The consumption of dietary fiber plays an essential role in maintaining homeostasis of the gut ecosystem. Bacterial metabolites and prebiotics from dietary fiber are involved in various physiological processes for modulating health in humans and animals (Desai et al., 2016; Martens, 2016). Low fiber intake often results in chronic diseases related to nutrient metabolism and gut inflammatory disorders, such as obesity, diabetes, and inflammatory bowel diseases (Han et al., 2017; Ma et al., 2017). Thus, there is an increasing interest focused on the link between different fibrous ingredients and gut health. High fiber diets improved intestinal morphology, mucin secretion, and growth of beneficial bacteria (Hedemann et al., 2009; Chen H. et al., 2015; Tian et al., 2017). This has been attributed to the role of short-chain fatty acids (SCFAs) in the gut health and they are end-products of dietary fiber fermented by gut microbiota (Chen J. et al., 2015). The major SCFAs are acetate, propionate, and butyrate, they serve as energy substrates for the colonic epithelium and modulate the immune system (Fan et al., 2015; Koh et al., 2016). The effects of high fiber on growth performance in pigs have been studied. Some reports indicated that fiber-rich diets increased weight gain (Bertram et al., 2009; Gerritsen et al., 2012), while other researchers showed reduced growth or no influence on growth performance accompanied by reduced digestibility of nutrients and energy (Jaworski et al., 2017; Morowitz et al., 2017).

The milling of corn for food constituents results in a large number of co-products. Corn bran (CB) is a common fibrous corn by-product that has been widely utilized in animal feed due to its relatively lower cost. CB primarily consists of insoluble fiber with ~280 g/kg cellulose and 700 g/kg hemicellulose. This indigestible corn bran contains many bioactive components, such as corn fiber gum, cellulosic fiber gel, and xylo-oligosaccharides (Rose et al., 2010). Previous studies have shown that arabinoxylan and cellulose were the major components extracted from wheat bran, which exerted positive effects on intestinal barrier function and increased the population of Lactobacillus in piglets (Chen H. et al., 2015; Hashemi et al., 2017). In contrast, information about how CB impacts the gut bacterial community is very limited.

To harvest the production of useful bioactive components from corn barn, different processing technologies have been used. Fermentation technology has been applied for modification of biological materials into useful products and for reducing the insoluble fiber content of feed (Pallin et al., 2016). Bacillus subtilis MA139 was identified in previous study having high antimicrobial activity against Escherichia coli, Salmonella typhimurium, and Staphylococcus aureus (Yang et al., 2015). Diets that contained B. subtilis MA139 improved daily gain and feed conversion in weaned pigs, and B. subtilis was recommended as a potential alterative to in-feed antibiotics (Horita et al., 2015). B. subtilis can also produce various carbohydrase enzymes for polysaccharide degradation (Rhee et al., 2014). B. subtilis MA139 co-inoculated with lactic acid bacteria were applied with solid-state fermentation for the production of fermented wheat bran and soybean bran feedstuffs (Vijayaraghavan et al., 2016). However, the effect of dietary CB fermented by B. subtilis in pigs has not been reported. It is important to understand how complex polysaccharides influence gut microbiota in humans and monogastric animals. Therefore, the objective of the present study was to compare the effects of diets that contained CB or fermented CB (FCB) on growth performance, nutrient digestibility, and gut microbiota composition in finishing pigs.

This study was performed to compare the effect of CB and FCB inclusion on growth performance, nutrient digestibility, and gut microbiota composition in finishing pigs. Dietary fiber is involved in various physiological processes and plays an essential role in enhancing immune regulation and gut health (Koh et al., 2016). Corn bran is the main by-products from the corn milling process, and is commonly utilized as a feed ingredient for animals. The majority of plant polysaccharides like CB cannot be digested directly by enzymes in humans and monogastric animals, but they can be degraded by colonic bacterial communities (Sonnenburg and Sonnenburg, 2014).

Polysaccharide utilization and fiber digestibility can be improved with special processing technology. Solid-state fermentation of feedstuffs has been applied for the purpose of modifying biological materials to reduce the insoluble fiber content of feeds resulting in useful feedstuffs (Pallin et al., 2016), because the fibrolytic enzymes produced from bacteria, such as cellulases and hemicellulases, are able to degrade complex polysaccharides (Kracher et al., 2016). In this study, the contents of cellulose and hemicellulose were reduced in the FCB. Meanwhile, the level of soluble dietary fiber increased after in vitro fermentation compared with CB. Chemical analysis demonstrated that the solid-state fermentation makes the insoluble glycans in corn bran easily digestible. In addition, 3 monosaccharide residues (arabinose, xylose, and glucose) were increased, and these should be closely associated with the reaction of fermentation by the mixed cultures of Bacillus subtilis MA139, E. faecium, and S. cerevisae. It has been reported that arabinose and xylose are major sugars in the plant cell wall of wheat bran and CB (Rose et al., 2010; Chen H. et al., 2015); B. subtilis produce xylanases, which are able to degrade soluble and insoluble xylan substrates to acidic xylo-oligosaccharides (Rhee et al., 2014). The strain of B. subtilis MA139 used in this study was identified and has been used as a potential probiotic to improve pig performance (Yang et al., 2015). Further, we found an innovative strategy to apply B. subtilis MA139 with lactic acid bacteria to produce solid-state fermentation of feedstuffs. These co-inoculating strains showed high antimicrobial activity against on Enterobacteriaceae, such as E. coli and S. typhimurium. Lactic acid bacteria, such as Lactobacillus reuteri and Lactobacillus fermentium, have been applied for feed fermentation due to their beneficial effects on preservation of feeds, favorable aroma, and flavor of bacterial end-products (Pallin et al., 2016). In the present study, E. faecium was applied in solid-state fermentation in CB because it can synthesize enzymes to degrade starch and the bacterial metabolite of lactic acid is beneficial for modulating immune response of the host (Siepert et al., 2014; Starke et al., 2015).

