Functional evaluation of Bacillus licheniformis PF9 for its potential in controlling enterotoxigenic Escherichia coli in weaned piglets

Abstract During the bacterial selection, isolate PF9 demonstrated tolerance to low pH and high bile salt and an ability to extend the lifespan of Caenorhabditis elegans infected with enterotoxigenic Escherichia coli (ETEC; P < 0.05). Thirty-two weaned piglets susceptible to ETEC F4 were randomly allocated to four treatments as follows: 1) non-challenged negative control group (NNC; basal diet and piglets gavaged with phosphate-buffered saline), 2) negative control group (NC; basal diet and piglets challenged with ETEC F4, 3 × 107 CFU per pig), 3) positive control (PC; basal diet + 80 mg·kg−1 of avilamycin and piglets challenged with ETEC F4), and 4) probiotic candidate (PF9; control basal diet + 2.5 × 109 CFU·kg−1 diet of B. licheniformis PF9 and piglets challenged with ETEC F4). The infection of ETEC F4 decreased average daily gain and gain:feed in the NC group when compared to the NNC group (P < 0.05). The inoculation of ETEC F4 induced severe diarrhea at 3 h postinoculum (hpi), 36, 40 hpi in the NC group when compared to the NNC group (P < 0.05). The supplementation of B. licheniformis PF9 significantly relieved diarrhea severity at 3 hpi when compared to the NC group (P < 0.05). The inoculation of ETEC F4 reduced duodenal, jejunal, and ileal villus height (VH) in the NC group when compared to the NNC group. A significant (P < 0.05) decrease was detected in the duodenal VH in the PC and NNC groups. Moreover, the NNC group had a reduced relative mRNA level of Na+-glucose cotransporter 1 (SGLT1) when compared to the NC group (P < 0.05). Compared to the NC and NNC groups, the supplementation of B. licheniformis PF9 increased the relative mRNA levels of aminopeptidase N, occludin, zonula occludens-1, and SGLT1 (P < 0.05). The supplementation of B. licheniformis PF9 also significantly increased the relative mRNA level of excitatory amino acid transporter 1 when compared to the NC group (P < 0.05). Piglets supplemented with B. licheniformis PF9 showed lower relative abundance of Bacteroidetes in the colon than piglets from the NNC group (P < 0.05). The NNC group had a higher relative abundance of Firmicutes in the ileum than all the challenged piglets (P < 0.05); however, a lower relative abundance of Proteobacteria in the ileum and colon was observed in the NC group (P < 0.05). This study provides evidence that B. licheniformis PF9 has the potential to improve the gut health of piglets under challenging conditions.


INTRODUCTION
Postweaning diarrhea (PWD) is a common disease that causes a significant economic loss in the global swine industry (Rhouma et al., 2017).Enterotoxigenic Escherichia coli (ETEC) is the main infectious agent of PWD (Luppi, 2017).Antibiotic growth promoters (AGPs) have been commonly added to feed to improve animal growth performance and prevent infectious diseases such as PWD (Thoms, 2012).However, the use of AGPs in livestock production has been suggested to contribute to the development of antibiotics resistance, which threatens human health (Martin et al., 2015).
The restriction of AGPs use in feed was enforced in European Union countries since 2006; in 2017, according to rules of the Food and Drug Administration of United States, animal producers required a veterinary feed directive for feed applications of antibiotics (Ekakoro et al., 2019); in Canada, prescriptions issued by veterinarians are needed for the use of all medically important antimicrobials (MIA) as defined by the World Health Organization, and growth promotion claims on the MIA product labels have been removed according to Health Canada regulations and policies implemented in 2018 (Lardé et al., 2021).To overcome the potentially negative effects of removing AGPs, several AGPs alternatives, including pro-and prebiotics, zinc, copper, acidifiers, and plant extracts, have been tested (Liu et al., 2018).Among them, dietary supplementation of probiotics has been reported to have positive effects on improving gut health and immunity, nutrient digestibility, and suppressing pathogens, hence enhancing growth performance and preventing enteric diseases (Liao and Nyachoti, 2017).Various dietary probiotics have been studied in the past few decades.Bacillus licheniformis has been reported to have the function of secreting digestive enzymes to enhance nutrient digestion and absorption and antimicrobial peptides to inhibit the growth of Brachyspira hyodysenteriae and Clostridium perfringens (Rozs et al., 2001;Horng et al., 2019).Recently, our group reported that B. licheniformis PF9 was capable of reducing inflammationrelated cytokines by blocking the NF-κB signaling pathways, while, significantly enhancing the intestinal porcine epithelial cell line integrity when infected with ETEC F4 (Li et al., 2022).It is hypothesized that dietary B. licheniformis PF9 will improve the gut health and growth performance of piglets.Therefore, the objective of the current report is to describe how B. licheniformis PF9 was selected and investigate its potential in application to improve the gut health and performance of weaned piglets including growth performance, digestive enzymes, gut morphology and integrity, nutrient transport, oxidative stress, innate immune responses, and gut microbiota after the ETEC F4 challenge.

Materials and Methods
The pig trial and animal care protocols (F19-020, AC11516) were approved by the Animal Care Committee of the University of Manitoba.The piglets used in this trial were cared for under the guidelines of the Canadian Council on Animal Care (CCAC, 2009).

Bacillus licheniformis PF9 and Its Culture Preparation
Bacillus licheniformis PF9 was recently shown through an in vitro study that it can improve barrier function and alleviate inflammatory responses against ETEC infection (Li et al., 2022).The bacterium was isolated from the feces of pigs raised on a free-range farm in Ontario, Canada, where antibiotics were not used in raising the pigs.It is heat resistant.The selection of a heat-resistant bacterium was conducted through a screening of multiple feces samples by suspending 0.5 g of pig feces in 4.5 mL of phosphate-buffered saline (PBS, pH 7.0) and heating the mixtures at 80 °C for 10 min followed by a series of dilution of each sample.The diluted samples were plated on the tryptic soy agar (TSA; Millipore Sigma, Oakville, ON, Canada) and incubated at 37 °C until colonies became visible.Finally, a single colony was selected for a pure bacterial culture.
To prepare a PF9 culture for the pig trial, a single colony of B. licheniformis PF9 from an agar plate was inoculated into 6 mL of tryptic soy broth (TSB; Millipore Sigma) medium in a 12-mL culture tube followed by the incubation at 37 °C for 16 h without shaking to get an initial culture.The initial culture was mixed with 250 mL of TSB in a 500-mL flask and incubated at 37 °C for 16 h without shaking.All the incubations were carried out aerobically.The culture (500 mL) with more than 99% of the cells in a vegetative form was evenly mixed with 50 kg of feed which included 2.5 × 10 9 CFU•kg −1 of B. licheniformis PF9 in the diet.

