Lactobacillus, an essential member of the normal microbiota, can improve the intestinal microbial balance with beneficial effects when administered in adequate amounts (Bogovic et al., 2016; Simpson et al., 2018). Lactobacillus has been widely employed as a supplement in foods as well as in farming and medicine (Hill et al., 2014). In fact, the Lactobacillus market has maintained a growth rate of 15–18% in recent years in China¹. There is increased acceptance of the application of Lactobacillus strains in animal feeds to maintain health, and animal studies have shown the beneficial probiotic effects. However, the mechanism of action of Lactobacillus remains to be elucidated.

During the weaning period, piglets are subjected to sudden dietary, social, and environmental changes, and the microbiota are sharply changed, generally resulting in a decrease in bacteria of the Lactobacillus group and an increase in pathogenic bacteria, mainly leading to colibacillosis resulting in diarrhea, poor appetite, lower feed intake, and growth retardation (Isaacson and Kim, 2012; Schokker et al., 2015). Previous studies on the microbiota in weaning piglets suggested that there was a large reduction in alpha diversity during the weaning period, and that this was reduced further after treatment with antibiotics (Pie et al., 2004; Tao et al., 2015; Dou et al., 2017). Therefore, weaning has been associated with a “disrupted state” for the microbiota that is referred to as “dysbiosis,” and weaning stress in piglets is always the major cause of economic loss for pig farmers (Wilson et al., 1989; Lalles et al., 2007).

Metagenomics and related methods have led to significant advances in our understanding of the gut microbiome. Supplementation with Lactobacillus modulates the gut microbial profile and thereby affects microbial metabolite production, which results in improved gut health (Chiang et al., 2015; Dowarah et al., 2017a). Researchers have also reported that Lactobacillus strains could be effective at enhancing intestinal barrier function, which restricts the colonization of pathogenic bacteria to the intestinal mucosa, as evidenced by better absorption of nutrients, enabling better sustainability of the piglets (Abrahao et al., 2004).

Previously, Lactobacillus reuteri ZLR003 was isolated from the caecum mucosa of a weaned piglet. Our previous work showed that this strain affected the gut microbiota composition in the jejunum, colon, and cecum of weaned piglets (Zhang et al., 2016). In addition, the results of Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis by RNA-seq in the jejunum tissue showed that the expression of genes related to arachidonic acid metabolism and linoleic acid metabolism, glutathione metabolism, glycine, and serine and threonine metabolism were improved after treatment with L. reuteri ZLR003 compared with the control group (Zhang D.Y. et al., 2017). We hypothesized that a shift in microbiota might be associated with a shift in microbiota products and blood metabolites, although the possible link was still unclear. Thus, in our present study, we analyzed changes in the fecal microbiota composition during the nursed and weaned periods following supplementation with L. reuteri ZLR003 and evaluated the underlying association with feces short chain fatty acids (SCFAs), serum long chain fatty acids (LCFAs), and serum free amino acids (FAAs) affected by the microbiota in weaned piglets.

Piglets are thought to be bacteria-free prior to birth. After birth, the flexible distribution of intestinal microbes leads to differences in microbiota composition, which is influenced by factors such as the porcine species, age, feed, and husbandry (Zhao et al., 2015). Reports have suggested that dietary supplementation with Lactobacillus helps in the early development of a stable gut microbiota in neonatal piglets and reduces the number of potential entero-pathogens, such as Escherichia coli and Clostridia, in neonatal weaned piglets by improving intestinal health and preventing diarrhea (Liu et al., 2014; Dowarah et al., 2017b).

High-throughput next-generation sequencing of 16S rRNA genes has provided abundant information on microbiota composition and has contributed to our understanding of the function of the intestinal microbiota in animal health. Previous research showed that Firmicutes and Bacteroidetes are the most abundant phyla in the fecal microbiota of pigs (Looft et al., 2014). The results of our present study similarly show that Firmicutes and Bacteroidetes were the dominant phyla in the feces of both treatment groups of nursed and weaned piglets. However, the total proportion of Firmicutes and Bacteroidetes was increased from 86 to 94% from the nursed to the weaned period, with a higher proportion of the Proteobacteria phylum (about 7%) observed during the nursed period. The proportion of Firmicutes present at the end of the nursed period in our study is consistent with the results published by Zhao et al. (2015). Who reported the presence of 73.0% of Firmicutes among 30-day-old piglets. The abundance of the Firmicutes phylum was decreased while that of the Bacteroidetes phylum was increased for weaned piglets compared with nursed piglets in both the control and L. reuteri ZLR003 groups. At the genus level, Prevotella was clearly increased from 7 to 31% between the nursed and weaned periods for piglets. Arumugam et al. (2011) reported that the Bacteroidetes phylum is important for the degradation of carbohydrates. Furthermore, Prevotella spp. has been reported to play a role in carbohydrate metabolism and was found to dominate the swine fecal metagenome (Surana and Kasper, 2017). Thus, we hypothesize that the changes in the abundance of the Bacteroidetes phylum and Prevotella genus are related to changes in the diet of piglets from the nursed to weaned period.

