We established an efficient way to deplete intestinal bacteria in three days, and subsequently transplant fecal microbiota via gavage of fecal fluid prepared from fresh or frozen intestinal contents in mice. We also present an optimized method to detect the frequency of IgA-coated bacteria in the gut.

Gut microbiota exert pleiotropic roles in human health and disease. Fecal microbiota transplantation (FMT) is an effective method to investigate the biological function of intestinal bacteria as a whole or at the species level. Several different FMT methods have been published. Here, we present an FMT protocol that successfully depletes gut microbiota in a matter of days, followed by transplantation of fecal microbiota from fresh or frozen donor intestinal contents to conventional mice. Real time-PCR is applied to test the efficacy of bacterial depletion. Sequencing of the 16S ribosomal RNA (rRNA) is then applied to test the relative abundance and identity of gut microbiota in recipient mice. We also present a flow cytometry-based detection method of immunoglobulin A (IgA)-coated bacteria in the gut.

A diverse gut microbiota plays a major role in maintaining host homeostasis. This microbiome aids in various physiological processes ranging from digestion and absorption of nutrients from food, defense against infection of pathogens, regulation of immune system development, and immune homeostasis¹. Perturbation in gut microbial composition has been linked to many diseases, including cancer², autoimmune diseases³, inflammatory bowel disease⁴, neurological diseases⁵, and metabolic diseases^6,7. Germ-free (GF) mice are powerful tools in fecal microbiota transplantation models to study the biological effects of microbiota⁸. However, the GF housing environment is very stringent, and performing fecal microbiota transplantation (FMT) in these mice is expensive. Moreover, GF mice have different barrier and mucosal properties, which regulate bacterial penetration, compared to conventional mice⁹. These factors limit the wide application of GF mice in studies. An alternative to using GF mice is to deplete the microbiota in conventional mice using an antibiotic cocktail followed by FMT. Previously reported FMT methods are not well described and inconsistent; therefore, it is necessary to establish a feasible, efficient, and reproducible protocol to perform FMT using conventional mice.

Several steps are crucial to a successful FMT. The efficiency of microbiota depletion is the first important step. For bacteria depletion, use of a single broad-spectrum antibiotic (e.g., streptomycin¹⁰) or an antibiotic cocktail (triple or quadruple-antibiotic treatment) has been reported^11,12. The quadruple-antibiotic treatment including ampicillin, metronidazole, neomycin, and vancomycin, has been found to be the most effective regimen and has been used in several studies^13,14,15. In addition to the type of antibiotic used, the route of administration, dosage, and duration of the antibiotic treatment affect the efficacy of bacterial depletion. Some researchers apply antibiotics in the drinking water to eliminate bacteria in the gastrointestinal tract¹⁵. However, it is hard to control the dosage of antibiotics that each mouse receives this way. Therefore, in subsequent studies, researchers have treated mice with antibiotics by oral gavage for 1–2 weeks¹² to achieve satisfactory depletion. However, the long-term use of antibiotics can be problematic, as the antibiotics themselves may affect some diseases in rodent models¹⁶. Therefore, faster and more efficient methods for microbiota depletion are warranted.

Fecal fluid preparation is another key step to ensure successful FMT. In the gastrointestinal tract, pH ranges from 1 in the stomach to 7 in the proximal and distal intestine⁹. The microbiota in the stomach is limited due to high acidity and includes Helicobacter pylori¹⁷. The proximal intestine produces bile acid for the liver-gut circulation, and contains microbiota associated with fat, protein, and glucose digestion. The distal intestinal tract contains abundant anaerobic bacteria and exhibits high microbial diversity¹⁸. Given the spatial heterogeneity of gut microbiota, it is imperative to isolate gut contents from different regions of the intestinal tracts for fecal fluid preparation. Additionally, other factors, including the nature of the donor sample (e.g., fresh or frozen sample), transplantation frequency, and duration are crucial when performing FMT. Frozen stool is most commonly used for colonizing conventional mice with human gut microbiota¹⁹. In contrast, FMT using fresh stool from animal donors is more appropriate and commonly used in animal models^20,21. FMT frequency and duration vary depending on the experimental design and models. In previous studies, FMT was either performed daily or every second day. The transplantation duration ranged from 3 days²² to 5 weeks²³. In addition to the above factors, maintaining an aseptic surgical environment and the use of sterilized surgical instruments is crucial to avoid unexpected environmental bacterial contamination.