In Vitro Characterization of B. licheniformis PF9
To determine the species identity of isolate PF9, its 16S ribosomal RNA (rRNA) gene was sequenced and then blasted against the database sequences available on the National Center for Biotechnology Information website (http://www.ncbi.nlm.nih.gov).Further determination of isolate PF9 to be B. licheniformis or B. paralicheniformis was carried out using the method of particular molecular markers developed by Olajide et al. (2021).Both 16S rRNA gene sequencing and the marker detection to distinguish B. licheniformis from B. paralicheniformis were conducted by the Agriculture and Food Laboratory (Laboratory Service Division, University of Guelph, Guelph, Ontario, Canada), an ISO/IEC 17025 accredited laboratory (https://afl.uoguelph.ca/accreditations).
To determine the antimicrobial activity of B. licheniformis PF9, an inhibitory assay was carried out by measuring the culture optical density at 600 nm (OD 600 ) of a target pathogen at 30-min intervals during an 18-h incubation at 37 °C with a Bioscreen C MBR (Oy Growth Curves Ab Ltd., Helsinki, Vuorimiehenkatu, Finland).The assay mixture (300 µL each) was constituted of 150 µL of TSB containing 1% inoculum of ETEC JG280 or Salmonella Typhimurium DT104 and 150 µL of filter-sterilized extracellular culture fluid of PF9.The control in the assay was the replacement of the extracellular culture fluid of PF9 with 150 µL of TSB without the bacterial inoculum, which was conducted in parallel.
To determine the tolerance of B. licheniformis PF9 to low pH and high bile salt, the isolate was tested based on the method previously reported (Yang et al., 2014) using the Bioscreen C MBR (Oy Growth Curves Ab Ltd.) to monitor the bacterial growth.Tolerance to low pH was evaluated by incubating bacterial cells in Lysogeny broth (LB; Millipore Sigma) at 37 °C, which had been treated with simulated gastric fluid (Wang et al., 2009) for 2 h with the pH values of 2.0 and 5.6, respectively.Tolerance to bile salt was investigated by incubating the isolate in the LB containing various concentrations (0%, 0.1%, 0.5%, 1.0%, and 1.5%) of bile salt (Catalog No. LP005, Oxoid, Nepean, ON, Canada) at 37 °C.The OD value of a sample was determined at a wavelength of 600 nm.The selected range of bile salts was based on the physiological concentration in the small intestine (Weinman et al., 1998).

Culturing and Lifespan Assay of Caenorhabditis elegans
The SS104 strain of C. elegans harboring a temperaturesensitive allele of glp-4 (bn2) was obtained from the Caenorhabditis Genetics Center (University of Minnesota, Minneapolis, MN, USA).C. elegans was maintained on nematode growth medium (NGM) plates seeded with E. coli OP50 (1 × 10 8 CFU•mL −1 ) that was used as food for the nematode according to the procedures described by Breger et al. (2007).The ETEC strain JG280 is a hemolytic E. coli of serotype O149: K88 (F4), a porcine isolate possessing toxin genes and resistant to some antibiotics (Noamani et al., 2003).
The lifespan assays of C. elegans were conducted based on the previously published method (Zhou et al., 2014).Briefly, gravid adult worms were treated with sterile water (containing 0.5% NaClO solution and 0.5 M NaOH) to synchronize worms to the same age and stage.The eggs were released and isolated by centrifugation (1 min at 1,300 × g) and resuspended in M9 buffer, and then hatched for 16 h at 20 °C.The L1 larvae of C. elegans were transferred to NGM agar with E. coli OP50 (1 × 10 8 CFU•mL −1 ) as food for the worms at 25 °C for 48-60 h until the worms reached the L4 stage.The worms collected from the NGM plate with M9 buffer were washed three times in the S medium, 18 to 26 worms were then transferred to each well containing 2 mL of S medium on a 24-well plate and incubated at 25 °C.
The treatments of the assay included 1) Control (treated with E. coli OP50 only), 2) ETEC infection (incubated with E. coli OP50 for 18 h followed by incubation of ETEC JG280 for 12 days), and 3) B. licheniformis PF9 pretreatment (incubated with B. licheniformis PF9 for 18 h followed by incubation of ETEC JG280 for 12 days).Each assay was designated as day 0 when the worms were fed E. coli OP50 or B. licheniformis PF9.The worms were collected after the 18-h incubation period via centrifugation and suspension and washed three times in the S medium.The assays were designated as day 1 when the washed worms were mixed with ETEC JG280 (2 × 10 8 CFU mL −1 ) in a 24-well plate.The lifespan assay lasted for 12 days.The number of live worms was recorded daily, and the survival rate of worms was calculated by the formula: survival rate (%) = (live worms/total worms used) × 100.A worm was considered dead when it failed to respond to touching.

ETEC F4 Culture Preparation
The ETEC F4 from −80 °C stock was streaked on the media of TSA to aerobically grow for 16 h at 37 °C.A single colony of ETEC F4 was used to inoculate 10 mL of TSB in a 50 mL sterile tube followed by aerobic incubation at 37 °C for 16 to 18 h with shaking (150 rpm).The culture was further subcultured by inoculating 10 mL TSB with 1% inoculum and incubated for 2.5 h at 37 °C with shaking (150 rpm) until its OD 600 value reached 0.3.The ETEC F4 culture was diluted by PBS to obtain a cell suspension at approximately 1 × 10 7 CFU•mL −1 .

