A 14-Day Double-Blind, Randomized, Controlled Crossover Intervention Study with Anti-Bacterial Benzyl Isothiocyanate from Nasturtium (Tropaeolum majus) on Human Gut Microbiome and Host Defense

Abstract

Despite substantial heterogeneity of studies, there is evidence that antibiotics commonly used in primary care influence the composition of the gastrointestinal microbiota in terms of changing their composition and/or diversity. Benzyl isothiocyanate (BITC) from the food and medicinal plant nasturtium (Tropaeolum majus) is known for its antimicrobial activity and is used for the treatment of infections of the draining urinary tract and upper respiratory tract. Against this background, we raised the question of whether a 14 d nasturtium intervention (3 g daily, N = 30 healthy females) could also impact the normal gut microbiota composition. Spot urinary BITC excretion highly correlated with a weak but significant antibacterial effect against Escherichia coli. A significant increase in human beta defensin 1 as a parameter for host defense was seen in urine and exhaled breath condensate (EBC) upon verum intervention. Pre-to-post analysis revealed that mean gut microbiome composition did not significantly differ between groups, nor did the circulating serum metabolome. On an individual level, some large changes were observed between sampling points, however. Explorative Spearman rank correlation analysis in subgroups revealed associations between gut microbiota and the circulating metabolome, as well as between changes in blood markers and bacterial gut species.

Keywords: BITC; Escherichia coli; antimicrobial; gut microbiome; human beta defensin 1; metabolome; nasturtium.

Figure 1

Microbial composition. Fractions of phyla (A) and genera (B) for each sample; n = 29 test subjects.

Figure 2

Microbial diversity analysis. Principal component analysis (A) using the k-means clustering algorithm (above colored by sampling day, below colored by individual), and boxplot of Shannon index as alpha diversity metric (B); n = 29 test subjects.

Figure 3

Abundance of bacteria from the class Bacilli. Absolute abundance per individual and sampling day of bacterial genera (A), and for each genus the mean and standard deviation per sampling day (B).

Figure 4

Abundance of bacteria from the order Enterobacterales. Absolute abundance per individual and sampling day of bacterial genera (A), and for each genus the mean and standard deviation per sampling day (B).

Figure 5

Spearman correlation analysis of the gut microbiome and metabolome. Spearman correlation analysis is conducted on the difference in the log2-fold changes of bacteria genera between pre vs. post treatment with verum and the log2-fold changes of metabolites between pre vs. post treatment with verum. Spearman correlation of the difference in bacteria genera and the difference in hydrophilic metabolites (A) for subjects with a >25% upregulated serum PGE₂, for the difference in bacteria genera and the difference in hydrophilic metabolites for subjects with a >25% downregulated serum PGE₂ (B), and for the difference in bacteria genera and the difference in SCFAs for subjects with >25% upregulated serum PGE₂ (C). Correlation with an unadjusted p-value < 0.05 or 0.01 are emphasized by bigger squares in the correlation matrix. Positive correlation coefficients are displayed in red, negative ones in blue. To the left of the matrix, color-coded bars of the phylum and order of the respective genera are displayed. Additionally, to the left and above the correlation matrix, the median of the absolute difference in the log2-transformed abundance of bacteria genera (left) and metabolites (top) between verum and control is displayed as a color-coded bar, where blue corresponds to negative and red to positive values. No correlations were significant after FDR correction.

Figure 6

Antimicrobial effects of nasturtium. Quantification of hBD-1 in the samples before (pre) and after (post) nasturtium intervention was conducted via ELISA in urine (A), exhaled breath condensate (B), and serum (C). Each dot represents results from one subject, tested in duplicate. The line represents the mean. * p < 0.05 and ** p < 0.01. Urine samples were incubated for 12 h at 37 °C with 10⁵ CFU/mL of one of the three E. coli cultures. Antibacterial activity was measured at OD 600 nm as an indicator for bacterial growth (D). Correlations for E. coli bacterial growth and post-intervention BITC-NAC concentration (n = 26) (E), pH (n = 31) (F), and hBD-1 concentration (n = 27) (G) were calculated using Pearson (F) and Spearman (E,G) analyses. Statistical significance was described as p < 0.05 and p < 0.01.