 72.47 g; p = 0.63) and a decrease in FCR (from 1.28 to 1.26; p = 0.01) ( Table 4) were observed for 28 days. Although the treated group was not significantly different from the control for the first 14 days, a significant increase was evident between day 14-21 in BW (885.46 g to 910.57 g; p = 0.04), ADG (66.99 to 70.05; p = 0.005) and ADFI (83.49 g to 87.58 g; p = 0.00061). A significant decrease in FCR (0-28 days) was detected when fed with probiotics (p = 0.01), indicating a significantly improved conversion of feed to energy levels. Lastly, the probiotic treatment group showed a lower mortality rate (1.56%) compared with the control group (3.13%).

Caecum Content Analysis
To evaluate whether the antimicrobial metabolites were present in the animal gut, caecum samples were collected from control and probiotic-supplemented birds for targeted metabolomic analysis. At the end of the animal trial, one bird per pen was euthanised by CO 2 chamber and eviscerated to collect the caecum. Only male birds with similar body weights were selected. The samples collected from the trial were stored at -80 • C before being processed. Each sample was extracted with EtOAc/H 2 O (1:1), then vortexed and centrifuged. The EtOAc layer was dried and redissolved in MeOH and the MeOH solution was analysed by LC-MS. Ions, representing compounds 1-8, were extracted on both positive and negative ionisation modes using the molecular weights of the compounds. Fifteen gut samples were randomly selected from control and probiotic treated group.
For the probiotic treated group, surfactin C analogues (3)(4) were present in 12 out of 15 samples ( Figure 4A,B). Compounds 1, 2, 5 and 6 were also detectable ( Figure S41); however, the signals were weak compared to the noise. For the control group, none of the compounds were detected in any of the 15 samples. solution was analysed by LC-MS. Ions, representing compounds 1-8, were extracted on both positive and negative ionisation modes using the molecular weights of the compounds. Fifteen gut samples were randomly selected from control and probiotic treated group.
For the probiotic treated group, surfactin C analogues (3)(4) were present in 12 out of 15 samples (Figure 4A,B). Compounds 1, 2, 5 and 6 were also detectable ( Figure S41); however, the signals were weak compared to the noise. For the control group, none of the compounds were detected in any of the 15 samples.