Selection of piglets
The piglets susceptible to ETEC F4 were selected for the ETEC challenge trial following the method described by Jensen et al. (2006).Briefly, the tails of piglets were collected 3 days after farrowing when docked.DNA was extracted according to previous publication (Truett et al., 2000).The mucin 4 (MUC4) gene was detected by a PCR assay.The assay mixture (25 µL) was composed of DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA), 200 µmol•L −1 of dNTP, 2 mmol•L −1 MgCl 2 , 400 µmol•L −1 of forward primer (5ʹ-GTGCCTTGGGTGAGAGGTTA-3ʹ), and reverse primer (5ʹ-CACTCTGCCGTTCTCTTTCC-3ʹ).Thermocycling was performed with initial denaturation at 95 °C for 5 min and subsequently by 35 cycles with each cycle having denaturation at 95 °C for 30 s, annealing at 65 °C for 30 s, and extension at 72 °C for 1 min.The PCR product size from the tail genomic DNA was 367 bp.The product (5 µL) was digested by XbaI (Thermo Fisher Scientific) for 5 min at 37 °C according to the protocols provided by the supplier.After digestion, the PCR products stained by SYBR Green (Invitrogen, Carlsbad, CA, USA) were electrophoresed on an agarose gel (2%) in a Tris-borate-EDTA buffer.The susceptible allele (S) was digested by FastDigest XbaI into two fragments with 151 and 216 bp, respectively, whereas the resistant allele (R) was indigestible.Similar body weight piglets contained both the SS-genotype (homozygote) and RS-genotype (heterozygote) were chosen for the experiment, as Matsumoto et al. (2020) revealed that RS-genotyped pigs showed no significant differences in fecal score, incidence of diarrhea, and diarrhea duration compared to SS-genotyped pigs.

Experimental Design
Thirty-two weaned pigs susceptible to ETEC F4 (TN Tempo × TN70; half castrated male and half female piglets; average body weight of 8.15 ± 0.18 kg) were transferred to T.K. Cheung Centre for Animal Science Research from the Glenlea Swine Research Unit of the University of Manitoba at the age of 28 d (weaning at 21 d).All piglets were housed in individual pens in a temperature-controlled room with the room temperature kept at 29 ± 1 °C within the prechallenge period (0 to 7 d), and after that decreased by 1.5 °C during the postchallenge period (8 to 10 d).All piglets for this trial were randomly allocated to four treatments (eight replicates per treatment, half castrated male and half female).The basal diet (Table 1) based on corn-soy was made to meet or exceed the recommendations for 7 to 11 kg piglets from the NRC (2012).The treatments were designed as follows: 1) non-challenged negative control group (NNC; basal diet and piglets gavaged with PBS), 2) negative control group (NC; basal diet and piglets challenged with ETEC F4, 3 × 10 7 CFU per pig), 3) positive control (PC; basal diet + 80 mg•kg −1 of avilamycin and piglets challenged with ETEC F4), and 4) probiotic candidate (PF9; control basal diet + 2.5 × 10 9 CFU•kg −1 diet of B. licheniformis PF9 and piglets challenged with ETEC F4).The whole trial lasted for 10 days.Piglets had ad libitum access to feed and water throughout the whole trial.Individual piglet's body weight and pen feed disappearance were recorded to calculate average daily feed intake (ADFI), average daily gain (ADG), and gain:feed within the prechallenge period and postchallenge period, respectively.For the ETEC F4 challenge, each piglet in the NC, PC, and PF9 groups was administered with 3 mL of ETEC F4 (1 × 10 7 CFU•mL −1 ) on day 7 (Koo et al., 2020).The rectal temperature was measured by thermometer before the inoculation, 3, 24, and 48 h postinoculum (hpi).The diarrhea index score was recorded at 3, 8, 12, 16, 20, 24, 28, 32, 36, 40, and 48 hpi following the method of Marquardt et al. (1999).

Collection of Tissue and Digesta Samples
All piglets from this trial were anesthetized with ketaminexylazine and euthanized by a captive bolt gun at the end of the trial (on day 11).First, the abdomen of the pig was quickly opened up, and the gut was quickly removed for sample collection.A 10 cm of jejunum (400 cm from the stomachduodenum junction) was obtained, put in a tube containing cold Krebs ringer bicarbonate (KRB) buffer, and immediately transferred to the laboratory to perform the permeability assay using a Ussing chamber.The other 15 cm of the jejunum was collected and quickly frozen in liquid nitrogen prior to storage at −80 °C for follow-up analyses.A section of 3 cm each of the duodenum (8 cm from the stomach), mid-jejunum (4 m from the stomach-duodenum junction), and mid ileum (10 cm from the ileum-cecum junction) was removed and fixed in 10% formaldehyde solution for histological analysis Xu et al. of gut morphology.The digesta from the ileum (5 cm of ileum from the middle) and colon (2 cm of colon from the middle) were also obtained and quickly frozen in liquid nitrogen before storage at −80 °C for the follow-up analyses (Choi et al., 2020).

In Vivo Gut Permeability
Each piglet received by oral gavage 10 mg in 5 mL PBS buffer of fluorescein isothiocyanate-dextran 70 kDa (FITC-D70; Sigma-Aldrich Co., St. Louis, MO, USA) on 3 d postinoculum.Four hours after gavage, blood samples from the jugular vein of each piglet were collected into vacutainer tubes (Becton Dickinson, Rutherford, NJ, USA) covered by aluminum foil for light blocking, and allowed to clot for 3 h at 22 °C.After centrifuging for 15 min at 750 × g, the obtained serum samples were stored at −80 °C for further analyses.The FITC-D70 content in the serum was measured by a Bio-Tek PowerWave HT Microplate Scanning Spectrophotometer (Bio-Tek Instruments Inc., Winooski, VT, USA) under the excitation wavelength of 485 nm and emission wavelength of 528 nm and calculated through the standard curve.

