If you give a mouse a poopsicle: a novel fecal microbiota transplant method for exploring the role of the gut microbiome in stress-related outcomes in mice

Abstract

The microbiome-gut-brain axis is a mediator of stress-related disorders. The number of preclinical studies exploring the potential causal mechanism of this connection using fecal microbiota transplantation (FMT) is growing. However, the most common method for delivering fecal transplants in rodent models is still oral gavage, which creates an adverse experience that may confound stress-related outcomes. Here, we establish an alternative methodology for FMT that decreases stress induced by traditional experimental procedures. We first used preference and anxiety behavior assays to identify antibiotic therapies having maximal tolerability and minimal anxiolytic properties. We then collected feces from donor mice and homogenized them with a microbe-stabilizing buffer to create a slurry, which was frozen into pellets ("poopsicles") for subsequent FMT. Recipient mice voluntarily consumed the pellets, and blood was collected to compare corticosterone levels relative to traditional gavage FMT. Plasma corticosterone levels were found to be significantly lower in mice receiving FMT via pellets compared to oral gavage. Furthermore, relative to gavage FMT, microbial signatures of mice receiving FMT via pellets were more similar to those of the donor pellets at one week following final FMT and were sustained for up to six weeks, as assessed by comparing Bray-Curtis beta-diversity distances. Together, these results establish effective antibiotic and FMT methods that minimize treatment-induced stress, while effectively transplanting fecal microbes between murine conspecifics.

Keywords: 16S sequencing; fecal microbiota transplant (FMT); microbiome; mouse; oral gavage; poopsicle; self-administration; stress.

Figure 1.. An alternative antibiotic cocktail does not reduce body weight

A: Experimental design of three different antibiotic regimens. B and D: timeline of bodyweight as a percentage of the starting body weight over the course of VNAM, or VNAA/VNBP (b) and water or VNAA (d). C and E: Average volume of administered liquids consumed per cage normalized by the average body weight per cage. Two-way RM ANOVA post hoc Sídak’s Multiple Comparisons Test (b, d); T-test (c, e). ns = no significance, *p ≤ 0.05, ****p ≤ 0.0001. Values represent mean ± SEM.

Figure 2.. VNAA effectively depletes fecal DNA and does not elevate anxiety behaviors in mice

A: Experimental design for determining the antibiotic cocktail, VNAA’s ability to deplete gut microbial populations and measuring the behavioral stress response to VNAA. B-C: Quantifying overall distance and time spent in the center of an open field in mice before, after, and a few days after a week of VNAA. D. DNA concentration of fecal pellets in the open field experiment. E. DNA concentration of mice either drinking VNAA for four or seven days at different timepoints. F. Experimental design for determining an effective duration for VNAA administration. Two-way RM ANOVA post hoc Sídak’s Multiple Comparisons Test (b, d); Mixed-effects model (e). ns = no significance, **p ≤ 0.01, ****p ≤ 0.0001. Values represent mean ± SEM.

Figure 3.. Anaerobic microbial communities are maintained during pellet preparation

A: Experimental design of pellet preparation and anaerobic culture. For each of the two donors we collected feces from, we measured the DNA concentration (b), Shannon index (c), and phyla (d) of their feces, the pellets made from their feces, and anaerobic cultures derived from each of the former sample types.

Figure 4.. Delivery via pellet results in less circulating corticosterone than via oral gavage

A: Experimental design for determining the differences in stress between oral gavage and voluntary consumption as FMT delivery techniques. B: Concentration of circulating corticosterone in mice from each group. One-way ANOVA post hoc Tukey’s Multiple Comparisons Test (b); ns = no significance, # = trending, and *p ≤ 0.05. Values represent mean ± SEM.

Figure 5.. Microbiomes of FMT recipients shift toward the donor

A: Experimental design of donors and recipients. B: Univariate Bray-Curtis distance between each group and the average pellet across experimental days. C: PCoA plots mapping Bray-Curtis distances at various timepoints. Two-way RM ANOVA; comparisons between AB_control and FMT 1 are represented by # and comparisons between AB_control and FMT3 are represented by *; *p ≤ 0.05, ***p ≤ 0.001. Values represent mean ± SEM.