Discussion
Bacillus probiotics are renowned for their ability to produce a range of antimicrobial compounds, hence they act as alternative AGPs [37]. In this study, we selected six Bacillus strains, investigated their antimicrobial potential, secondary metabolites and their effects on animal growth [38]. Our results showed that four strains, BPR-11, BPR-14, BPR-16 and BPR-17, inhibited the activity of C. perfringens, E. coli and S. aureus, with BPR-16 and BPR-17 also inhibiting P. aeruginosa and S. enterica. Further chemical analysis on BPR-17, one of the most active strains, led to the isolation of antimicrobial lipopeptides (1-4) and dipeptides (5)(6). Subsequent animal trials involving the feeding of a probiotic mixture including BPR-11, BPR-16 and BPR-17 over 0-28 days showed an increase in BW, ADG and ADFI and a decrease in FCR and mortality. By analysing the caecum samples from the animal trials, we were able to detect signals indicative of lipopeptides (3)(4). The results suggest that antimicrobial surfactins may be responsible for the beneficial effects shown in the animal trials, supporting Bacillus probiotics as an effective alternative to AGPs.
The selection of strains for the animal trials was based on multiple factors, including the antimicrobial potential and the compatibility in fermentation. BPR-17 and BPR-16 were the most active followed by BPR-11 and BPR-14. Co-incubation of the mixed strains with pathogens suggested that the combination of BPR-17, BPR-16 and BPR-11 gave the best activity against all tested pathogens (unpublished results). In addition, the benefit of having multiple strains is well known in the feed industry, as there is an increasing tendency to work with multi-strain probiotics [39].
The most potent antimicrobial metabolites revealed in this study were the lipopeptides C16-C13 surfactin C (1-4). These metabolites were first reported in 1968 as products of Bacillus subtilis and are renowned for their potency to target both Gram-positive and negative strains [29]. These lipopeptides are characterised by a peptide moiety consisting of the seven amino acids (L-Glu, L-Leu, D-Leu, L-Val, L-Asp, D-Leu and L-Leu) with a βhydroxy fatty acid tail and act in a concentration-dependent manner [40]. Previous studies showed that these lipopeptides form unilamellar vesicles on the membrane wall, leading to cell distortion and triggering apoptosis [41][42][43][44]. These compounds can also form pores at the plasma membrane that lead to intracellular leakage and eventual cell death [42,[45][46][47]. Additionally, surfactins can breakdown bacterial biofilms, either by reducing the ratio of alkali-soluble polysaccharides or down-regulating the expression of genes involved in biofilm formation [48]. In our study, we found that the length of the lipophilic fatty acid affects its antimicrobial activity. This finding is consistent with the literature, showing that an extended fatty acid improves its ability to permeabilise the bacterial membrane leading to cell death [49].
Additionally, maculosin (5) and its analogue maculosine 2 (6) were isolated from the culture of BPR-17. Maculosin, a dipeptide, has been previously reported to exert an antimicrobial effect against methicillin-resistant S. aureus (MRSA) (37.0 µg/mL), E. coli  [30,50]. The mechanisms of maculosine 2 have been purported to be due to its interaction with lipid heavy cell walls eventually leading to cell death. Furthermore, maculosin has been shown to antagonise quorum sensing, which is needed for bacterial interaction and growth [51,52]. In comparison, the activity of compound 6 has been reported to be lower compared to 5, highlighting the importance of the phenol group for its antimicrobial activity [30,53]. Genistein (7) and daidzein (8) have been identified for their antioxidant properties and have been explored for use in animal feed to improve animal health [32,[54][55][56]. In our study, we were able to isolate these two isoflavones, but there was no antimicrobial activity detected for concentrations up to 100 µg/mL [57]. Although genistein has been previously reported to inhibit S. aureus and MRSA after only a limited incubation of 5-10 h, the authors noted that this inhibition was only a short term inhibition of topoisomerase IV and, after 10 h, the genistein was degraded by the bacteria in the culture, leading to a lack of inhibition over extended periods [58]. Additionally, daidzein has been shown to inhibit topoisomerase I, and, more weakly, topoisomerase II [59]. These actions halt the nucleic acid system, showing that these compounds are bacteriostatic rather than bacteriocidic as observed with our antimicrobial testing [60].
This study also investigated the effects of Bacillus probiotics on animal performance indicators and their faeces. Increases in BW, ADG and ADFI and a decrease in FCR were observed and have been associated with the altered fermentation and microbial ecology in the gut [61]. Similar benefits to animal performance has also been observed by Cavazzoni et al. (1998) [62]. Here, the addition of B. coagulans probiotics to the diets of poultry promoted growth and had prophylactic effects compared to chickens fed without additives. Additionally, the benefits to animal performance were comparable to animals fed with the antibiotic virginiamycin, commonly used as a growth promoting agent. These results can be explained by assuming that the Bacillus supplementation enhanced the feed utilization and reduced the excretion of proteins in the faeces [63]. Previous research has also highlighted the reduction of NH 3 -N of animals fed with Bacillus probiotics, suggesting a microbial aspect that reduces the production of urease [64]. This has been reinforced with recent studies, which have investigated the ability of Bacillus probiotics to affect the microbial community and to improve animal performance [65,66]. Here, not only could Bacillus probiotics competitively exclude pathogenic bacteria, but they were able to hinder the growth of the pathogenic bacteria E. coli and S. enterica, leading to an improvement in overall animal health [67][68][69].
Research into the use of probiotic interventions and their exact influence on the microbial communities using nucleic acid sequencing has increased over the past decade [70]. However, these changes in the microbial communities are often indirect and have resulted in closer analyses of the metabolome itself. Targeted metabolomic analyses of the faecal contents in the caecum confirmed the presence of secondary metabolites (3)(4) in the gut. This follows previous findings, showing that probiotic supplementation changes the metabolite profiles in the gut using GC-MS and 1 H NMR [71]. Lactobacillus probiotic supplementation was able to increase the amount of health enhancing compounds such as amino acids, carbohydrates and lipids [67,72]. This includes glutamine and 5-aminoimidazole-4-carboxamid, produced by L. acidophilus, which increases enterocyte proliferation, regulation of tight junctions and suppression of inflammatory signalling pathways [72,73]. However, the potential for these antibacterial components was unexplored in this study, and they may be synergistic in their antimicrobial effects [74]. Additionally, the gut microflora have been shown to interact heavily with the blood metabolites, and further metabolomic analysis of either the serum, saliva or urine will help deconvolute this network [75]. These directions will help identify potentiators, and reveal how these metabolites interact with each other, and hence assist in optimizing the performance and benefits of probiotic supplementation [76].
NMR spectra were obtained using a Bruker Ascent 800 MHz spectrometer coupled with a TCI CryoProbe at 298 K in deuterated dimethyl sulfoxide (DMSO-d 6 ) (Cambridge Isotope Laboratories, Tewksbury, MA, USA). The optical rotation was recorded using a Jasco P-1020 polarimeter using a quartz cell with 10-cm path length. The LC-MS analysis was conducted on a Thermo Scientific MSQ Plus single quadrupole ESI mass spectrometer equipped with an Accucore C 18 LC column (2.6 µm, 150 mm × 2.1 mm) with H 2 O/MeOH/formic acid as the solvent system. A Themo Scientific Dionex Ultimate 3000 RS UHPLC system equipped with a pump, autosampler, diode array detector and automated fraction collector was used for chemical analysis and HPLC fractionation. A Jasco V-650 Spectrophotometer was used for optical density (OD) determination. A Biotek Spectre 2 microplate reader was used for the antimicrobial activity readout. The eggs were incubated in a Brinsea Ova-Easy 380 Egg Incubator in the animal trial.