Analysis of Transepithelial Electrical Resistance
Transepithelial electrical resistance (TEER) was determined as the electrophysiological property by a Ussing chamber (Physiologic Instruments Inc., San Diego, CA, USA) with voltage and current electrodes filled with KRB buffer and housed in 3% agar bridges.Five milliliters KRB buffer solution was filled in the mucosal chambers and the serosal chambers with 10 mmol•L −1 d-mannitol and d-glucose, respectively.The chambers were kept at a temperature of 37 °C through a water-jacketed reservoir and continued to pump the 95% O 2 and 5% CO 2 mixed gas.The tissue without serosal and longitudinal muscle layers was mounted in Ussing chambers, equilibrated for 10 min, and then the TEER was recorded after mounting for 10 min (Choi et al., 2020).

Gut Morphology
The gut tissue samples were embedded in paraffin after being fixed in a 10% formaldehyde solution, sliced into samples with 5 µm thickness, and then transferred onto glass slides.The dewaxed parts were immersed in xylene, 100% ethanol, and 95% ethanol for 5 min each for two cycles.After that, the gut tissues were left in Alcian blue at 22 °C for 15 min and washed with pure water for 2 min and in Schiff for 10 min followed by another washing in pure water for 10 min.Finally, hematoxylin was used for a second staining for 10 s, and then the samples were washed with water and dehydrated.Each slide was photographed and visualized using a digital camera (Lumenera Corp., Ottawa, ON, Canada) and microscope equipped with 10× lenses (Carl Zeiss Microscopy Deutschland GmbH, Göttingen, Germany) for quantification of Alcian blue/Periodic acid-Schiff staining.Fifteen villi and associated crypts of each sample were selected for measuring villus height (VH), crypt depth (CD), and VH:CD using a Infinity Analyze software (Lumenera Corp.; Choi et al., 2020).

Analysis of Intestinal Digestive Enzyme Activities
Maximal enzyme activity (V max ) of intestinal digestive enzymes in the mid-jejunum including sucrase, maltase, intestinal alkaline phosphatase (IAP), aminopeptidase N (APN), and maltase-glucoamylase (MGA) were all measured.Specifically, around 200 mg of frozen mid-jejunum tissues were homogenized on ice in a cold homogenizing buffer (phenylmethylsulfonyl fluoride and d-mannitol, pH 7.4) with a polytron homogenizer.The original tissue homogenate sample suspension was transferred into the tube to determine the protein content.The activities of sucrase and maltase were measured following the method written by Dahlqvist (1964).The activity of IAP was analyzed following the method of Hübscher and West (1965), and APN activity was measured using the procedure of Maroux et al. (1973).MGA activity was determined using the procedure of Lackeyram et al. (2012).

Assays of Total Glutathione, Glutathione/Oxidized Glutathione, and Antioxidant Capacity
Total antioxidant capacity (TAC) in the mid-jejunum was analyzed using the Total Antioxidant Capacity Kits of Colorimetric Microplate Assay (Oxford Biomedical Research Inc., Oxford, MI, USA; Yang, 2011).Specifically, about 200 mg of mid-jejunal tissues were homogenized with 1 mL of cold PBS for 30 s on ice and centrifuged for 12 min at 3,600 × g at 4 °C.Firstly, 25 µL supernatant was used for protein level analysis.Secondly, the TAC in an aliquot of supernatant was determined by all antioxidants' capacity to convert Cu 2+ to Cu + following the manufacturer's protocol.Cu + was complexed with bathocuproine to create a stable form which was detected by Bio-Tek PowerWave HT Microplate Scanning Spectrophotometer (Bio-Tek Instruments Inc.) with a 96-well plate reader at a wavelength of 450 nm.The values were based on the standard curve (Choi et al., 2020).Glutathione (GSH) and oxidized glutathione (GSSG) levels of the mid-jejunum were analyzed using the detection kit for glutathione colorimetric according to the manufacturer's instructions (Invitrogen).Specifically, around 30 mg of tissues were homogenized for 30 s on ice in 750 µL of ice-cold PBS followed by centrifugation for 10 min at 3,600 × g at 4 °C.After determining the protein concentration, 5-sulfosalicylic acid dihydrate used for precipitating protein was added into the supernatant followed by another centrifugation for 10 min at 3,600 × g at 4 °C.The protein level of all homogenized tissues was measured by a BCA Protein Assay Kit (Thermo Fisher Scientific).The total GSH and GSSG levels were then measured using this equation: Reduced GSH = Total GSH − 2 × GSSG (Choi et al., 2020).

RNA Extraction and Gene Expression Analysis
Total RNA in 50 mg mid-jejunal tissues was extracted by the total RNA isolation kit (Thermo Fisher Scientific Inc.).The RNA concentration and OD 260 :OD 280 ratios were measured by a Nanodrop spectrophotometer (Thermo Fisher Scientific Inc., Ottawa, ON, Canada).One microgram of RNA was used for synthesizing cDNA using an iScript cDNA Synthesis Kit (Bio-Rad Laboratories, Mississauga, ON, Canada) following the protocol provided by the manufacturer.Primer-Blast was used for designing the primers (Table 2) and synthesized by Integrated DNA Technologies, Inc. (Coralville, IA, USA).A total volume of 20 μL containing 1 μL cDNA, 300 nmol•L −1 of each forward and reverse primers and 10 μL SYBR Green supermix (Bio-Rad, Ontario, Canada) was used for Real-time PCR assays, which were performed on a Real-Time PCR Detection System (Bio-Rad Laboratories, Ontario; Omonijo et al., 2019).All reaction conditions were denaturation at 95 °C for 3 min, 40 cycles of 20 s at 95 °C, 30 s at 60 °C, and then 30 s at 72 °C.The target mRNA abundance was normalized with GAPDH and relative mRNA levels were determined by the 2 −ΔΔ CT method (Livak and Schmittgen, 2001).The values of the threshold cycle (Ct) were acquired when targeted genes were amplified above a threshold of 30 fluorescence units.Duplication of each gene was performed for each sample (Choi et al., 2020).

