Small Intestinal Epithelial Proliferation in a Microbial Depletion Animal Model Is Serotonin-dependent

Background: Intestinal failure is an unsolved surgical problem characterized by long-term dependence on parenteral nutrition due to inadequate small intestinal (SI) surface area. Intestinal microbial depletion in mice is associated with enhanced SI crypt cell proliferation, increased villus height, and greater SI absorptive surface area compared to mice with conventional microbiomes. The underlying mechanism for this observation is unknown. Specific intestinal bacteria are known to induce epithelial serotonin (5-HT) production and modulate expression of the serotonin reuptake transporter (SERT), which is responsible for terminating the action of 5-HT. Furthermore, potentiation of 5-HT signaling, either through genetic knockout or pharmacologic blockade of SERT, is associated with increased intestinal epithelial proliferation. We aimed to determine whether the proliferative epithelial changes associated with microbial depletion of the murine small intestine are due to alterations in 5-HT signaling.

Methods: Adult C57Bl/6J mice were pre-treated with 300mg/kg/day para-chlorophenylalanine (PCPA), an inhibitor of 5-HT synthesis, or normal saline (NS) intraperitoneally for 3 days, followed by treatment with a previously validated oral antibiotic regimen to induce microbial depletion, or vehicle alone. The four resulting treatment groups were NS/Antibiotics, NS/Vehicle, PCPA/Antibiotics and PCPA/Vehicle (n=6, respectively). Mice were injected with bromodeoxyuridine (BrdU) and euthanized at day 0, 7 or 14 of treatment. SI tissue was harvested for histologic examination and Sert gene expression analysis by quantitative RT-PCR. Average crypt proliferation index (CPI) was calculated as percentage of BrdU-positive crypt cells. Two-tailed t-tests were performed to compare villus height and CPI between groups, with significance set at p=0.05.

Results: In NS pre-treated animals, antibiotics significantly increased villus height after 14 days compared to vehicle (302 vs 252 microns; p<0.0001). Depletion of 5-HT stores by PCPA pre-treatment blocked this antibiotic-induced effect; villus height after PCPA/Antibiotics was no different from NS/Vehicle (241 vs 252 microns; p=0.42) and was significantly shorter than NS/Antibiotics (241 vs 302 microns; p<0.001). PCPA treatment was also associated with decreased CPI compared to NS treatment at all time points. Furthermore, CPI was higher in the NS/Antibiotics group than in the NS/Vehicle group after 14 days (0.49 vs 0.42, p<0.0001), but was no different between the PCPA/Antibiotics and PCPA/Vehicle groups (0.34 vs 0.32; p=0.15). On treatment day 14, Sert gene expression in SI tissue from NS/Antibiotics animals was 11.5% (95% confidence interval: 7.5%, 15.5%) of that in the NS/Vehicle group.

Conclusion: Inhibition of 5-HT synthesis blocks the enhanced crypt proliferation and villus growth observed after antibiotic-induced microbial depletion in mice. Furthermore, decreased intestinal Sert gene expression in antibiotic-treated animals suggests that potentiation of 5-HT signaling is a feature of microbial depletion. Overall, these findings provide evidence that the intestinal microbiome influences SI proliferation via modulation of 5-HT signaling, and may be a promising therapeutic target for intestinal failure.