Dietary fiber plays an important role in maintaining bacterial diversity and intestinal homeostasis (Chen J. et al., 2015). It is reported that high concentrations of dietary fiber result in body weight loss (Kovatcheva-Datchary et al., 2015), which may be attributed to an improvement in satiety due to fiber viscosity. Fiber-rich diets may affect satiety through an increase in mastication or changes in gut hormones levels, such as glucagon-like peptide 1 (Näslund and Hellström, 2007). A previous study showed CB resulted in a lesser desire for food intake in humans (Rose et al., 2010). In the present study, the pigs fed CB or FCB treatments did not show significant differences in daily feed intake and average daily gain and digestibility of organic matter decreased in both CB and FCB treatments compared to CON group. This observation was in agreement with the previous studies in pigs (Bindelle et al., 2008; Bertram et al., 2009). The result suggests that diets containing 10% CB or FCB did not affect satiety response in finishing pigs over 21 d study.

The gut bacterial community and the abundance of bacterial metabolites are influenced by diets, in turn, the gut microbiota composition impacts nutrient digestibility (Fan et al., 2015, 2017a). Dietary fiber is characterized by different monosaccharides within complex polymers of ß-1,4-linked units of hexose (cellobiose and glucose) and/or pentose (arabinose and xylose) sugars (Martin et al., 1998), which cannot be digested in the small intestine of monogastric animals due to a lack of cellulolytic enzymes. However, the most dense and diverse bacterial community inhabits the large intestine and can produce a range of cellulolytic enzymes to degrade different fiber component in diet. Consequently, the activity of bacterial physiology relates to the fiber metabolism in the gut. The degradation of complex carbohydrates requires a range of bacterial enzymatic activities, normally accomplished by a consortium of bacteria rather than a single species (Scott et al., 2008).

There is a considerable degree in variation of the intestinal microbiota composition between studies, but Firmicutes and Bacteroidetes were the most dominant phyla in healthy individuals (Flint et al., 2007). In our study, the Firmicutes and Bacteriodetes were numerically dominant in finishing pigs with the abundance higher than 94%, and the result is consistent with fecal microbiota from pigs reported in other studies observed (Heinritz et al., 2016). It has been reported that fiber-rich diets result in increased abundance of Bacteroidetes, especially the genera of Prevotella and Xylanbacter (De Filippo et al., 2010; Trompette et al., 2014). For instance, B. thetaiotaomicron is regarded as a glycan-degrading generalist, which can degrade complex glycan of plant wall, animal tissue, and host mucin (Koropatkin et al., 2012). The change in the ratio of Firmicutes and Bacteroidetes in FCB group might be due to the glycan composition in FCB. Addition to Firmicutes and Bacteroidetes phylum, there were 13 other phylum detected among the 3 dietary treatments, including Spirochaetae, Proteobacteria, Tenericutes, Lentisphaerae, Actinobacteria, SHA-109, Cyanobacteria, Fibrobacteres, Saccharibacteria, unclassified_k_norank, Synergistetes, Elusimicrobia, and Chlamydiae. Compared to a previous study on gut microbiota composition in different age of pigs (Niu et al., 2015), wherein the SHA-109, Saccharibacteria, unclassified_k_norank, and Elusimicrobia were unique phylum, and this might be associated with different experiment diets between the studies. Specifically, Fibrobacteres phylum was identified by gene sequencing, albeit in a small proportion. The genus Fibrobacter belongs to the poorly defined Fibrobacteres phylum, which was regarded as Spirochetes or Bacteroidetes. In fact, the Fibrobacteres phylum is characterized by high cellulolytic activity and capable of degrading refractory plant structural polysaccharides and they play a prominent function in fiber-digesting and energy metabolism in gastrointestinal tract (Qi et al., 2005). In the rumen, the genus Fibrobacter is responsible for the degradation of dietary fiber and polysaccharides. The increased abundance of Fibrobacter resulted in more sugars and SCFAs in gastrointestinal tract of dairy calves (Deng et al., 2017). Moreover, Fibrobacteres was observed in different ages of pig; however, diet composition and fiber components were not mentioned (Niu et al., 2015). The mechanistic link between Fibrobacteres and polysaccharide metabolism needs further study.