Microbiota Analysis
The PCR amplification and 16S rRNA sequencing were performed by Genome Quebec.Bacterial DNA from ileal and colonic digesta was extracted using a Fast DNA Stool Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's protocol.Extracted DNA from the digesta was amplified by PCR, targeting the V3-V4 hypervariable region of the 16S rRNA gene with a forward primer 341 F (5ʹ-CCTACGGGNGGCWGCAG-3ʹ) and a reverse primer 805 R (5ʹ-GACTACHVGGGTATCTAATCC-3ʹ; Caporaso et al., 2012).The PCR reaction system was 25 µL, containing 1× reaction buffer (Q5), 5% DMSO, l µL DNA template, 0.2 mM dNTP, 0.5 U Q5 DNA polymerase, and 0.6 µM of forward and reverse primers.The PCR amplifications were performed according to the following procedures: 30 s of 98 °C denaturation, followed by 25 cycles of 10 s at 98 °C, annealing at 60 °C for 15 s, and elongation at 72 °C for 30 s, with a final extension at 72 °C for 2 min.After verification of the amplificons on 2% agarose gel, the barcoding step was conducted with a total volume of 20 µL, containing 1× reaction buffer, 5% DMSO, 0.2 mM dNTP, 1.8 mM Mg 2+ , 0.2 µM primers, 0.5 U polymerase, and 1 µL DNA template.Initially, the PCR program started with 95 °C for 10 min; 95 °C for 15 s, 60 °C for 30 s, and 72 °C for 1 min, repeat for 15 cycles; 72 °C for 3 min.The verification of barcode incorporation for each sample was then on 2% agarose gel.Each amplicon was quantified with the QuantiT PicoGreen dsDNA Assay Kit (Thermo Fisher Scientific).The library was then generated by pooling the same quantity of each amplicon and cleaning-up the pool with sparQ PureMag Beads (Quantabio, Beverly, MA, USA).Afterward, the library was quantified using Kapa Illumina GA with Revised Primers-SYBR Fast Universal kit (Sigma-Aldrich Canada Co., Oakville, ON, Canada).The average fragment size was determined using a LabChip GX (PerkinElmer, Waltham, MA, USA) instrument.Before sequencing, 5% of the PhiX control library was added to the amplicon pool (loaded at a final concentration of 10 pM).The PhiX control library helps to improve the unbalanced base composition of the flowcell.Illumina MiSeq sequencing was carried out by Genome Quebec.Sequences were inputted into Quantitative Insights Into Microbial Ecology 2 (QIIME2; version 2021.8) for bioinformatics analysis (Caporaso et al., 2010;Bolyen et al., 2019).

Statistical Analyses
Survival curves for C. elegans were compared using the Kaplan-Meier survival analysis of SAS (V.3.8;SAS Institute, Cary, NC, USA) followed by a log-rank test.One-way analysis of variance (ANOVA) and the Tukey's multiple comparisons were used to test for significant differences between the means.Tolerance to low pH and high bile salt data were analyzed by the SAS GLM procedure and Tukey multiple comparisons.Means with P < 0.05 were considered a significant difference.
PROC MIXED of SAS with each piglet as the experimental unit in a randomized complete block design were used for analyzing the data.The model included sex (barrows and gilts) and treatment as the main effects and the interaction of sex by treatment.The least-square means with the option of Tukey-adjusted PDIFF was applied to split and calculate the treatment mean value.The NC group was compared with either the NNC group by preplanned contrasts to determine the effects of ETEC F4, or the PC group or the PF9 group to determine the avilamycin or the B. licheniformis effects, respectively.Results shown in tables are indicated as LSMEANS and pooled SEM, while results shown in figures are indicated as mean ± SEM.A P-value of <0.05 was considered as a significant difference.
Statistical analysis and data visualization for ileal and colonic microbiota were performed by the R program (version 4.0.2).Chao1, Shannon, and Simpson alpha diversity indices and relative abundance were performed by the phyloseq package.Ggplot2 package in the R program was applied to visualize the relative abundance, which aggregated at different taxonomical levels.Significant relative abundance was observed by the Kruskal-Wallis rank sum test.The conover test with the agricolae package was used for multiple pairwise comparison.Beta diversity was performed by the Bray-Curtis dissimilarity.

Species Identification of Isolate PF9 and Its Tolerance to Low pH and High Bile Salt
According to the final test report from the Agriculture and Food Laboratory, University of Guelph, sequencing of the 16S rRNA gene (1,485 bp) of isolate PF9 revealed >99.8% similarity to B. licheniformis or B. paralicheniformis compared with known 16S rRNA gene sequences in the NCBI GenBank database.In contrast to a B. paralicheniformis strain that served as a positive control in the assay, isolate PF9 showed no amplicons of fenC and fenD, the two markers from the fengycin operon of B. paralicheniformis, confirming that the isolate PF9 belongs to B. licheniformis.In the antimicrobial assays, the isolate partially inhibited the growth of ETEC JG280 and Salmonella Typhimurium DT104 (18%-30%).
In the tolerance assay to low pH, B. licheniformis PF9 had similar optical OD 600 values after 28 h growth at pH 7.0 regardless of the culture being pretreated at pH 2.0 or 5.6, indicating its tolerance to low pH (Fig. 1A).In the tolerance assay to high bile salt, B. licheniformis PF9 at 0.1% (v/v) concentration of bile salt showed similar growth with the control (0%) group (Fig. 1B).At 0.5% (v/v), the isolate had poorer growth than the control group before 36 h incubation.However, after 72 h, the growth of the isolate reached more than 1.0 at OD 600 without a significant difference (P > 0.05) compared to the control (0%).At 1.0% and 1.5% (v/v), the growth of the isolate was suppressed until 48 h incubation.After 72 h, the same cultures had about half of the OD 600 values of the control (P < 0.05).These results suggest that B. licheniformis PF9 has a certain degree of tolerance to bile salt.

