I. INTRODUCTION
Continuous interest in new probiotics to alleviate intestinal health problems is emerging, especially since antibiotics become more restricted in use. Various Bacillus spp. are known to improve broiler performance and health due to their ability to produce antimicrobial substances, modulate the immune system, change the intestinal microbiota or increase nutrient digestion and utilization (Lee et al., 2010; Caulier et al., 2019; Giurescu et al., 2020; Jha et al., 2020; Zaghari et al., 2020). Several years ago, the probiotic organism Bacillus sp. PB6 was isolated from the intestinal tract of healthy chickens, and it was shown that this strain has an effectivity against Clostridium perfringens (Teo and Tan, 2005; Abudabos et al, 2013). During the past years, an extensive in vitro and in vivo screening process took place to find new strains to add together with this Bacillus sp. PB6, in order to extend its target range from only Clostridium to Enterobacteriaceae like E. coli and Salmonella. The current paper will describe a selection of these trials.
II. METHOD
In vitro trials. In the in vitro screening, 23 pathogenic E. coli strains from several locations (US and Asia) and 10 Salmonella strains (US and Asia, serovars Enteritidis, Typhimurium, Worthington, Pullorum, Gallinarum, Choleraesuis and London) were included. The inhibitory effect of the probiotic strains was measured using growth inhibition kinetics, where the pathogen culture was incubated with the probiotic cell free supernatant for 20 hours at 37°C and the optical density was measured at 600 nm every 20 minutes. Next to the inhibitory effect, also other qualitative parameters of the potential probiotic organisms were tested, such as pH (pH 3 and 6) and heat tolerance (pelleting at 80°c and 90°C), cytotoxicity to Vero cells, absence of antimicrobial resistance (AMR) genes (checked by full genome sequencing) and production yield (small scale fermentation).
In vivo Trial 1. E. coli challenge trial. In total, 100 one-day-old commercial broiler chickens (Ross-308) were divided over five isolators under negative pressure (20 birds/isolator). Each group of birds in an isolator received a different diet: (1) multi-strain probiotic dose 1, 3x108 CFU/kg feed, (2) multi-strain probiotic dose 2, 3x109 CFU/kg feed, (3) single-strain probiotic Bacillus sp. PB6, 3x108 CFU/kg feed, (4) not supplemented, challenged control, (5) not supplemented, not challenged control. At 14, 16 and 18 days of age, birds in group 1-4 were infected with 4.5-5.7x108 CFU avian pathogenic E. coli. At 21, 28 and 35 days of life, 5 birds from each group were killed for necropsy and bacteriological investigation. E. coli was counted using MacConkey agar plates. Feed consumed by birds in each group and body weight of all the birds were measured at weekly intervals and the body weight gain and FCR were calculated.
In vivo trial 2. Salmonella challenge trial. 240 one-day-old male Ross 708 broilers were divided over 2 groups, housed in cages (12 birds/cage; 10 cages/treatment): (1) challenged control, (2) multi-strain probiotic 3x108 CFU/kg feed. At day 7, the birds were challenged with 3.7 x 106 nalidixic acid-resistant Salmonella Heidelberg. At the end of the trial, at 42 days of age, 10 birds per treatment were taken from each individual cage (in total 100 samples per treatment), euthanized and the ceca were aseptically removed and the presence of Salmonella was checked.
In vivo Trial 3. Clostridium perfringens challenge trial. 336 one-day-old male Cobb 500 broilers were divided over 3 groups, housed in cages (8 birds/cage; 14 cages/treatment) : (1) not supplemented, not challenged control, (2) not supplemented, challenged control, (3) multistrain probiotic 3x108 CFU/kg feed. At day 14, the birds were challenged with coccidia (5000 oocysts of Eimeria maxima) and at 19, 20 and 21 days of age with 108 CFU Clostridium perfringens. At the end of the trial, at 28 days, the lesion scoring due to necrotic enteritis was assessed, as well as general performance and mortality.
III. RESULTS
In vitro trials. The new strains were able to reduce the growth of both E. coli and Salmonella in the respective kinetic studies. One of the obtained results can be found in Figure 1. Furthermore, the strains were shown to be tolerant to intestinal pH conditions and to high temperature conditions, they were non-cytotoxic, showed no presence of AMR genes and had a good production yield, showing their potential as probiotics.
Figure 1 - Effect of probiotic supernatant (mixture of the two new Bacillus strains FXA and G3) on E.coli (left, avian pathogenic field isolates) and Salmonella (right, SE: Salmonella enteritidis, ST: Salmonella thyphimurium, TSN: unknown field isolate) growth in an inhibition kinetic experiment.
In vivo trials. In the E. coli challenge trial, a significant decrease in cecal E. coli counts, compared to the infected control could be observed in the groups treated with both dosages of the multi-strain probiotic at 21 and 35 days of age, and for the highest dosage at day 28 (Figure 2). In contrast, this significant reduction in cecal E. coli counts was not observed with the single strain probiotic product. In addition, the highest body weight gain and lowest FCR during the last week of the trial was noted for group 2 (multi-strain probiotic dose 2).
Figure 2 - Effect of different treatments on the counts of E. coli in the cecum of broiler chickens (CFU/g) (left) and on broiler performance (right). Error bars show the standard error of the mean. Bars with different superscripts indicate significant differences (P < 0.05).
In the Salmonella challenge trial, the probiotic treatment numerically reduced the number of Salmonella positive caeca at the end of the trial whereas in the Clostridium challenge trial, the probiotic treatment resulted in significantly lower lesion scores and mortality (Table 1).
Table 1 - Selection of results from the Salmonella and Clostridium perfringens challenge trials. Numbers with different superscripts within one column indicate significant differences (P < 0.05).
IV. DISCUSSION
The results showed that the new strains were able to reduce the growth of both E. coli and Salmonella in the kinetic studies. This observed effect can be explained by antimicrobial substances produced by the probiotic strains, such as lipopeptides and polyketides (Caulier et al., 2019). In addition, the strains were tolerant to intestinal pH conditions and to high temperature conditions, were non-cytotoxic, showed no presence of AMR genes and had a good production yield, all important characteristics of in-feed probiotics. The newly developed multi-strain probiotic was also able to reduce mortality and performance losses due to avian pathogenic E. coli, reduce Salmonella load in the ceca of broiler chickens and prevent mortality and intestinal lesions due to necrotic enteritis. The effect against Clostridium perfringens has been shown before for the Bacillus sp. PB6 (Teo and Tan, 2005; Abudabos et al., 2013), and this effectivity was maintained with this multi-strain probiotic. The ability of the Bacillus strains to affect Enterobacteriaceae, such as E. coli and Salmonella is in accordance with previous literature (Amerah et al., 2013; Thirabunyanon and Thongwittaya, 2011; Shanmugasundaram et al., 2020;) and can be explained by antimicrobial substances formed, as shown in the in vitro study, and in addition, the probiotic strains might also favor a more beneficial development of the intestinal microbiota, modulate the immune system and improve nutrient digestibility, as explained in the introduction.
In conclusion, this research describes the development process of a new multi-strain probiotic targeting multiple potential pathogens at once.
Presented at the 34th Annual Australian Poultry Science Symposium 2023. For information on the next edition, click here. 
