Chronic exposure to N-butyl-N-(4-hydroxybutyl)-nitrosamine (BBN) induces bladder cancer in mouse models. Thus, the investigators used this model to examine the influence of gut microbiota on BBN-induced carcinogenesis and toxicokinetics.
C57BL/6 mice were either treated with antibiotic or not treated and concomitantly received BBN through drinking water (0.05% v/v). Bacterial load reduction in the gut was confirmed in the antibiotic-treated group, showing 99.99% fewer colony-forming units in treated mice than in untreated mice. Furthermore, gut microbiome α-diversity was significantly reduced, and the overall community composition was significantly altered in antibiotic-treated mice.
Mice were exposed to BBN for 12 weeks and then no BBN was given for 8 weeks, after which blinded pathological analysis of their bladders was undertaken. No bladder tumour pathology was found in 81% of antibiotic-treated mice, but 77% of bladders from mice that did not receive antibiotics showed neoplastic changes. Invasive tumours were observed in 53% of the BBN-only mice but only in 12% of those treated with antibiotics. Concentrations of N-butyl-N-(3-carboxypropyl)-nitrosamine (BCPN; the oxidation product of BBN) were lower in bladder tissue from antibiotic-treated mice than BBN-only mice during the 12-week BBN exposure. Furthermore, urinary elimination of BCPN was significantly reduced in mice that received antibiotics, but BBN elimination did not differ between the two groups.
BCPN was found to accumulate in the contents of the caecum and colon luminal of BBN-treated mice and levels reduced from the caecum to the rectum. In germ-free mice treated with BBN for 10 days with or without antibiotics, BCPN levels in various tissues did not differ. In antibiotic-treated mice, BBN levels in the lower intestine were more than a magnitude less than in BBN-only animals and glucuronidated BBN was detected in their large intestines. Glucuronidated BBN in the caecum could be deconjugated by β-glucuronidases that are prevalent in gut bacteria. In antibiotic treated mice, more BBN was liberated from their large intestines than in BBN-only mice. The amount of BCPN found was comparable between the groups.
Incubation of ex vivo microbiota cultures from mouse caecal, colon and rectal samples with BBN showed that it was oxidized to BCPN under aerobic or microaerobic conditions.
Characterization of axenic bacterial cultures isolated from mouse intestine under aerobic and microaerobic conditions enabled identification of 12 species belonging to 8 different bacterial genera and 3 distinct phyla capable of BBN oxidation. Mapping of the reference sequence of the amplicon sequence variants (ASVs) identified in the samples to the full-length 16S rRNA sequences of the BBN-converting bacterial isolates narrowed this number down to 8 ASVs belonging to 4 different genera (Escherichia, Lactobacillus, Corynebacterium and Staphylococcus). The collective abundance in the gut of these putative BBN-metabolizing bacteria was estimated to be low — 0.48% in the caecum, 1.66% in the colon and 0.50% in the rectum.
The capacity of human microbiota to oxidize BBN to BCPN was investigated by sequencing 11 culturable faecal samples from human volunteers. Only one putatively active bacterial genus (Escherichia) overlapped between human and mouse. Ex vivo culture of the faecal communities with BBN in aerobic or microaerobic conditions showed only 6 of 10 evaluable communities produced BCPN after 24 h. Large differences were observed in the BBN-oxidizing activity of these six communities. The strongest activity was observed under microaerobic conditions. Colonization of germ-free mice with the stool microbiota of one participant recapitulated the metabolic phenotype observed in model mice. Human gut microbiota also produced BCPN.
Characterization identified 25 unique BBN-converting species and 18 different bacterial genera from 2 distinct phyla (Firmicutes and Proteobacteria) in human microbiota. Mapping of the full-length 16S rRNA sequences of the isolates to the 16S rRNA microbiome data revealed seven ASVs of the faecal microbiomes, belonging to three different genera (Escherichia, Enterococcus and Haemophilus). The relative abundance of these ASVs was between 0% and 10.52% with an average of 3.94%.
Different Escherichia strains had the ability to oxidize BBN. In vivo, monocolonization of germ-free mice with a representative human Escherichia isolate, Escherichia flexneri, and 10 days of BBN administration in drinking water resulted in BCPN accumulation and reabsorption in the lower intestine, as was observed in previous experiments. Recapitulation of the interspecies interactions in the gut of germ-free mice using inoculation with two non-BBN metabolizing bacterial strains (Coprococcus comes and Flavonifractor plautii) or these two strains plus BBN-metabolizing E. flexneri showed that only the three-member community resulted in BCPN accumulation in the lower intestine. Microbiota also altered the toxicokinetics of other nitrosamine carcinogens including N-ethyl-N-(4-hydroxybutyl)-nitrosamine (EHBN), N-nitrosodibutylamine and N-propyl-N-butyl-nitrosamine.
Exposure to EHBN for 20 weeks also resulted in bladder carcinogenesis in mice with intact micro-biota, but in antibiotic-treated mice this effect was reduced. The oxidized metabolite of EHBN, N-ethyl-N-(3-carboxypropyl)-nitrosamine was only found in the lower intestine of EHBN-only animals.
“No bladder tumour pathology was found in 81% of antibiotic-treated mice”Together, these observations show that gut microbial metabolism can affect chemically induced tumorigenesis by altering the toxicokinetics of nitrosamine carcinogens. Thus, in the future, the microbiome could be targeted to improve predisposition risk for and also potentially prevent cancer.
Louise Lloyd, Nature Reviews Urology