The gut microbiota has the potential to regulate the accumulation of cells that express Immunoglobulin A (IgA). IgA, a predominant antibody isotype, is critical in protecting the host from infection through neutralization and exclusion. High-affinity IgA is transcytosed into the intestinal lumen and can bind and coat offending pathogens. In contrast, coating with IgA may provide a colonization advantage for bacteria²⁴. In contrast to pathogen-induced IgA, indigenous commensal-induced IgA has lower affinity and specificity²⁵. The proportion of intestinal bacteria coated with IgA is reported to be significantly increased in some diseases^25,26. IgA-coated bacteria can initiate a positive feedback loop of IgA production²⁷. Therefore, the relative level of IgA-coated bacteria can predict the magnitude of the inflammatory response in the gut. In fact, this combination can be detected via flow cytometry²⁸. Using IgA-based sorting, Floris et al.²⁷, Palm et al.²⁵, and Andrew et al.²⁹ acquired IgA⁺ and IgA^- fecal bacteria from mice and characterized taxa-specific coated-intestinal microbiota via 16S rRNA sequencing.

In this study, we describe an optimized method to efficiently deplete intestinal dominant bacteria and colonize conventional mice with fresh or frozen fecal microbiota isolated from the contents of the ileum and colon. We also demonstrate a method based on flow cytometry to detect IgA-binding bacteria in the gut.

Animal experiments were conducted in accordance with the current ethical regulations for animal care and use in China.

NOTE: Animals were housed in a specific pathogen-free (SPF), controlled environment under 12-hour light and dark cycles at 25 °C. Food was irradiated before being fed to mice. Drinking water and cages were autoclaved before use. Eight-week-old male C57BL/6J mice were used in the study following 1 week of acclimatization. They were divided into several groups based on the experiment design. Each group consisted of at least three mice.

1. Gut microbiota depletion

1. Antibiotic cocktail preparation
   1. Prepare an antibiotic solution in sterile phosphate buffered saline (PBS) with 0.5 mg/mL vancomycin hydrochloride, 1 mg/mL metronidazole, 1 mg/mL ampicillin sodium salt, and 1 mg/mL neomycin sulfate.
      NOTE: Store the antibiotics at 2–8 °C and avoid direct light. Prepare the antibiotic cocktail solution fresh and use immediately.
2. Treatment regimen
   1. For 3 days, orally administer 200 μL of prepared antibiotic cocktail with a #10 needle to a mouse. Hold the mouse vertically in one hand while using the other hand to adjust the angle of the needle to avoid reaching the stomach.
      NOTE: The mouse must be handled gently to avoid struggling, because surface damage of the esophagus and stomach may result in inflammation or death.
   2. Add antibiotics to the drinking water at the same concentrations as indicated in step 1.1.1.
   3. Three days later, sacrifice the mouse by CO₂ asphyxiation.
   4. Dissect the abdomen aseptically using sterile instruments. At a distance of 2 cm away from the caecum, cut off a 2 cm tract of the ileum. Using sterile tweezers to clamp off one end of the tract, squeeze the fresh contents out of the tract into pre-weighed tubes.
   5. To collect the contents of the cecum, cut it in half with surgical scissors. Squeeze the contents of each half of the cecum into pre-weighed tubes.
3. Evaluate gut microbiota depletion efficacy.
   1. Weigh gut contents isolated from the ileum or cecum from naive mice and antibiotic cocktail treated mice, and extract stool microbiota DNA using a kit according to the manufacturer’s instructions. Determine the DNA concentration according to the formula:
      
      concentration_sample (μg/mL) = OD₂₆₀ x 50
       
   2. Construct a standard curve.
      1. Amplify E. Coli genes with a polymerase chain reaction (PCR) using V3-V4 specific primers. Predenature at 95 °C for 1 min, amplify for 40 cycles of 95 °C for 10 s, 60 °C for 30 s, 72 °C for 30 s.
      2. Construct a plasmid by linking the amplified product to a TA vector. Clone the culture with a TA vector kit according to the manufacturer’s instructions.
      3. Measure the plasmid concentration (ng/μL) based on the OD₂₆₀ value as described in step 1.3.1.
         NOTE: The unit for plasmid concentration is different from the unit for DNA concentration.
      4. Calculate the copy numbers using the following formula, where MW stands for molecular weight:
         
         copies in 1 μL = concentration (ng/μL) x 10^-9 x 6.02 x 10²³/(MW_plasmid x 660)
          