Our data reveal that the microbiota composition profile in weaned piglets supplemented with L. reuteri ZLR003 was distinct from that of weaned control piglets. Specifically, at the genus level, Selenomonas, Mitsuokella, Megasphaera, and Lactobacillus, all of which belong to the Firmicutes phylum, were more abundant in the L. reuteri ZLR003 group compared with the control group. We analyzed carbohydrate fermentation in accordance with Bergey’s ManuaCl of Determinative Bacteriology (2004) and found that Selenomonas (acetate and propionate), Mitsuokella (acetate), Megasphaera (lactate), and Lactobacillus (lactate) were able to produce SCFAs, including lactate, acetate, and propionate. Furthermore, our results indicate that L. reuteri ZLR003 dietary supplementation affected the SCFAs content of piglets, based on comparison with the control group. However, there was no significant difference in the abundance of Lachnospiraceae between the L. reuteri and control groups. Furthermore, the abundance of the Spirochaetae phylum was significantly lower in the L. reuteri ZLR003 group compared with the control group, as was the abundance of the Treponema genus. Our previous study suggested that, at 40 days post-partum, the genus Treponema was significantly less abundant in the colon and cecum of weaned piglets from the L. reuteri group than in those from the chlortetracycline and control groups (Zhang et al., 2016). This is also consistent with the findings of Riboulet-Bisson et al. (2012), who showed that a Lactobacillus salivarious strain had the ability to modulate the level of Treponema.

It was previously proposed that the major symbiotic function of the gut microbiota is to provide SCFAs by fermenting carbohydrates in the intestinal epithelium, and SCFAs play an important role as an energy source for enterocytes in terms of animal health (Fouhse et al., 2016). Brestensky et al. (2017) reported that, for pigs, the lactate content in the cecal digesta and feces is lower than that in the jejune digesta, whereas acetate is present in higher proportions in the cecal digesta and feces. Liu et al. (2014) found that probiotics could accelerate the decomposition of carbohydrates, influencing SCFAs synthesis and helping to increase the butyric acid concentration in the colon. In the present study, the lactic acid, butyric acid, and the total SCFAs concentrations were all significantly higher whereas the acetic acid level was lower in the L. reuteri ZLR003 group compared with the control group. Previous work revealed that acetate provides an important substrate for butyrate production, and about 50% of butyric acid produced by bacteria in the intestine, including Rosebyria spp. and Faecalibacterium prausnitzii, is associated with the net consumption of acetate (Duncan et al., 2004). Thus, our findings may suggest that dietary supplementation with L. reuteri ZLR003 enhanced the production of butyric acid from acetate by the intestinal microbiota.

To date, many studies have used high-throughput sequencing methods to explore the gut microbiota composition of animals. An increasing number of studies have reported the contribution of SCFAs production by the microbiota. Fernandes et al. (2014) reported that SCFAs are absorbed in the colon and that the SCFA content in the feces is lower compared with that in the cecal digesta. Zhang Y.J. et al. (2017) showed that Lactobacillus and Megasphaera are affected by the fiber source. Louis et al. (2014) reported that Megasphaera elsdenii can produce several SCFAs, such as acetate, propionate, butyrate, and valerate. Here, Lactobacillus and Megasphaera showed positive correlations with butyrate, but there were no significant differences in their abundances between piglets treated by L. reuteri and control animals (P > 0.05). Additionally, Flint et al. (2008) suggested that the Ruminococcaceae genus can produce SCFAs by fermenting dietary polysaccharides in the intestinal tract. However, in the present study, the microbiota results suggest that the abundances of fecal Ruminococcaceae_UCG-014 and Ruminococcaceae_UCG-010 were significantly lower in the L. reuteri group compared with the control group and that Ruminococcaceae_UCG-010 and Ruminococcaceae_UCG-014 both showed a negative correlation with the acetate and butyrate levels. Thus, it is possible that there is a difference in Ruminococcaceae abundance between the intestinal tract and feces, but these differences need to be analyzed in future work.

Here, the L. reuteri group showed a higher PUFAs content compared with the control group; this finding is consistent with work by Ross et al. (2012) and Chang et al. (2018). Moreover, an analysis of the LCFAs profiles of piglets also revealed that the PUFAs C18:2n6c, C18:3n3, C20:4n6, and C22:6n3 were significantly higher in the L. reuteri ZLR003 group compared with the control group. This is consistent with our finding from a previous study that the PUFAs C18:2n6c, C18:3n3, and C20:4n6 in the colon were significantly higher in the L. reuteri group compared with the control group (Zhang D.Y. et al., 2017). Other research has shown that dietary fatty acid composition has an influence on its deposition in tissue, and the PUFAs: SFAs ratio is commonly used to evaluate the nutritional and health value of meat for human consumption (Aldai et al., 2007). Therefore, we hypothesize that Lactobacillus may promote the PUFA content in pork during pig production.

