Cranberry extracts promote growth of Bacteroidaceae and decrease abundance of Enterobacteriaceae in a human gut simulator model

Abstract

The opportunistic pathogen Escherichia coli, a common member of the human gut microbiota belonging to the Enterobacteriaceae family, is the causative agent of the majority of urinary tract infections (UTIs). The gut microbiota serves as a reservoir for uropathogenic E. coli where they are shed in feces, colonize the periurethral area, and infect the urinary tract. Currently, front line treatment for UTIs consists of oral antibiotics, but the rise of antibiotic resistance is leading to higher rates of recurrence, and antibiotics cause collateral damage to other members of the gut microbiota. It is commonly believed that incorporation of the American cranberry, Vaccinium macrocarpon, into the diet is useful for reducing recurrence of UTIs. We hypothesized such a benefit might be explained by a prebiotic or antimicrobial effect on the gut microbiota. As such, we tested cranberry extracts and whole cranberry powder on a human gut microbiome-derived community in a gut simulator and found that cranberry components broadly modulate the microbiota by reducing the abundance of Enterobacteriaceae and increasing the abundance of Bacteroidaceae. To identify the specific compounds responsible for this, we tested a panel of compounds isolated from cranberries for activity against E. coli, and found that salicylate exhibited antimicrobial activity against both laboratory E. coli and human UTI E. coli isolates. In a gut simulator, salicylate reduced levels of Enterobacteriaceae and elevated Bacteroidaceae in a dose dependent manner.

Fig 1. The average relative abundance changes of the most abundant families in a human gut simulator.

A) A gut simulator seeded with a stool sample lacking Enterobacteriaceae B) or a stool sample high in Enterobacteriaceae and treated with whole cranberry powder (red), phenolic-enriched extract (green), or phenolic-deficient extract (purple) in duplicate. A gut simulator seeded with a human microbiota was treated each day for five days with 1 mg/mL of each extract or powder respectively. 16S rRNA gene sequencing was performed by an Ion Torrent PGM. The average relative abundance change was calculated from before treatment to after treatment with error bars representing standard deviation. Statistical significance (*) was calculated using the Student’s T-test for non-paired samples to compare the gut simulator community before treatment and after each respective treatment (p<0.05).

Fig 2. Relative abundance of Enterobacteriaceae in gut simulator before and after treatment with salicylate or β-resorcylate.

A gut simulator seeded with a human gut microbiota was treated for five days with 1 mg/mL (1X MIC) of salicylate or β-resorcylate. All vessels had a 1% DMSO final concentration. 16S rRNA gene sequencing was performed by an Ion Torrent PGM. Salicylate and β-resorcylate significantly reduced the relative abundance of Enterobacteriaceae. β-resorcylate and control treatments were run in triplicate and the salicylate treatment was run in duplicate and averaged. Error bars represent standard deviation. Statistical significance (*) was calculated using the Student’s T-test for non-paired samples (p<0.05).

Fig 3. Average relative abundance change of five most abundant families in a gut simulator treated with salicylate.

A gut simulator seeded with a human gut microbiota was treated with salicylate each day for five days at at 0.01 mg/mL (0.01X MIC, red), 0.1 mg/mL (0.1X MIC, green) and 1 mg/mL (1X MIC, purple). All vessels had a 1% final DMSO concentration. 16S rRNA sequencing was performed by an Ion Torrent PGM. Salicylate treatment increased Bacteroidaceae and decreased Enterobacteriaceae in a dose dependent manner. Experiment was performed in triplicate and the average percent abundance change from before treatment to after treatment was calculated. Error bars represent standard deviation. Statistical significance (*) was calculated using the Student’s T-test for non-paired samples to compare the gut simulator community before treatment and after each respective treatment (p<0.05).