      5. Reconstitute the plasmid with double distilled water to a final concentration of 40 µg/μL. Prepare a 10x gradient dilution series with eight stages. Construct a standard curve by plotting the C_T value (Y-axis) against log (Copies_plasmid) (X-axis) after PCR amplification, where C_T stands for threshold cycle for target amplification. Extract the equation from the plot (in this case, it was y = -3.07x + 45.07, R² = 0.994).
   3. Amplify 1 μL of the DNA samples obtained in step 1.3.1 extracted from the ileum or cecum from a naive control group and the antibiotic cocktail treatment group by real-time PCR as described in step 1.3.2.1. Calculate the copy number in 1 μL of the DNA samples based on the standard curve in step 1.3.2.5 and then calculate the total copy number using the following formula:
      Total copy number (per mg) in the weighted samples = copy numbers × total DNA volume/initial weight of the sample
4. Analyze the microbiota DNA by sequencing the 16S rRNA V3 and V4 regions.

2. Fecal microbiota transplantation

1. Fecal fluid preparation for both fresh and frozen stool samples
   NOTE: All instruments are soaked in 75% alcohol prior to usage to avoid preexisting bacterial contamination. It is crucial to avoid contamination when the ileum and colon contents are collected.
   1. Sacrifice 3–5 donor mice by CO₂ asphyxiation.
   2. Collect the contents of the ileum as described in step 1.2.4.
   3. To collect the contents in the colon, make the first incision near the anus and cut the upper 2 cm tract. Extract the contents as explained in step 1.2.4. Collect the samples in 1 min to reduce exposure to air.
   4. For frozen fecal fluid, collect the ileum or colon contents as described in steps 2.1.2 and 2.1.3, and flash freeze the contents in liquid nitrogen.
   5. Store samples in a -80 °C freezer until ready for use.
2. Pool and weigh the contents, add in fresh sterile tubes with beads. For frozen samples, thaw the fecal pellets on ice before weighing.
3. Add sterile PBS into the tube and resuspend the fecal pellets in PBS (1 mL of PBS/0.2 g of fecal pellet) using a 5 mL syringe needle. Homogenize the fecal pellet completely with beads and vortex 3x for 1 min each.
4. Centrifuge at 800 x g for 3 min, then filter the supernatants by passing through a 70 µm cell strainer. Collect the filtered fecal fluid in a sterile tube and use for FMT immediately.

3. Fecal microbiota transplantation procedure

1. Administer 200 µL of prepared fecal fluid to the microbiota-depleted mice via oral gavage every second day for 7 days. Gavage control mice with 200 µL of sterile PBS.
2. Sacrifice the mice by CO₂ asphyxiation and harvest the contents from the ileum and colon according to steps 2.1.2 and 2.1.3. Store samples at -80 °C until further processing.
3. Extract the microbiota DNA of the ileum and colon contents as described in step 1.3.1. Verify the microbiota composition in the gut of transplanted mice by 16S rRNA sequencing.

4. IgA-coated bacteria measurement

1. Sample preparation
   1. Collect 50 mg of fecal pellets from donor mice as described in steps 2.1.2 and 2.1.3, and incubate in 1 mL of sterile PBS at 4 °C for 1 h. Homogenize the pellets using a bead beater for 5 s.
   2. Centrifuge the solution at 300 x g for 10 min at 4 °C. Collect supernatant after filtrating through a 70 µm strainer.
      NOTE: Avoid high speed centrifugation.
   3. Add 5 µL of the supernatant to 1 mL of 1% bovine serum albumin (BSA) in PBS (FACS buffer). Pellet at 8,000 x g for 5 min at 4 °C and discard the supernatant.
   4. Resuspend the pellet in 1 mL of FACS buffer, centrifuge at 8,000 x g for 5 min at 4 °C, and discard the supernatant.
      NOTE: The volume used here may need to be optimized. Too much of the supernatant at this step may mask the positive signal of IgA-coated bacteria.
   5. Resuspend the pellet in 100 µL of PBS containing 10% goat serum and incubate on ice or at 4 °C for 30 min. Add 1 mL of FACS buffer in the tube and centrifuge at 8,000 x g for 5 min at 4 °C to pellet.
2. Flow cytometry
   1. Resuspend the pellet obtained in step 4.1.5 with 200 µL of FACS buffer containing biotin anti-mouse IgA antibody (1:100) and APC-conjugated anti-biotin antibody (1:100) and incubate for 30 min at 4 °C.
   2. Add 1 mL of FACS buffer in the tube and centrifuge at 8,000 x g for 5 min at 4 °C. Discard the supernatant.
   3. Stain the pellet from step 4.2.2 by adding 200 µL of FACS buffer containing green fluorescent nucleic acid stain (1:200) and incubate at 4 °C for 5 min.
   4. Add 1 mL of FACS buffer and centrifuge at 8,000 x g for 5 min at 4 °C to pellet.
   5. Resuspend the pellet in 250 µL of FACS buffer before measurement.
   6. Acquire the data using a flow cytometer and analyze it using FlowJo software.