Protection to C. elegans Infected With ETEC F4
Figure 2 shows the effects of B. licheniformis PF9 on the lifespan of C. elegans infected with ETEC JG280.The lifespan of worms infected with ETEC JG280 was dramatically decreased after 8 d of incubation.However, the nematode pretreated with B. licheniformis PF9 had a significantly extended lifespan (P < 0.05) compared to those infected with ETEC JG280 only.On day 12, only about 10% of the worms survived the ETEC infection, while the survival rate was increased to nearly 60% by the B. licheniformis PF9 pretreatment.
Growth Performance, Diarrhea Index Score, and Rectal Temperature The statistical analysis demonstrated no significant interactions (P > 0.05) of sex and treatments for the variables studied (data not shown).Therefore, both male and female pigs were combined for the following analyses.There was no significant difference (P > 0.05) in the ADG, ADFI, and gain:feed found among all groups during the prechallenge period (0 to 7 d) and whole period (0 to 10 d; Table 3).The ADFI among all treatment groups during the postchallenge period did not show any significant differences (P > 0.05).
Compared to the NNC group, the inoculation of ETEC F4 significantly decreased ADG and gain:feed in the NC group (P < 0.05).As shown in Fig. 3, no significant difference (P > 0.05) was found in the rectal temperature observed among all treatments during 48 hpi.Compared to the NNC group, the ETEC F4 challenge significantly (P < 0.05) increased the severity of diarrhea at 3, 36, and 40 hpi in the NC group.The supplementation of B. licheniformis significantly (P < 0.05) reduced diarrhea score at 3 hpi compared to the NC group (Fig. 4).Xu et al.

Gut Morphology
The inoculation of ETEC F4 significantly (P < 0.05) decreased duodenal, jejunal, and ileal VH in the NC group compared to the NNC group, and there was a significant decrease (P < 0.05) in duodenal VH between the PC and NNC group (Table 4).However, there was no significant difference (P > 0.05) in all small intestinal CD and VH:CD among all treatments.

Gut Permeability
As shown in Table 5, there was no significant difference (P > 0.05) observed in the TEER measured by the Ussing chamber and in vivo gut permeability among all treatments.
Intestinal Digestive Enzyme Activities, Total GSH, GSH/GSSG, and TAC There was no significant difference (P > 0.05) in the V max values of APN, IAP, MGA, sucrase, and maltase among all treatments (Table 6).No significant difference (P > 0.05) was found in the GSH, GSSG, reduced GSH, reduced GSH:GSSG ratio, and TAC among all treatments (Table 6).

Relative mRNA Abundance in the Jejunum
The relative mRNA levels of genes in the mid-jejunum related to gut barrier function, nutrient transportation, digestive enzymes, and immunity were determined by Real-time PCR assays.As shown in Table 7, no significant difference (P > 0.05) was observed in the relative mRNA levels of claudin 3, MUC2, maltase-glucoamylase (MGA), peptide transporter 1, neutral amino acid transporter 2, neutral amino acid transporter, interleukin-1β (IL-1β), IL-6, IL-8, IL-10, toll-like receptor 5 (TLR5), and TLR7 among all treatments.Compared to the NNC group, the inoculation of ETEC F4 significantly (P < 0.05) decreased the relative mRNA levels of Na + -glucose cotransporter 1 (SGLT1) in the NC group.When compared Values within a row with different superscripts differ significantly at P < 0.05. to the NC and NNC groups, the group supplemented with B. licheniformis PF9 significantly (P < 0.05) increased the relative mRNA levels of SGLT1, occludin (OCLN), zonula occludens-1 (ZO-1), and APN.Moreover, the supplementation of B. licheniformis significantly (P < 0.05) increased the relative mRNA level of excitatory amino acid transporter 1 (EAAC1) when compared to the NC group.

Gut Microbiota
In the current study, 3,906,109 qualified reads in total were obtained.A mean of 69,751 reads per sample and 2,862 operational taxonomic units in total were identified.The top phyla in the ileal digesta of non-challenged pigs were Firmicutes, which was replaced by Proteobacteria in challenged pigs.The colonic digesta was dominated by Bacteroidetes, Firmicutes, and Proteobacteria (Fig. 5).Piglets supplemented with B. licheniformis PF9 had a lower relative abundance of Bacteroidetes in the colon than piglets in the NNC group (P < 0.05).Non-challenged piglets had a higher relative abundance of Firmicutes in the ileum than the challenged piglets (P < 0.05); however, a lower relative abundance of Proteobacteria (P < 0.05) in the ileum and colon was observed in the NC group only (Fig. 6).For alpha diversity (Chao1, Shannon, and Simpson indices), there was no significant difference (P > 0.05) in the ileum and colon among all treatments (Fig. 7).No significant difference (P > 0.05) was found in the beta diversity (Bray-Curtis distance) of ileal and colon digesta among treatments (Fig. 8).

DISCUSSION
The aim of this study was to characterize B. licheniformis PF9 and investigate its potential to relieve the negative effects of  Values within a row with different superscripts differ significantly at P < 0.05.

