In vivo Trial of Bifidobacterium longum Revealed the Complex Network Correlations Between Gut Microbiota and Health Promotional Effects

Abstract

Complete genome sequence analysis of Bifidobacterium longum subsp. longum BCBL-583 isolated from a Korean female fecal sample showed no virulence factor or antibiotic resistance gene, suggesting human safety. In addition, this strain has oxygen and heat tolerance genes for food processing, and cholesterol reduction and mucin adhesion-related genes were also found. For in vivo evaluations, a high fat diet (HFD) mouse model was used, showing that BCBL-583 administration to the model (HFD-583) reduced the total cholesterol and LDL-cholesterol in the blood and decreased pro-inflammatory cytokines but increased anti-inflammatory cytokines, substantiating its cholesterol reduction and anti-inflammation activities. Subsequent microbiome analysis of the fecal samples from the HFD mouse model revealed that BCBL-583 administration changed the composition of gut microbiota. After 9 weeks feeding of bifidobacteria, Firmicutes, Actinobacteria, and Bacteroidetes increased, but Proteobacteria maintained in the HFD mouse models. Further comparative species-level compositional analysis revealed the inhibitions of cholesterol reduction-related Eubacterium coprostanoligenes and obesity-related Lactococcus by the supplementation of B. longum BCBL-583, suggesting its possible cholesterol reduction and anti-obesity activities. The correlation analysis of HFD-583 between the gut microbiota compositional change and cholesterol/immune response showed that Verrucomicrobia, Firmicutes, Actinobacteria, and Bacteroidetes may play an important role in cholesterol reduction and anti-inflammation. However, correlation analysis of Proteobacteria showed the reverse correlation in HFD-583. Interestingly, the correlation analysis of B. longum ATCC 15707 administration to HFD model showed similar patterns of cholesterol but different in immune response patterns. Therefore, this correlation analysis suggests that the microbial composition and inflammatory cytokine/total-cholesterol may be closely related in the administration of BCBL-583 in the HFD mice group. Consequently, BCBL-583 could be a good probiotic strain for gut health promotion through gut microbiota modulation.

Keywords: Bifidobacterium longum; anti-inflammation; cholesterol reduction; gut microbiota; obesity.

Figure 1

Experimental procedure for high fat diet (HFD) model and administration of bifidobacteria. Mice were given a single week for normal diet (ND) and then randomly divided into four groups. Each bacterial sample was administered orally in the mice (10⁸ CFU/kg/day).

Figure 2

Circular genome map of Bifidobacterium longum BCBL-583. The outer circle indicates the location of all annotated open reading frames (ORFs) in double strands colored by clusters of orthologous groups (COG), and the inner circle with red peaks indicates GC content. Between these circles, sky blue arrows indicate the rRNA operons, and orange arrows indicate the tRNAs. Specific genes or gene clusters were colored according to their functions in upper legend. The colors of COG categories were indicated in lower legend.

Figure 3

Transcription analysis of bsh gene to bile acid and cholesterol adsorption assay of BCBL-583. (A) The bsh gene expression levels were quantified in the different concentration of bile acid by qRT-PCR. (B) To clarify the cholesterol adsorption ability of BCBL-583, total cholesterol amounts were determined in supernatant, cell extract, and cell membrane.

Figure 4

Cholesterol levels in serum samples of four groups after 9 weeks. (A) Total cholesterol level, (B) LDL cholesterol level, and (C) HDL cholesterol level in serum. Differential letters indicate statistically relevant differences among the different groups (p < 0.05).

Figure 5

Inflammatory cytokine levels from spleen. (A) Pro-inflammatory cytokines: tumor necrosis factor (TNF)-α and interleukin (IL)-6, (B) anti-inflammatory cytokines: IL-10 and IL-4 levels from spleen. Differential letters indicate statistically relevant differences among the different groups (p < 0.05).

Figure 6

Principal coordinates analysis (PCoA) and microbial compositional changes of representative phyla and genera of four groups in Week 1, Week 5, and Week 9. (A) β-Diversity plot using UniFrac distance with four groups. (B–F) The major phyla and genera were indicated on the box plot.

Figure 7

Spearman’s correlation analysis between selected phyla/genera and cholesterol/immune response results of ATCC 15707 and BCBL-583 in HFD mouse models after 9 weeks. Cholesterol and inflammatory cytokines are correlated with selected phyla and genera bacteria.

Figure 8

Summary on the probiotic effect of B. longum BCBL-583 in HFD mouse model.