The study was conducted in the Feed Research and Innovation Center, Charoen Pokphand Foods (CPF) Public Company Limited (PLC.). The experimental protocols and methods in this study were carried out in compliance with the ARRIVE guidelines. The in vivo experimental study was approved according to the guidelines for experimental animals established by the Institute Animal Care and Use Committee of the Feed Research and Innovation Center of CPF (FRIC-ACUP-1707013). The use of the LAB strains was approved by the Institutional Biosafety Committee, Chulalongkorn University (IBC1631047).
The euthanasia procedures were performed as following the guidelines for the euthanasia of animals complied with the American Veterinary Medical Association (AVMA). All pigs were humanely terminated by electrocution and exsanguination techniques. Briefly, pigs were rendered unconscious by electrical stunning with the head-only application. They were then immediately cut the major blood vessels in the neck, resulting in a rapid fall in blood pressure, leading to a lack of blood to the brain and death. All efforts were made to minimize the suffering.
Bacterial strain used in the experiment
The three strains of LAB that were used were previously isolated from the faeces of antibiotic-free healthy pigs. These bacteria were identified as Lactobacillus plantarum (strains 22F and 25F) and Pediococcus acidilactici (strain 72N) and were characterized in vitro for their probiotic properties in relation to: resistance to acid and bile; lack of antimicrobial-resistance genes using European Food Safety Authority (EFSA) criteria; antibacterial properties against E. coli and Salmonella; and interference with porcine endemic diarrhoea virus18-20.
The probiotic bacteria were stored at -80°C in De Man, Rogosa and Sharpe (MRS) broth (Becton, Dickinson and Company, Maryland, USA) containing 20% glycerol. Bacterial strains were grown in aerobic conditions at 37°C for 18-20 h in MRS medium. Each LAB strain was harvested by centrifugation (3,000 g, 4°C, 10 min), washed, and resuspended in sterile normal saline separately to obtain a final concentration of 109 CFU/mL58. Three milliliters of these suspensions were orally delivered to each of the animals in the corresponding probiotic supplement feed groups on the designated days.
Microencapsulation of probiotic strains
Previously it has been shown that of the three LAB strains, L. plantarum strain 22F gave the best in vitro performance19. Hence, this strain was selected to use in the microencapsulation procedure. Alginate (1% w/v) (Sigma-Aldrich, Missouri, USA) and chitosan (0.4% w/v) (Union Chemical 1986, Bangkok, Thailand) were used as inner and outer wall materials. A total of 109 CFU/mL of L. plantarum strain 22F was added at a ratio of 1:5 (v/v) to alginate solution. The mixture was atomized through a spray dryer (Mini Spray Dryer B-290, Buchi, Flawil, Switzerland) with the inlet temperature set at 130°C, and then the alginate powder was collected. One gram of this powder was added to 100 mL of chitosan solution before atomizing through the spray dryer under the same conditions as previously described. These double-coated powders containing L. plantarum strain 22F were recovered from the collecting vessel and stored at room temperature for six months before use59.
Animals and Housing
After cross-fostering, a total of 240 healthy neonatal pigs (Large White × Landrace × Duroc) were randomly distributed into six experimental groups, with 2 male and 2 female replicate pens per group (10 pigs per pen). The piglets were housed in an environmentally-controlled building using an evaporative cooling system. For the nursery phase, each pen (1.6×1.6 m) was with stainless steel floor mats and a heated plastic mat cover, a feeder, and a water nipple. For the grower and finisher phase, each pen (6×6 m) was with a concrete floor stall, a feeder, and three water nipples. The housing was maintained at 27 to 28°C and 80% humidity. The photoperiod was controlled to provide 12 h of light and 12 h of dark.
Experimental design and sample collection
The six groups of pigs comprised: Group 1 (control) - no supplementation; Group 2 (ABO) - diet supplemented with antibiotics (Table 1); Group 3 (spraydry) - supplemented with spray-dried L. plantarum strain 22F; Groups 4-6 - supplemented with freshly prepared L. plantarum strain 22F (L22F), L. plantarum strain 25F (L25F) and P. acidilactici strain 72N (P72N), respectively.
