Enhancing gut health and performance of grower-finisher pigs: assessing the efficacy of the mixture of microencapsulated essential oils and saponins

Animal experiment

The experiment was conducted in the specifically constructed pigsty to study the impact on the productivity of the grower to finisher pigs with phytogenic supplements. In total, 400 pigs (Large White × Landrace, males, 16–20 kg, 8 weeks of age) were randomised and separated into two groups (control and treatment). Each group consisted of 200 pigs, and 50 pigs from both groups were randomly selected and tagged for identification throughout the production process. Ten of these identified pigs were then chosen to follow through the production cycle. Animals were provided with ad libitum access to water and feed. Basal diets were composed of barley, wheat, sorghum and other grains as energy sources and soya meal with meat and bone meal (MBM) as protein sources, which were designed by the company nutritionists to meet the production and animal welfare requirements. A standard trace minerals and vitamin premix, organic acids and prebiotics and probiotics were added to the feed to cover the vitamin and mineral requirements.

Treatment groups were supplemented with commercial PFA (Aromex Pro, Delacon Biotechnik GmbH) containing a blend of micro-encapsulated essential oils of rosemary and thyme, and saponins, to the basal diet at the recommended dosage of 200 g/1000 kg. The phytogenic supplements were administered to the animals 7 days after their arrival and supplementation continued for 9 weeks, from 8 to 17 weeks of age.

All animals were supplemented with amoxycillin and a blend of organic acids (Selko-pH, AusPac Ingredients, Tamworth, NSW, Australia) through water for the first 2 weeks on arrival to the farm, and with oxytetracycline and Selko-pH through water thenafter.

The study monitored various performance data, including general health, cumulative mortality, and bodyweight of the grower-finisher pigs. Individual body weights of 50 pigs from each group were recorded, and a total of 150 rectal swab samples were collected using a sterile cotton-swab applicator three times during the trial period, at the ages of 8, 14 and 17 weeks. The samples were kept on dry ice during collection and transportation and stored at −80°C until DNA extraction. Week 17 samples were obtained 1 day prior to the transportation of the animals to the abattoir.

Intestinal content samples for preliminary insights on microbiota profile in different gut sections were collected from the duodenum, jejunum, ileum, caecum and colon in the abattoir from 10 randomly selected animals in each group.

DNA extraction, 16S rRNA gene sequencing and data analysis

The DNA from intestinal-content and rectal-swab samples was extracted as previously described (Bauer et al. 2019). Briefly, the cells were lysed as per the improved lysis protocol specifically developed for digesta and faecal samples (Yu and Morrison 2004) and purified using the DNA mini spin column kits (Enzymax LLC, CAT# EZC101, Lexington, Kentucky, USA). Following the extraction, the quantity and quality of the DNA were measured with a NanoDrop One UV-Vis spectrophotometer (ThermoFisher Scientific, Wilmington, USA).

The V3–V4 region of the 16S rRNA gene was amplified using the dual index primer pairs (forward primer pro341F (5′-CCTACGGGNBGCASCAG-3′) and reverse primer pro805R (5′-GACTACNVGGGTATCTAATCC-3′) with index, heterogeneity spacer and Illumina sequencing linkers (Fadrosh et al. 2014). An AMPure XP Kit (Beckman Coulter, Brea, CA, USA) was used to purify the 16S amplicon library and it was sequenced with the Illumina Miseq platform (2 × 300 bp paired-end). A minimum Phred score of 25 across the length of 200 nucleotides of the forward read was used for downstream analysis.

The raw DNA sequences were demultiplexed using Cutadapt (Martin 2011), followed by analysis using quantitative insights into microbial ecology 2 (QIIME2) for microbiota analysis (Bolyen et al. 2019). Dada2 (Callahan et al. 2016) plugin with all recommended parameters was used for quality filtering, denoising, and chimera removal. SILVA v 138.1 database (Quast et al. 2013) was used as reference data to assign taxonomy. The data were analysed at amplicon sequence variant (ASV) and genus levels. The interpretation and analysis of the data were completed with a minimum of 3000 rarefied sequences per sample. For further downstream analysis and visualisation, R packages Phyloseq (McMurdie and Holmes 2013), Phylosmith (Smith 2019), Vegan (Dixon 2003) and Microeco (Liu et al. 2021) were used. Permutation multivariate ANOVA (PERMANOVA) and a paired multivariate ANOVA (Paired MANOVA) were completed using Primer-e v7 (Anderson et al. 2008).

