Table 2.
Human dietary interventions that tested the effect of soy on the human gut microbiota
| Study product | References | Subject number/description | Microbiota alterations | Health effects | Comment |
|---|---|---|---|---|---|
| Cheonggukjang (fermented soy product) | [109] | 60 postmenopausal Korean women | Reduced proportion of phylum Firmicutes | Menopausal symptom relief and reduced blood glucose only in “high beneficial bacteria” product group | Study product was stratified based on fermentation process and product “beneficial bacteria” content. Microbiota effects were related to this, and only presented at phylum level |
| Black soymilk (instead of pork in diet) | [110] | 8 test, 8 control pre-diabetic/obese Taiwanese adults | No changes in alpha- or beta-diversity; increased relative abundance of Butyricicoccus and SCFA producers; lower abundance of Rothia spp. | Reduced oxidative stress and lower levels of glycative stress markers | V3-V4 16S amplicon sequencing limits resolution |
| Fermented versus unfermented soy milk | [111] | 56 test, 56 control healthy Japanese adults with low Bifidobacterium count | Increased abundance of lactobacilli and bifidobacteria, Faecalibacterium in both groups, more pronounced in fermented soy group; increased Clostridiaceae and Enterobacteriaceae in non-fermented group | Higher levels of faecal SCFAs in fermented soy group; improvement in defaecation habits in both groups, achieved earlier in fermented soy group | V1-V2 16S amplicon sequencing was supplemented by targeted qPCR |
| Commercial fermented soybean beverage | [112] | Lean (10) and obese (9) participants in USA | Increased abundance of Bifidobacteria and Blautia in lean subjects; increased oral abundance of Veillonellaceae in lean subjects; study product significantly increased Fusobacteria in obese participants | Not tested | V4 16S amplicon sequencing |
| Fermented soy milk (FSM) plus L. casei plus aglycones, versus unfermented soy milk | [113] | 30 test, 30 controls in Japan | Lactobacillaceae significantly increased, Bifidobacteriaceae trended upwards, Enterobacteriaceae and Porphyromonadaceae significantly lower in FSM group | Both groups had improved skin condition and significantly increased levels of urinary isoflavones | V1-V2 16S amplicon sequencing |
| Soy isoflavone concentrate tablet, 80 m/day | [114] | 16 menopausal women for 6 months, in Spain | Increased abundance of Lactonifactor longoviformis, Faecalibacterium prausnitzii, Bifidobacterium sp., Ruminococcus sp., isoflavone metabolizers; Clostridium leptum and C. coccoides abundance increased in equol producers | Not tested | PCR-DGGE* and qPCR |
| Soy bars containing isoflavones | [115] | 17 postmenopausal women in USA | Significant increases in Bifidobacterium after soy consumption; Bifidobacterium and Eubacterium were significantly greater in equol vs non-equol producers | Correlations detected between urinary isoflavones and isoflavone metabolites with microbiota | Microbiota was studied by PCR-DGGE* and pyrosequencing of 16S V4 region |
| Low glycinin soymilk (LGM), conventional soymilk (S) or bovine milk | [116] | 64 overweight and obese men in USA | Total bacteria increased in all treatments; Bacteroides-Prevotella increased in LGM; Firmicutes: Bacteroidetes ratio decreased in LGM and S groups; Eubacterium and Clostridium were more abundant in the S group | Not tested | qPCR for total bacterial load and quantitation of Bacteroides-Prevotella, Bifidobacterium, and Lactobacillus; Short-read sequencing of V1-V3 region. Microbiota of all 3 groups separated by beta-diversity analsysi |
| Non-fermented soybean milk (NFSM) and fermented soybean milk (FSM) | [117] | 10 healthy subjects, 21–25 years old in Japan | Bifidobacteria and lactobacilli were more abundant in both groups, Clostridia were less abundant | Not tested | Bacteriological culture on selective media |
| Non-fermented soybean milk (NFSM) and fermented soybean milk (FSM), cross-over | [118] | 28 healthy adults in China | Bifidobacterium spp. and Lactobacillus spp. Increased; higher ratios of Bifidobacterium spp. and Lactobacillus spp. to Clostridium perfringens (P < 0.05) | Not tested | Bacteriological culture on selective media |
*DGGE: Denaturing gradient gel electrophoresis