Multi-Omics Characterization of Host-Derived Bacillus spp. Probiotics for Improved Growth Performance in Poultry

Abstract

Microbial feed ingredients or probiotics have been used widely in the poultry industry to improve production efficiency. Spore-forming Bacillus spp. offer advantages over traditional probiotic strains as Bacillus spores are resilient to high temperature, acidic pH, and desiccation. This results in increased strain viability during manufacturing and feed-pelleting processes, extended product shelf-life, and increased stability within the animal's gastrointestinal tract. Despite numerous reports on the use of Bacillus spores as feed additives, detailed characterizations of Bacillus probiotic strains are typically not published. Insufficient characterizations can lead to misidentification of probiotic strains in product labels, and the potential application of strains carrying virulence factors, toxins, antibiotic resistance, or toxic metabolites. Hence, it is critical to characterize in detail the genomic and phenotypic properties of these strains to screen out undesirable properties and to tie individual traits to clinical outcomes and possible mechanisms. Here, we report a screening workflow and comprehensive multi-omics characterization of Bacillus spp. for use in broiler chickens. Host-derived Bacillus strains were isolated and screened for desirable probiotic properties. The phenotypic, genomic and metabolomic analyses of three probiotic candidates, two Bacillus amyloliquefaciens (Ba ATCC PTA126784 and ATCC PTA126785), and a Bacillus subtilis (Bs ATCC PTA126786), showed that all three strains had promising probiotic traits and safety profiles. Inclusion of Ba ATCC PTA12684 (Ba-PTA84) in the feed of broiler chickens resulted in improved growth performance, as shown by a significantly improved feed conversion ratio (3.3%), increased of European Broiler Index (6.2%), and increased average daily gain (ADG) (3.5%). Comparison of the cecal microbiomes from Ba PTA84-treated and control animals suggested minimal differences in microbiome structure, indicating that the observed growth promotion presumably was not mediated by modulation of cecal microbiome.

Keywords: Bacillus; broiler chickens; growth performance; multi-omics; poultry; probiotics.

FIGURE 1

Screening of Bacillus spp. DFM candidates for pathogen growth inhibition and enzyme activities. (A) Microorganism growth inhibition. Representative examples of microorganism overlay assays were presented with the following microorganisms, Avian Pathogenic Escherichia coli serotype O78, O2, and O18, Salmonella enterica serovar Typhimurium ATCC 14028, and Clostridium perfringens NAH 1314-JP1011. Overlay microorganism inhibition assays were performed in duplicate for each of the Bacillus spp. strains. A zone of inhibition was measured from the edge of Bacillus spp. until the end of clearance zone as shown as white double-edge arrows. (B) Screening of digestive enzyme activities of Bacillus spp. strains. The assay was performed in duplicate for each Bacillus spp.

FIGURE 2

A safety assessment of Bacillus spp. DFM candidates. (A) Antimicrobial susceptibility tests of Bacillus spp. MIC (μg/mL) values for each antibiotic tested of respective Bacillus spp. were shown. Nine medically important antibiotics at a concentration range of 0.06–32 μg/mL were tested and the respective antimicrobial susceptibility cut-off concentrations required for Bacillus spp. were shown at the top panel. (B) Cytotoxicity assessment of Bacillus spp. culture supernatant toward Vero cells. Culture supernatant of B. cereus ATCC 14579 and B. licheniformis ATCC14580 were used as positive and negative controls, respectively. Cytotoxicity level above 20%, dashed line, is considered cytotoxic. *, NR, not required by EFSA guidelines.

FIGURE 3

Global untargeted metabolomic analysis of Ba PTA84, Ba PTA85, and Bs PTA86. (A) Principal component analysis of scaled metabolite intensities for culture supernatants in two different media. Black symbols represent media controls; numbers in parentheses indicate the variance explained by the first two principal components. (B) Like (A), but for the culture cell pellets. (C) Number of identified secreted metabolites in culture supernatants in two different media (M) and (R), minimal and rich media, respectively. Secreted metabolites were defined as having a scaled intensity at least 1.5-fold higher than observed in media controls. Unique metabolites represent secreted compounds with abundances at least 1.5-fold higher than observed for other single strains (in the case of individual strain cultures) or corresponding individual strains (in the case of co-cultures). (D) Pathway representation of metabolites secreted by strains or strain combinations under minimal and rich media culture conditions.

FIGURE 4

Phylogenetic analysis of poultry origin Bacillus spp. (A) Core genome analysis of Ba PTA84 and PTA85, and Bs PTA86. (B) Phylogenetic relationship of Ba PTA84 and PTA85, and Bs PTA86 to other Bacillus species along with Lactobacillus reuteri as an outgroup.

FIGURE 5

In silico analysis of genes encoding enzymes and antimicrobial peptides in Bacillus spp. genomes. Genes encoding enzyme (A) and antimicrobial peptides (B) were compared among the genomes of three Bacillus spp., Ba PTA84 and PTA85, and Bs PTA86.

FIGURE 6

Effect of in-feed administration of Ba PTA-84 on growth performance of broiler chicken. Growth performance as measured by four parameters, (A) weight gain, (B) feed intake, (C) Feed Conversion Ratio (FCR), and (D) European Broiler Index (EBI).

FIGURE 7

Microbiome analysis of cecal content from birds fed with or without Ba PTA84. (A) ASV richness in the cecum of control birds and birds treated with Ba PTA84. Richness is quantified using the Chao index (Mann–Whitney U p-value = 0.07). (B) ASV diversity quantified with the Simpson (left) and Shannon indexes (right) for control birds and birds treated with Ba PTA84. (p-value = 0.44, 0.36). (C) Principal component analysis of the Bray–Curtis dissimilarity matrix between microbiome profiles for control birds and birds fed with Ba PTA84. Each dot represents a cecal sample. Numbers in parentheses indicate the variance explained by each principal component.

FIGURE 8

A screening workflow diagram for selection of Bacillus spp. probiotic candidate. Bacillus spp. isolates were screened for their activities to inhibit poultry pathogens and ability to secrete digestive enzymes in vitro. The best candidates were further selected based on their safety profiles (i.e., antimicrobial resistance profile and cytotoxicity level). Genomic and metabolomic analyses were performed on the select isolates to further investigate potential host-benefit properties and possible health/safety concerns. The top candidate was then tested for its effects on growth promotion in vivo.