An In Vitro Enrichment Strategy for Formulating Synergistic Synbiotics

Abstract

Research on the role of diet on gut and systemic health has led to considerable interest toward identifying novel therapeutic modulators of the gut microbiome, including the use of prebiotics and probiotics. However, various host responses have often been reported among many clinical trials. This is in part due to competitive exclusion as a result of the absence of ecological niches as well as host-mediated constraints via colonization resistance. In this research, we developed a novel in vitro enrichment (IVE) method for isolating autochthonous strains that can function as synergistic synbiotics and overcome these constraints. The method relied on stepwise in vitro fecal fermentations to enrich for and isolate Bifidobacterium strains that ferment the prebiotic xylooligosaccharide (XOS). We subsequently isolated Bifidobacterium longum subsp. longum CR15 and then tested its establishment in 20 unique fecal samples with or without XOS. The strain was established in up to 18 samples but only in the presence of XOS. Our findings revealed that the IVE method is suitable for isolating potential synergistic probiotic strains that possess the genetic and biochemical ability to ferment specific prebiotic substrates. The IVE method can be used as an initial high-throughput screen for probiotic selection and isolation prior to further characterization and in vivo tests.IMPORTANCE This study describes an in vitro enrichment method to formulate synergistic synbiotics that have potential for establishing autochthonous strains across multiple individuals. The rationale for this approach-that the chance of survival of a bacterial strain is improved by providing it with its required resources-is based on classic ecological theory. From these experiments, a human-derived strain, Bifidobacterium longum subsp. longum CR15, was identified as a xylooligosaccharide (XOS) fermenter in fecal environments and displayed synergistic effects in vitro The high rate of strain establishment observed in this study provides a basis for using synergistic synbiotics to overcome the responder/nonresponder phenomenon that occurs frequently in clinical trials with probiotic and prebiotic interventions. In addition, this approach can be applied in other protocols that require enrichment of specific bacterial populations prior to strain isolation.

Keywords: bifidobacteria; prebiotic; probiotic; synbiotic; xylooligosaccharide.

FIG 1

Bifidobacteria were successfully enriched by XOS in fecal environments, whereas strain establishment was dependent on the strain and the host. (A) Hypothetical trends of successful (green) and unsuccessful (red) enrichments in fermentation experiments. (B) Enrichment of total Bifidobacterium (•) and B. adolescentis (▪) in a sample from which B. adolescentis CR11 was isolated. (C) Unsuccessful establishment of B. adolescentis (CR11) (▪) with commensurate enrichment of total Bifidobacterium (•) in a sample from which B. longum subsp. longum CR15 was isolated. (D) Establishment of B. longum (CR15) (▪) and total Bifidobacterium (•). Horizontal dashed lines indicate limits of detection (10⁴ CFU/ml).

FIG 2

Growth of B. longum subsp. longum CR15 in minimal media supplemented with different XOS fractions. (A) Optical density measurements at a wavelength of 600 nm every 4 h within the first 12 h and at 24 h using minimal media (◆; mMRS) with the addition of the following sugars: MRS containing equivalent amounts of residual sugars (■; mMRS-res), 1% glucose (▲; mMRS-glucose), 1% XOS (•; mMRS-XOS), 1% of an XOS fraction containing DPs 2, 3, and 4 (⬣; mMRS-DP2,3,4), and 1% of an XOS fraction containing DPs 4 and above (▼; mMRS-DP4). (B) TLC analysis was carried out using 7.5 μl of spent fermentation media and standards including 2.5 μl of 2% xylose and 5 μl of 2% XOS. The plates were developed twice using a solvent containing 1-butanol/2-propanol/H₂O (3:12:4) and sprayed with 0.5% α-naphthol and 5% H₂SO₄ in ethanol. Lane 1, xylose; lane 2, XOS; lane 3, mMRS-XOS; lane 4, 0 h spent medium; lane 5, 24 h spent medium. Growth profiles demonstrated the strain’s preference to utilize smaller DPs of XOS.

FIG 3

Various trends of establishment of B. longum subsp. longum CR15 were observed across fecal samples. (A) A summary of the establishment trends of B. longum subsp. longum CR15 in all 20 samples in the presence (▲) or absence (•) of XOS. B. longum subsp. longum CR15 was clearly established in 7 samples (B), potentially established in 11 samples (C), and displaced or washed out in 2 samples (D). In the absence of XOS, the strain could not be established in any of the samples. Time zero samples were taken prior to inoculation of 10⁷ CFU/ml of the test strains. Horizontal dashed lines indicate the limits of detection (10⁴ CFU/ml); * indicates the sample from which B. longum subsp. longum CR15 was isolated.

FIG 4

Analysis of microbial community composition and diversity across treatments. By two measures of α-diversity, Shannon index (A) and number of ASVs (B), diversity was lower in the presence of XOS (▲) than in the absence of XOS (•). Principal-coordinate analysis (PCoA) (C) and principal-component analysis (PCA) (D) revealed distinct community profiles between groups at baseline (blue) and at the end of the fermentation period, with (green) or without (red) XOS (PERMANOVA, P = 0.001). *, significant difference between 0 and 24 h; †, significant difference between treatments at a particular time point.

FIG 5

Significant changes in taxa driven by XOS in establishment experiments with B. longum subsp. longum CR15. Wilcoxon rank sum test with FDR adjustment was used to identify significantly different taxa (FDR < 0.05) in the presence (A) and absence (B) of XOS. Nodes in orange indicate greater abundance at baseline than at 96 h, whereas nodes in green and red indicate greater abundance at 96 h than at baseline.

FIG 6

Enrichment of B. longum subsp. longum CR15 (▲) and B. pseudocatenulatum (•) in the presence of XOS. (A) When present at baseline in most samples (n = 9), B. pseudocatenulatum reached high cell numbers at the end of fermentation. (B) When B. pseudocatenulatum was below detection at baseline (n = 11), the species remained throughout. Horizontal dashed lines indicate the limits of detection (10⁴ CFU/ml).