Herein, the relative abundance of the top 30 bacterial communities at the genus level was shown in a heat map. Clostridium has been associated with dietary fiber metabolism previously (Niu et al., 2015). Lachnospiraceae and Rumininococcaceae are two dominant bacteria from Clostridia among the 3 dietary treatments and are regarded as fibrolytic specialists of commensal bacteria to degrade the complex plant material involved in activating carbohydrate enzymes, sugar transport, and metabolic pathways. Therefore, Lachnospiraceae and Rumininococcaceae are associated with the maintenance of gut health (Biddle et al., 2013). In this study, the CB group showed the greatest abundance of genus Lachnospiraceae_NK4A136_group and Lachnospiraceae_XPB1014_group. This was attributed to the dietary inclusion of CB which can stimulate fibrolytic bacteria to degrade the complex plant bran of recalcitrant substrate (Brulc et al., 2009). In contrast, non-adherent Bacteriodes sp. and Bifidobacterium sp. ferment easily hydrolysable starch (Cameron et al., 2014). Pigs fed CB treatment showed distinctly more bacteria in the Lachnospiraceae and Rumininococcaceae families of the Firmicutes phylum, while Prevotellaceae of the Bacteroidetes phylum was more abundant in the FCB group. The relative abundance of Lactobacillus and Anaerovibrio were higher in both CB and FCB groups compared to CON group. These genera are able to utilize glycerol and have been associated with propionate and butyrate production (Minuti et al., 2015). The result was supported by a previous study where insoluble dietary fiber extracted from CB stimulated the growth of beneficial bacteria in the gut of health individuals (Hsu et al., 2004). The Prevotellaceae family within Bacteroidetes phylum are involved in glucose metabolism (Heinritz et al., 2016) and the relative abundance of norank_f_Prevotellaceae was significantly higher in FCB group compared to the other 2 groups, consistent with the increased levels of soluble dietary fiber and glucose in FCB. Rikenellaceae are commensal bacteria in the gut in relatively small proportion, which can thrive on high-fat diets and are enriched in gut microbiota of the obese. A study reported that the abundance of Rikenellaceae_RC9_gut_group increased with high concentration of dietary protein in pigs (Fan et al., 2017a,b). In our study, the proportion of Rikenellaceae_RC9_gut_group decreased in CB and FCB treatments compared to the CON group, which suggested that Rikenellaceae_RC9_gut_group have an impact on carbohydrate and lipid metabolism. The alterations in microbial populations with the inclusion of CB and FCB are consistent with the relative fermentable substrate composition of feedstuff.

The end-product of bacterial metabolism plays an important role in shaping the bacterial community. Bacterial metabolite production is dependent on the available substrates, but in general, most end-products detectable in feces are acetate, propionate, and butyrate (Huang et al., 2015). Fermentation of different carbohydrate substrates by bacterial consortia normally results in variations of SCFAs production. Many colonic bacteria produce lactate as a fermentation product, but the concentration of lactate was <3–5 mM in fecal sample (Duncan et al., 2007). Lactic acid was not detected in the majority of fecal samples in our study. Lactate-utilizing bacteria includes Eubacterium hallii and Anaerostipes caccae, which can be detected in fecal samples from healthy individuals (Duncan et al., 2004). The species of E. hallii and A. caccae were identified by 16S rRNA sequencing in the present study, but their abundance were low and not affected by dietary treatments. Butyrate as one of the important bacterial metabolites that can provide energy for colonic epithelium, and also plays key functions in protection against colonic diseases (Koh et al., 2016). Eubacterium rectale and Roseburia species are two important groups of butyrate-producing bacteria. In the present study, the abundance of Roseburia was not influenced by dietary treatments, and a population of E. rectale was not observed in fecal samples. There were few changes on gut bacteria related to SCFAs production, and this observation was consistent with the stable SCFA concentration among the treatment groups.

In conclusion, diets containing CB and FCB did not impact growth performance or nutrient digestibility in finishing pigs. Gut microbiota composition was affected slightly by dietary treatments, as the bacterial richness and diversity in FCB was decreased compared to control and CB groups. Moreover, the relative abundance of bacteria related to cellulolytic degradation was higher in CB fed pigs; whereas FCB fed pigs had greater proportions of starch-fermenting bacteria. The differences in bacterial community between CB and FCB inclusion might be due to the alteration on chemical component of polysaccharides via in vitro fermentation. Thus, the underlying mechanisms on how the chemical characterizations of corn bran components influence the bacteria related to polysaccharide and lipid metabolism still needs further investigation.

XM, and WL: Designed the experiments; PL, JinZ, PG, and ZG: Performed the experiments; LL and CW: Analyzed the data; PL: Wrote the manuscript, which was edited by CL, LJ, SQ, JieZ, and XM. XM: Resourced the project. All authors read and approved the final manuscript.