Ethics approval

The Animal Ethics Committee of Central Queensland University approved the study under Approval number 0000023290. All animal manipulations were performed according to the Australian Code for the Care and Use of Animals for Scientific Purposes and reported in proportion to guidelines and regulations of Animal Research: Reporting of In Vivo Experiments (ARRIVE).

Performance of animals

No significant improvements in weight gain were observed in the pigs that received phytogen (Phy) compared with the control group, and the mortality rate was similar between the two groups (Table 1, Table S1, Fig. 1).

Fig. 1.

Bar chart showing weight distribution between treatment groups at Weeks 8 and 17.

Overall microbiota composition

The microbiota composition at the phylum level in various gut regions, such as caecal, colonic, jejunal, duodenal, and ileal, and in rectal swab (faecal) is shown in the stacked bar chart (Fig. 2a). At phylum level, the most abundant phyla discovered in both groups were Firmicutes, Bacteroidota, Fusobacteriota, and Proteobacteria, with lesser amounts of Actinobacteriota, Spirochaetoata and Campilobacterota. Fig. 2b presents the most abundant genera in the gut for the two groups observed in the duodenum, jejunum, ileum, caecum and colon, and in rectal swabs (faecal). In the caecum of Phy-treated animals, the genera Clostridium sensu stricto 1 and Escherichia–Shigella showed a noticeable reduction. In the colon, the genus Romboutsia was enriched in the Phy-treated group compared with the control, while the majority of rectal-swab samples did not show apparent variation in the microbial population.

Fig. 2.

Relative abundance stacked bar chart at phylum (a) and genus level (top 15) (b).

Fig. 3 displays stacked bar charts of the top 20 genera in rectal swabs, sampled at Weeks 8, 14, and 17. The most common genera found were Escherichia–Shigella, Prevotella, Lactobacillus, Streptococcus, Acinetobacter, Muribaculaceae, Subdoligranulum, Terrisporobacter, Faecalibacterium, UCG-005, Roseburia, Blautia, Clostridium sensu stricto 1, Prevotellaceae NK3B31 group, Corynebacterium, UCG-002, Coprococcus, Campylobacter, Rikenellaceae RC9 gut group, and Staphylococcus. At Week 14, there was a significant shift in the microbial community compared with Weeks 8 and 17, with a decreased presence of Escherichia–Shigella and an increased presence of Prevotella in the Phy-treated pigs.

Fig. 3.

Relative abundance at the genus level, showing the top 20 genera across different time frames (Week 8, W08; Week 14, W14; Week 17, W17) between control (Ctr) and treatment (Phy).

The relative abundance bar chart was generated as in Supplementary Fig. S1 to demonstrate the microbial composition of individual pigs at Week 17. The analysis of rectal swabs from each group displayed that the microbial populations varied among the individual animals.

Alpha and beta diversity

To determine whether the treatment significantly affected the gut microbiota, we used both weighted and unweighted UniFrac distances at ASV level to conduct a PERMANOVA for the main variables and a paired MANOVA for paired comparisons by using Primer-e v7 (Table 2). The analysis used the following three factors: the age of the pigs (measured in weeks), the type of treatment they received (either a control group or the Phy treatment), and the origin of the sample (rectum, colon, caecum, jejunum, ileum, or duodenum). The paired MANOVA and PERMANOVA were performed to assess the impact of these factors on the gut microbiome.

The results of the PERMANOVA indicated that both age and origin had a significant (P < 0.001) impact on the gut microbiota, as determined by both weighted and unweighted UniFrac distances (Table S2). The treatment variable was found to significantly influence the microbiota when analysed with the unweighted UniFrac distance (P < 0.01) and Bray–Curtis distance at the genus level (P < 0.05).