On the designated treatment days, pigs in groups 1 and 2 were orally administered with 3ml of sterile peptone water (Becton, Dickinson and Company, Maryland, USA) by syringe. Pigs in group 3 received 3 mL of sterile peptone water containing 1 g of double-coated L. plantarum strain 22F that had been stored for six months. Pigs in groups 4-6 received 3ml of suspensions (109 CFU/mL) of L. plantarum strain 22F, L. plantarum strain 25F, and P. acidilactici strain 72N, respectively. Administrations commenced on the day of cross-fostering and were repeated five times (on days 0, 3, 6, 9, and 12 after cross-fostering). The piglets were allowed to suckle sow’s milk conventionally until weaning.
On weeks 1, 2, 3, 8, and 22, ten pigs (5 males and 5 females) in each experimental group were randomly selected for collection of faeces for microbial profile analysis. Then two pigs (1 male and 1 female) from each of these ten were euthanized, and the small intestines were collected for histological analysis. After the weaning period, pig body weight and feed intake were recorded weekly for performance evaluation. In addition, observations of morbidity and mortality were made daily throughout the experimental period. Throughout the experiment, all of the pigs had ad libitum access to tap water and a basal diet formulated following the NRC guidelines according to the pig’s body weight (Table 1). The pigs in the antibiotic group (Group 2) received the diet supplemented with the antibiotics shown in Table 1. These antibiotics and dose rates were those used in a commercial setting to control subclinical infections and improve growth rates, and had been developed on a semi-empirical basis and used over many years.
The performance data were divided into 3 phases: nursery (weeks 3-8), grower (weeks 8-15), and finisher (weeks 15-22). The body weight and feed intake from each experimental group were used to calculate average daily gain (ADG) and feed conversion rate (FCR). The pigs were examined daily for signs of ill-health. Moreover, the presence of any sick or dead pigs was intended to be included into the percentage of morbidity and mortality35,37.
Return on investment (ROI) analysis
The probiotic and antibiotic usage performances of the pigs were estimated based on the increased body weight at market age (BWm) compared to the control group using the ROI as follows Equation 160, where Net return represented the profit after excluding the total cost, and Total cost represented the total expense per pig for probiotic and antibiotic supplementation along the rearing cycle. The Net return, Increase income and Increased BWm were determined as follow Equation 2, 3, and 4, respectively. According to data from the Department of Economics and Trade, Thailand, the average liveweight price for the pigs at market age in January 2021 was 2.51 USD/kg.
ROI = Net return/Total cost (1)
Net return = Increased income - Total cost (2)
Increased income = Increased BWm × Average liveweight price for the pigs at market age (3)
Increased BWm = BWm of the probiotic or antibiotic supplemented group - BWm of the control group (4)
Faecal microbial count
On weeks 1, 2, 3, 8, and 22, faecal samples were obtained from the rectal swabs and placed into transport medium to maintain viability. These samples were kept on ice and immediately taken to the laboratory61. The samples from 10 pigs in each experimental group were pooled and mixed well with normal saline (1:9 w/v). The supernatants were subjected to serial dilution and plated at the appropriated dilution on MRS and MacConkey (Becton, Dickinson and Company, Maryland, USA) agar using the spread plate method for the determination of viable lactobacilli and Enterobacteriaceae cell counts, respectively. The plates were incubated at 37°C for 48 h37,41. Microbial enumerations were determined in triplicate and calculated as colony forming units (CF) per g.
Small intestinal tissues (duodenum, jejunum, and ileum) taken from two pigs in each group on weeks 1, 2, 3, 8, and 22 were immediately fixed with 10% neutral-buffered formalin, dehydrated in alcohol, cleared in xylene, and embedded in paraffin wax. Embedded tissues were cut with a microtome to achieve thin sections (4-6 μm thick) and stained with hematoxylin and eosin. The tissues were examined under the light microscope for assessment of villous height (VH), crypt depth (CD), and VH:CD ratio37,41 using Motic® Images Plus Version 2.0 (Motic, Texas, USA).
Data from the experiments were analysed with Prism 9 for macOS version 9.0.2 (134). Effects were considered significant at P < 0.05. Results were presented as mean ± standard error of the mean (SEM). The means of ADG, FCR, and BWm from all replications were determined in each group. Bacterial enumeration for the faecal microbial count, in log (CFU/g) units, were performed in triplicate. Twenty measurements of villi and crypts per sample were averaged to acquire VH and CD (μm) for each pig. Those parameters were used to calculate the VH:CD ratio. Analysis of data across groups was carried out using one-way ANOVA, and the comparison of means was tested by Tukey’s multiple range tests. Analyses of the combined effect of two variables, including experimental groups and period, were conducted with two-way ANOVA and Tukey’s multiple range tests.