Bacillus Strains and Fermentation
Six Bacillus strains (BPR-11, BPR-12, BPR-13, BPR-14, BPR-16 and BPR-17) were obtained from Bioproton Pty Ltd. (Brisbane, Australia). The Bacillus strains were streaked on a plate with TSA (1.5% agar) and incubated for 16 h. Subsequently, a single colony was isolated and incubated for 24 h in TSB (1 L) with vigorous shaking (300 rpm). A starting inoculation culture was prepared with 0.1 OD at 600 nm. The sample was inoculated into a culture broth of TSB (1 L) added to a Eppendorf bioreactor (1.5 L). The culture was incubated for 8 h at 37 • C, 300 rpm, pH 7, with O 2 levels at 300 ppm. Antifoam C was added to remove excess foam from the bioreactor. The fermented cultures were stored at −20 • C until ready for extraction.

Extraction
The culture broth (1 L) was sonicated three times for 5 min and centrifuged at 4 • C at 8000 rpm. The cell pellet was separated from the broth and stored at −20 • C. Half of the broth (500 mL) was extracted with EtOAc (500 mL) three times and the combined EtOAc extracts were brought to dryness using a rotary evaporator to obtain an EtoAc extract. The second half of the broth (500 mL) was freeze-dried and resuspended in H 2 O to give the crude extract.

Isolation
The chemical analysis of the EtOAc extracts (1 mg/mL) was performed on a Phenomenex Onyx Monolithic C 18 column (100 mm × 4.6 mm) with a gradient elution from 10% H 2 O/90% MeOH/0.1% TFA to 100% MeOH/0.1% TFA within 7 min, then back to 10% MeOH/90% H2O/0.1% TFA within 4 min, at a flow rate of 4 mL/min. For large-scale isolation, the EtOAc extract (500 mg) was loaded onto an Alltech Hyperprep PEP C 18 column (250 mm × 22 mm) eluting with a solvent gradient from 10% MeOH/90% H 2 O/0.1% TFA to 100% MeOH/0.1% TFA within 50 min then kept at 100% MeOH/0.1% TFA for 10 min, with a flow rate of 9.0 mL/min. Sixty fractions were collected based on time (1 fraction per min) and the tubes were combined into 10 fractions based on their UV peaks for further antimicrobial testing.