Huang et al. (2013) suggested that a diet rich in polyunsaturated omega-6 fatty acids causes an alteration in the composition of the intestinal microbiota compared with diets rich in saturated fat. Mayorga-Reyes et al. (2016) also suggested that a diet rich in unsaturated fatty acids promotes an abundance of beneficial bacteria such as Bacteroidetes. Prevotella, a dominant genus of Bacteroidetes, has been reported to be a polysaccharide-degrading and oligosaccharide-utilizing bacterial genus, associated with the ability to degrade mucin and plant-based carbohydrates (Lamendella et al., 2011; Patel et al., 2014). Previous research showed that Prevotella is one of the abundant genera in pigs after weaning and that it represents up to 30% of all classifiable bacteria at 10 weeks of age while accounting for only 3.5–4.0% of the bacteria by the time pigs reach 22 weeks old (Isaacson and Kim, 2012; Pajarillo et al., 2014). Furthermore, the administration of Lactobacillus strains was found to significantly increase the abundance of the Prevotella genus (Kaluzna-Czaplinska et al., 2017). Our present study found that the genus Prevotella, especially Prevotella_9, was more abundant in the L. reuteri group compared with the control group. Moreover, our results also demonstrate that Prevotella_9 showed a stronger positive correlation with C18:2n6c compared with the correlation between the Prevotella genus and the SCFAs content.

The serum concentration of FAAs is affected by their intestinal absorption, transport rates across tissue membranes, and cellular metabolism, and this concentration is an important modulator of host physiology (Bertolo et al., 2000). A previous study focused on frail older patients suggested that altered amino acid metabolism is correlated with body composition and nutritional status; it also reported that body mass index was reduced in the severely frail group and that it was significantly correlated with plasma Val levels (Adachi et al., 2018). Essen-Gustavsson et al. (2010) evaluated horses during short and intense exercise (mostly anaerobic) on a high-speed treadmill and reported that there was an interaction between exercise and dietary crude protein in terms of the Lys and Val plasma concentrations. Here, we analyzed the effect of dietary supplementation with the L. reuteri ZLR003 strain on the serum FAA profiles of weaned piglets, and our results show that FAAs, including Gly, Ala, Val, Iso, Asp, Asn, Glu, Met, Phe, and Leu, were significantly abundant in the L. reuteri group compared with the control group. Chang et al. (2018) reported that muscles from the probiotic-treated group had higher levels of Gly, Ala, Pro, Val, Leu, and Iso. However, intervention studies of amino acids and their relationship with probiotic strains in pig production are still needed.

Lastly, our present study found that the genera Treponema_2, Lachnospiraceae_XPB1014_group, Clostridium_sensu_stricto_1, and Mollicutes_RF9_norank were negatively correlated with Pro, Asp, Phe, Arg, Met, Tyr, Iso, His, and Leu and that Streptococcus was negatively correlated with Phe, Tyr, Met, and Leu, whereas the genera Megasphaera and Mitsuokella were positively correlated with Pro, Phe, Met, Tyr, Iso, and Leu. The microbiota results also show that dietary L. reuteri supplementation significantly increased the abundance of Megasphaera and Mitsuokella and decreased the abundance of Streptococcus. Mitsuokella genera, such as Mitsuokella jalaludinii, which are considered novel phytase-producing rumen bacteria, were reported to promote the digestion of amino acids (Lan et al., 2011). M. elsdenii has been reported as a major inhabitant of the pig intestine, and this bacterium converts lactate to various kinds of SCFAs (Kim et al., 2011), although its relationship with amino acid fermentation is still unclear. Anhe et al. (2015) reported that the main SCFAs produced by the microbiota, such as acetate, propionate, and butyrate, are derived from both carbohydrates and amino acids. Therefore, further investigations are needed to clarify the mechanism of action and obtain a more detailed understanding of the role of gut microbiota composition in nutrient metabolism in pig production.

In conclusion, the results of this study indicate that dietary supplementation with L. reuteri ZLR003 in piglets influenced the fecal microbiota composition, which was associated with the metabolism of SCFAs in the feces and the metabolism of LCFAs and FAAs in the serum. This is the first study to analyze the microbiota associated with host metabolism in weaned piglets after the administration of oral Lactobacillus. Although further studies are needed, our findings will facilitate the application of Lactobacillus strains for nutritional and health benefits in pig production.

The study was carried out in accordance with the recommendations of guidelines set by the Animal Care and Use Committee (permit number: SYXK-2017-0005) of the Institute of Animal Husbandry and Veterinary Medicine, Beijing Academy of Agriculture and Forestry Sciences (IAHVM-BAAFS), Beijing, China. The protocols were approved by the Animal Care and Use Committee of IAHVM-BAAFS.

DZ and HJ designed the experiments and carried all the analyses, interpreted the results, and drafted the manuscript. HL contributed to data collection during the animal experiments. SW analyzed the microbiota composition data. WZ measured and analyzed the serum amino acids contents. JW, HT, and YW collected the samples during the animal experiments. All co-authors provided their comments, have read, and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.