Human Gut Bacteria Are Sensitive to Melatonin and Express Endogenous Circadian Rhythmicity

Abstract

Circadian rhythms are fundamental properties of most eukaryotes, but evidence of biological clocks that drive these rhythms in prokaryotes has been restricted to Cyanobacteria. In vertebrates, the gastrointestinal system expresses circadian patterns of gene expression, motility and secretion in vivo and in vitro, and recent studies suggest that the enteric microbiome is regulated by the host's circadian clock. However, it is not clear how the host's clock regulates the microbiome. Here, we demonstrate at least one species of commensal bacterium from the human gastrointestinal system, Enterobacter aerogenes, is sensitive to the neurohormone melatonin, which is secreted into the gastrointestinal lumen, and expresses circadian patterns of swarming and motility. Melatonin specifically increases the magnitude of swarming in cultures of E. aerogenes, but not in Escherichia coli or Klebsiella pneumoniae. The swarming appears to occur daily, and transformation of E. aerogenes with a flagellar motor-protein driven lux plasmid confirms a temperature-compensated circadian rhythm of luciferase activity, which is synchronized in the presence of melatonin. Altogether, these data demonstrate a circadian clock in a non-cyanobacterial prokaryote and suggest the human circadian system may regulate its microbiome through the entrainment of bacterial clocks.

Fig 1. Swarming behavior in E. aerogenes is induced by melatonin and occurs with a circadian frequency.

A Swarming behavior in control treated (left cultures vs. treatment with 1nM melatonin (right. Images were equally enhanced using “Bump Map” in GIMP software to highlight banding patterns. B The increase in swarming was only seen at 100pM and 1nM concentrations of melatonin, * = p value < 0.001 compared to vehicle treated cultures, n = 16 cultures per treatment. C Period of swarming as calculated by the number of rings observed per culture period of 4 days in n = 16 cultures per treatment, * = p value < 0.001. D Area of bacterial spread was unaffected by tryptophan (left, serotonin (middle and N-acetylserotonin (right, n = 16 cultures per treatment. E Melatonin did not affect growth in K. pneumoniae (left or clinical or lab strains of E. coli (middle and right, respectively, n = 16 cultures per strain per treatment.

Fig 2. Neither lab nor clinical strains of E. coli nor K. pneumoniae show swarming response to other indoles.

Cultures of clinical isolates of E. coli (left panels, DH5-α (middle panels, and K. pneumoniae (right panels were tested for swarming/growth in the presence of tryptophan (top row, serotonin (middle row, and N-acetylserotonin (bottom row, n = 16 cultures per strain per treatment.

Fig 3. Bioluminescence recording of MotA::luxcdabe transformed E. aerogenes confirms a temperature compensated circadian rhythm.

A) Normalized bioluminescence rhythms from control-treated (top panels) and melatonin-treated (bottom panels) cultures show circadian rhythms at (from left to right) 27° (n = 5/treatment), 34° (n = 5/treatment), 37° (n = 5 control and 7 melatonin-treated) and 40° (n = 6 control and 6 melatonin-treated). Time scales represent days after plates were inoculated with bacteria, which varied in the amount of time needed to stabilize and begin outgrowth. B) Periodogram analysis-derived peak phases of rhythmic cultures from (A) reveal that control-treated cultures (white circles) show greater variation in peak phase than melatonin-treated cultures (black circles), which are more synchronized at all three temperatures. C) Periods of rhythms varied between 22 and 28 hours among temperature and melatonin treatments, but were not significantly affected by temperature or melatonin.

Fig 4. The data presented here show that a clinical isolate of E. aerogenes expresses a circadian rhythm in MotA expression and displays a swarming response to melatonin in a dose- and temperature-dependent manner.