The paired-comparison analysis using both weighted and unweighted UniFrac distances between treatment and age showed significant (P < 0.001 and P < 0.01 respectively) differences at Week 14 (Table S3). Furthermore, treatment and origin paired comparisons as shown in Table S4 determined that the differences between the control and Phy groups were most pronounced in the rectal samples when analysed using the unweighted UniFrac distance matrix (P < 0.01).

Microbial community composition in faecal samples was examined using the UniFrac distance metric to assess differences at different ages (Table 2). At Weeks 8 and 17, only the unweighted UniFrac distances showed significant differences, suggesting variations in the presence or absence of specific bacterial species between the two groups. However, weighted distances did not show significant differences at these time points. In contrast, at Week 14, both weighted and unweighted UniFrac distances exhibited significant (P < 0.001) differences, indicating distinct microbial community composition between the control and treatment groups.

The graph that displays the differences in the microbial community structure of different gut regions was created by using the principal coordinate analysis (PCoA) based on both the weighted and unweighted UniFrac distances (Fig. 4). The results from the PCoA plot demonstrated that microbiota from rectal swabs were most distinct relative to other gut regions, such as caecum (Cec), colon (Col), duodenum (Duo), ileum (Ile), and jejunum (Jej). Among intestinal regions, microbiota in the large intestine (caecum and colon) were relatively different from those of the small intestine (duodenum, jejunum and ileum), because samples from large intestine clustered together, whereas samples from the small intestine clustered together away from the large intestine.

Fig. 4.

PCoA plot created using the weighted and unweighted UniFrac distance matrices.

The alpha diversity of the microbiota in the rectal-swab samples was evaluated to compare the diversity of microbes in the Ctr and Phy groups during the study period. Shannon and Simpson indices showed significant differences between the Ctr and Phy groups in Weeks 8 and 17. The analysis showed that the microbial community in the Phy-treated pigs was more diverse, as demonstrated by the higher values of the Simpson and Shannon entropy index, than that in the untreated group, as shown in Fig. 5a. However, there was no difference in the diversity in Week 14.

Fig. 5.

(a) Characterisation of alpha diversity and (b) beta diversity in cloacal swabs before and after treating with Phy in pigs. Mann–Whitney test was used to calculate the significance. (*P < 0.05; **P < 0.01; ns, not significant).

To illustrate the alterations in the composition of the microbial community in the cloacal swabs, a principal-components analysis (PCA) was performed utilising Primer-7e, as presented in Fig. 5b. The Bray–Curtis dissimilarity matrix was employed to quantify the difference between the groups. The loading vectors indicated the direction and magnitude of the contribution of some important variables (microbes) to the variance explained by each principal component.

In Week 8, both the Phy-treated and the untreated groups had a similar distribution, with the highest contribution of Escherichia–Shigella to the variance captured by the principal components in one direction and Acinetobacter in another direction. Other important genera contributing to variance are Psychrobacter, Prevotella, Lactobacillus and Streptococcus. However, at Week 17, Escherichia–Shigella had the highest contribution to the variance in one direction and Prevotella in another direction. Acenetobacter, Muribaculaceae and UCG-005 are other major taxa contributing to the variance defined by the principal components.

Taxonomic alterations

The linear discriminant analysis effect size (LEfSe) analysis, a widely used biomarker discovery tool for genomic studies, was utilised to assess the statistical significance of differences among the microbial communities at Week 17 (Fig. 6). The threshold for linear discriminant analysis (LDA) score was set at 3.3. The LDA score higher than 2.0 is considered to be biologically relevant, indicating that particular feature is significantly enriched in one group compared with the other. Therefore, the higher the LDA score is, the higher is the contribution of specific taxa to the differences between groups (Ctr and Phy). The control group was found to have a higher abundance of the genera Escherichia–Shigella, Fusobacterium, Muribaculaceae, Romboutsia, Acinetobacter, Christensenellaceae R-7 group, and Streptococcus, while the Phy-treated group was enriched with the genera Campylobacter, Clostridium sensu stricto 1, Prevotella, Roseburia, Terrisporobacter, Prevotellaceae NK3B31 group, Lactobacillus, and Alloprevotella.

Fig. 6.

LEfSe at the genus level from all tissue sites, LDA score of >3.3.