Xu et al.
ETEC infection on weaned piglets, which impaired gut health including diarrhea and inflammation, by conducting both in vitro and in vivo studies.The B. licheniformis PF9 showed tolerance to low pH and high bile salt in vitro, which provided the basic physiological foundation for the following in vivo animal trial.The same justification was also offered by the lifespan assay of C. elegans, in which B. licheniformis PF9 partially prevented the nematode from death caused by ETEC infection.Our previous studies have demonstrated that Lactobacillus was able to protect C. elegans against ETEC infection by suppressing the gene expression of STa and STb enterotoxins in ETEC (Zhou et al., 2014).In addition, the same Lactobacillus isolate could regulate the cell signaling of C. elegans to increase the production of defense molecules to combat ETEC infection (Zhou et al., 2018).It is unknown at present whether B. licheniformis PF9 has the same molecular mechanisms as the Lactobacillus isolate in the protection of C. elegans or not.Further studies are required to reach a conclusion.
The ETEC F4 challenge model in weaned piglets was established for inducing enteric infection (Opapeju et al., 2015).The ETEC F4 virulence and F4 fimbriae receptors in piglets are the two main factors leading to the pathogenesis of ETEC F4 (Luppi et al., 2016).Single nucleotide polymorphisms on F4 receptors MUC4 have been frequently used as the genetic markers for ETEC F4 susceptibility or resistance in pigs; therefore, the ETEC F4 susceptible piglets were chosen in the current study after gene screening of the susceptible alleles of MUC4 (Jørgensen et al., 2003).
In the present study, the supplementation of B. licheniformis PF9 did not affect the growth performance of piglets during the prechallenge period.The inoculation of ETEC F4 significantly decreased the ADG and gain:feed, which was similar to the findings of Ren et al. (2014).The potential mechanisms related to decreased ADG and gain:feed of piglets challenged with ETEC F4 have been attributed to reduced nutrient utilization efficiency (Choi et al., 2020); induced diarrhea (Sayan et al., 2018) and inflammatory responses (Xu et al., 2020); and reduced nutrients available for host growth because of ETEC F4 competition for available nutrients with the piglets (Mingmongkolchai and Panbangred, 2018).Inconsistent effects of the supplementation of Bacillus on pig growth performance have been reported.Pan et al. (2017)    Values within a row with different superscripts differ significantly at P < 0.05. in growth performance.These inconsistent results may be due to the differences in tested Bacillus isolates, their physiological status, ETEC strains (e.g., F4 and F18), and tested doses (de Lange et al., 2010) or experimental designs (e.g., the experimental period including pretreatment with the probiotic bacterium).
The higher diarrhea index score due to the ETEC challenge was an indicator showing the success of ETEC F4 infection in piglets (Luise et al., 2019b).In the present study, the supplementation of B. licheniformis PF9 reduced the severity of diarrhea to score values similar to those obtained with the antibiotic diet.It showed that both B. licheniformis PF9 and avilamycin mitigated the early stages of diarrhea caused by the ETEC F4 infection.Thus, B. licheniformis PF9 could be a promising alternative to antibiotics.This observation is consistent with a previous study described by Chen et al. (2021), which may in part be explained by enhanced gut barrier function with increased mRNA expression of the tight junction proteins ZO-1 and OCLN in the jejunum detected in the present study.This speculation was supported by the observation that the same isolate (B.licheniformis PF9) increased the expression levels of ZO-1 and OCLN in the ETEC F4-infected IPEC-J2 cells (Li et al., 2022).It has been well reported that the supplementation of probiotics can improve the gut barrier function of piglets (Wang et al., 2018;Yi et al., 2018).
To study the gut permeability of weaned piglets, 2 mg•mL −1 FITC-D70 in PBS was orally delivered to the piglets by gavage.In brief, FITC-D70 cannot pass through the epithelial barrier in the gut and be digested by gut digestive enzymes during normal healthy conditions; however, once the tight junction proteins have been affected by a pathogen, toxin, or inflammation, the FITC-D70 molecule could enter the blood circulation and could be detected in serum samples (Choi et al., 2020).The impairment of gut barrier function by the ETEC F4 challenge could increase the gut permeability with increased FITC-D70 flux, which affects several alternations in the morphology of the gut.It has been reported that the supplementation of Bacillus subtilis reduced paracellular permeability determined by measuring the flux of FITC-D4 across the jejunal mucosa (Kim et al., 2019).The inconsistent results in our current study may be due to the different strains.
The higher VH could indicate better nutrient digestion and absorption as the enterocytes located in the villus of piglets were critical in nutrient utilization (Chen et al., 2018).In the present study, the inoculation of ETEC F4 decreased the VH of the small intestine, which is consistent with a report by Choi et al. (2020).ETEC F4 adhering to the enterocyte brush border membrane of gut mucosa causing villous atrophy may be the possible reason for decreasing VH (Luppi, 2017).In the current study, the supplementation of B. licheniformis demonstrated a tendency to increase VH in the ileum when compared with the NC group, which is similar to a study from Zong et al. (2019).This may partially explain the increased relative mRNA abundance of the brush border digestive enzymes, APN and MGA, in B. licheniformis PF9treated piglets.
In this study, the ETEC F4 infection significantly decreased the relative mRNA level of SGLT1 in the mid-jejunum,  which were similar results to those found in a study by Choi et al. (2020).The SGLT1 plays an important role in the gut glucose transport system in pigs (Yang et al., 2011).The decreased SGLT1 may be associated with secreted toxins and villous atrophy from ETEC F4 (Wu et al., 2015).In the present study, the supplementation of B. licheniformis PF9 increased the relative mRNA expression of SGLT1 and EAAC1.These data could be supported by Cao et al. (2020) reporting that the increased expression of nutrient transporters in the intestine can reflect the strong absorption function of epithelial cells in the small intestine due to the supplementation of probiotics.
The diversity and composition of gut microbiota in piglets are strongly related to animal health conditions and nutrient compositions that are provided by animal diets (Guevarra et al., 2018).In the present study, the supplementation of B. licheniformis PF9 showed no impact on the microbiota diversity in the digesta of ileum and colon, which were similar results to those found in a study by He et al. (2020).Nonetheless, with findings that were inconsistent with those previously published by Duarte et al. (2020), the ETEC infection significantly increased the relative abundance of Proteobacteria and reduced the relative abundance of Firmicutes.This may be partially explained by the production of enterotoxins due to ETEC infection and secretion of fluid to the gut lumen (Wang et al., 2019), which creates an ideal environment for the growth of Proteobacteria (Gresse et al., 2017).The results from the current study indicated that the supplementation of B. licheniformis PF9 decreased the relative abundance of Bacteroidetes, which is consistent with the study reported by Cui et al. (2013).
In conclusion, results from the current study demonstrated that the heat-resistant B. licheniformis PF9 was tolerant to low pH and high bile salt.It provided protection to C. elegans against ETEC infection.The infection of ETEC F4 impaired growth performance, damaged gut morphology, induced diarrhea, and changed the gut bacterial population in weaned piglets.The supplementation of B. licheniformis PF9 reduced diarrhea score, increased the relative mRNA level of several nutrient transporters, the gut barrier-associated proteins, and digestive enzymes, and lowered the relative abundance of Bacteroidetes in the ETEC F4-challenged weaned piglets.These results suggest that B. licheniformis PF9 has the potential for improving pig gut health.Further studies are needed to evaluate the different dose of this isolate and a cocktail of probiotics containing different probiotic candidates for their potential in controlling enteric infections and improving the gut health and performance of pigs.