Antimicrobial Assay
Bacteria were streaked on plates containing MHA (28 g/L, 1% Agar). A colony was picked, added to MHB and incubated at 37 • C until a starting culture of 0.1 OD was obtained. To test for activity, bacterial broth (78 µL) was added to either the crude extracts (800, 400, 200 and 100 µg/mL, 2 µL) or the EtOAc extracts (100, 50, 25, 12.5 µg/mL, 2 µL). The plate was then incubated at 37 • C overnight. A resazurin dye (25 µL) was then added into each well and incubated at 37 • C for 6 h, and the antimicrobial activity was quantified by the optical density at 525/580 nm on a plate reader. The serial dilutions were prepared with DMSO. Gentamicin (Sigma-Aldrich) was used as a positive control and DMSO was used as a negative control. Each concentration was tested in triplicate (n = 3).
For MIC determination of the pure compound, a range of concentrations were prepared (200, 100, 50, 25, 12.5, 6.25, 3.13 or 1.563 µg/mL, 2µL) and the assay was conducted as described for the extracts. The MIC was determined by the inhibition of visual growth of bacterium.

The Animal Trial
This study used a subset of an animal trial approved by the Ethics Committee of Central Queensland University under approval number 0000023123. A total of 256 mixedsex broiler eggs were randomly selected and placed into two identical incubators with automatic digital humidity control and temperature sensitivity down to 0.1 • C and were digitally monitored 24 h per day for 21 days of incubation. The incubators were set at 37.5 • C and 55% relative humidity (RH) on day 1-18, and 37.5 • C and 65% RH on day 18-21. The hatchlings were individually weighed and birds with similar weight (±2 g) were randomly assigned to one of two experimental groups (control and probiotic). Each experimental group had 16 replicate pens with eight birds in each pen (n = 128 per experimental group). The feed-supplemented probiotic was added to the basal diet in the facility, and the diet was thoroughly mixed prior to the start of the experiment. The experimental treatments included a standard micro crumble wheat-soybean diet (control) and control +500 g/t probiotics (F1). The F1 probiotic added to the feed was supplied by Bioproton Pty Ltd. (Acacia Ridge, Queensland, Australia) ( Table 3).
The broiler chickens in each group were reared in floor pens (120 cm × 120 cm × 80 cm) with wood shavings as bedding and had ad libitum access to feed and water for 28 days. Throughout the trial period, the ROSS 308 Management Handbook was followed to meet the nutritional and environmental recommendations of the broiler, including temperature, relative humidity (RH) and lighting program [36]. The temperature was set at 32 • C and 40-50% RH for the first 7 days and a 2 • C reduction per week until the temperature reached 22 • C at 28 days.
The BWs of individual birds and the FIs of each replicate pen were recorded weekly. The ADG, ADFI and FCR ratio between feed intake over BW of birds per replicate were calculated. Animal mortality was recorded and the data was used to adjust subsequent measurements.

Targeted Metabolomic Analysis
The caecum samples from the animal trial were stored at −80 • C and were analysed following a protocol previously described by Martias [77]. The caecum samples were freeze-dried, and a portion of the samples (6 mg) was added to H 2 O (6 mL) and ethyl acetate (6 mL). Each suspension was vortexed for 5 min and centrifuged at 12,000× g for 10 min at 4 • C. Subsequently, the supernatant (400 µL) was removed, concentrated and redissolved in 100% methanol (30 µL). The samples were analysed by the LC-MS system with a gradient elution from 5% H 2 O/95% MeOH/0.1% formic acid to 100% MeOH/0.1% formic acid within 13 min, at a flow rate of 0.3 mL/min. Ions were extracted using both positive and negative modes.