Figure 2 .
Figure 2. Lifespan of C. elegans with or without the pretreatment of B. licheniformis isolate PF9 after infection with enterotoxigenic Escherichia coli (ETEC).Treatments: •, treated with E. coli OP50 only; ■, treated with ETEC JG280; ▲, treated with B. licheniformis isolate PF9 and then ETEC JG280.All the groups showing different letters were significantly different (P < 0.05) in their survival curves.

Figure 3 .
Figure 3. Effects of B. licheniformis PF9 on rectal temperature in piglets.Rectal temperature was measured in the non-challenged negative control (NNC) group with basal diet and piglets gavaged with PBS; negative control (NC) group with basal diet and piglets challenged with enterotoxigenic E. coli (ETEC) F4; positive control (PC) group with basal diet + 80 mg•kg −1 of avilamycin and piglets challenged with ETEC F4; PF9 group with basal diet + 2.5 × 10 9 CFU•kg −1 diet of B. licheniformis and piglets challenged with ETEC F4.Each value is represented by the mean ± SEM.

Figure 4 .
Figure 4. Effects of B. licheniformis PF9 on diarrhea index score in weaned piglets.The diarrhea index score was measured in the non-challenged negative control (NNC) group with basal diet and piglets gavaged with PBS; negative control (NC) group with basal diet and piglets challenged with enterotoxigenic E. coli (ETEC) F4; positive control (PC) group with basal diet + 80 mg•kg −1 of avilamycin and piglets challenged with ETEC F4; PF9 group with basal diet + 2.5 × 10 9 CFU•kg −1 diet of B. licheniformis and piglets challenged with ETEC F4.Diarrhea index score 0 represents normal feces; 1 represents soft feces; 2 represents mild diarrhea; and 3 represents severe diarrhea.Each value is represented by the mean ± SEM.Different letters shown in the bars indicate significant difference (P < 0.05).

Figure 5 .
Figure 5. Stacked bar plot showing the relative abundance of bacterial phyla in the ileum and colon.Non-challenged negative control (NNC) group with basal diet and piglets gavaged with PBS; negative control (NC) group with basal diet and piglets challenged with enterotoxigenic E. coli (ETEC) F4; positive control (PC) group with basal diet + 80 mg•kg −1 of avilamycin and piglets challenged with ETEC F4; PF9 group with basal diet + 2.5 × 10 9 CFU•kg −1 diet of B. licheniformis and piglets challenged with ETEC F4.

Figure 6 .
Figure 6.Bar plot showing the relative abundance of bacterial phylum in ileum and colon.a,b Means without a common superscript are different (P < 0.05).Non-challenged negative control (NNC) group with basal diet and piglets gavaged with PBS; negative control (NC) group with basal diet and piglets challenged with enterotoxigenic E. coli (ETEC) F4; positive control (PC) group with basal diet + 80 mg•kg −1 of avilamycin and piglets challenged with ETEC F4; PF9 group with basal diet + 2.5 × 10 9 CFU•kg −1 diet of B. licheniformis and piglets challenged with ETEC F4.

Figure 7 .
Figure 7. Alpha diversity as indicated by Chao1, Shannon, and Simpson index in digesta of ileum and colon.Non-challenged negative control (NNC) group with basal diet and piglets gavaged with PBS; negative control (NC) group with basal diet and piglets challenged with enterotoxigenic E. coli (ETEC) F4; positive control (PC) group with basal diet + 80 mg•kg −1 of avilamycin and piglets challenged with ETEC F4; PF9 group with basal diet + 2.5 × 10 9 CFU•kg −1 diet of B. licheniformis and piglets challenged with ETEC F4.

Figure 8 .
Figure 8. PCA plot of beta diversity of microbiota in digesta of ileum and colon based on Bray-Curtis dissimilarities.Non-challenged negative control (NNC) group with basal diet and piglets gavaged with PBS; negative control (NC) group with basal diet and piglets challenged with enterotoxigenic E. coli (ETEC) F4; positive control (PC) group with basal diet + 80 mg•kg −1 of avilamycin and piglets challenged with ETEC F4; PF9 group with basal diet + 2.5 × 10 9 CFU•kg −1 diet of B. licheniformis and piglets challenged with ETEC F4.

Table 3 .
Effects of B. licheniformis PF9 on the growth performance of weaned piglets during the prechallenge period (0 to 7 d), postchallenge period (8 to 10 d), and whole period 1 NNC, non-challenged negative control group with basal diet and piglets gavaged with PBS; NC, negative control group with basal diet and piglets challenged with ETEC F4; PC, positive control group with basal diet + 80 mg•kg −1 of avilamycin and piglets challenged with ETEC F4; PF9, isolate PF9 group with basal diet + 2.5 × 10 9 CFU•kg −1 diet of B. licheniformis and piglets challenged with ETEC F4; ADG, average daily gain; ADFI, average daily feed intake.a,b

Table 4 .
Effects -challenged negative control group with basal diet and piglets gavaged with PBS; NC, negative control group with basal diet and piglets challenged with ETEC F4; PC, positive control group with basal diet + 80 mg•kg −1 of avilamycin and piglets challenged with ETEC F4; PF9, isolate PF9 group with basal diet + 2.5 × 10 9 CFU•kg −1 diet of B. licheniformis and piglets challenged with ETEC F4.
of B. licheniformis PF9 on gut morphology with villus height (VH, μm), crypt depth (CD, μm), and VH:CD in the duodenum, jejunum, and ileum of weaned piglets a,b reported that the supplementation of probiotics containing B. licheniformis and Saccharomyces cerevisiae had positive effects on ADG and gain:feed in weaned piglets challenged with ETEC K88.However, Luise et al. (2019a) reported that B. amyloliquefaciens and B. subtilis did not show improvement

Table 7 .
Effects of B. licheniformis PF9 on the relative mRNA levels of genes related to gut barrier function, nutrient transportation, immunity, and digestive enzymes in the mid-jejunum of weaned piglets 1