The FMT schedule used in this study is outlined in Figure 1. After treatment with the antibiotic cocktail, the efficiency of intestinal microbiota depletion was analyzed by sequencing the 16S rRNA region. We detected 196 species in the ileum of naive mice, whereas 3-day antibiotic treatment rapidly reduced the bacterial species to 35 (Figure 2A). There were eight species detected solely in mice that underwent the antibiotic cocktail treatment (Figure 2A). Beta-diversity analysis at the genus level further indicated that the predominant bacteria in the ileum of naive mice, e.g., Norank_f__Bacteroidales_S24-7_group, Desulfovibrio, Lactobacillus, and Staphyloccoccus, were eliminated after antibiotic cocktail treatment. In contrast, Escherichia-Shigella accounted for more than 99% of the gut bacteria in the antibiotic-treated mice (Figure 2B). Besides the significant reduction of bacterial species, mice receiving the antibiotic cocktail had a dramatic decrease in copy numbers in the equal amount of gut contents in the ileum and cecum compared to naive mice (Figure 2C). The efficacy of gut microbiota depletion reaches up to 99%.

Firmicutes, Bacteroidetes, Proteobacteria, Actinobacteria, and Verrucomicrobia were found to be the most abundant bacteria in donor mice at the phylum level (Figure 3A). Mice receiving either fresh or frozen fecal transplantation for 7 days displayed similar predominant bacterial taxa to the naive group. Principal component analysis (PCoA) was performed to identify differences among donor, PBS, fresh, and frozen fecal transplantation groups. By comparison, samples from the fresh fecal transplantation group formed a cluster that was close to the cluster of donor groups in PCoA1 (3.0%) and PCoA2 (6.3%) (Figure 3B). Compared to the PBS group, the bacterial cluster formed in the frozen fecal transplantation group was closer to that of control group. While samples from the PBS group were clustered differently from the control group cluster, PCoA1 and PCoA2 accounted for 27.9% and 8.6% of the variance, respectively (Figure 3B).

Figure 4 represents a typical flow cytometry plot analysis of IgA-coated bacteria in intestinal contents based on antibody-binding bacterial cell sorting. Figure 4B shows the sample without antibody staining. As shown in Figure 4C, there was positive APC fluorescence signal in the sample stained with biotin anti-mouse IgA antibody and APC streptavidin. We also detected the positive staining population in the sample stained with only green fluorescent nucleic acid stain (Figure 4D). Moreover, no fluorescence compensation was observed between APC and bacterial nuclear staining signal channel (Figure 4C,D). IgA-coated bacteria in the contents of the ileum from naive mice were 1.97% based on flow cytometry analysis (Figure 4E).

Figure 1: Experimental schedule of fecal microbiota transplantation. ATB = antibiotic treatment; FMT = Fecal microbiota transplantation; PBS = phosphate buffer saline. Please click here to view a larger version of this figure.

Figure 2: Gut microbiota community alteration after antibiotic cocktail treatment. (A) Venn diagram displaying the degree of bacterial overlap at the species level between naive and ATBs cocktail treatment groups. (B) The relative abundance of different taxa at the phylum level in mice with or without antibiotic cocktail treatment. (C) Copy numbers in the contents of the ileum and cecum from naive mice or mice treated with the antibiotic cocktail. Please click here to view a larger version of this figure.

Figure 3: Characteristics of gut microbiota in donor mice and mice receiving sterile PBS or fecal fluid prepared from fresh or frozen contents of ileum. (A) Relative abundance of the dominant bacteria at the phylum level in the ileum of donor mice, PBS, fresh, and frozen fecal transplantation groups. (B) Principal coordinate analyses (PCoA) of beta diversity based on binary Jaccard analysis in the gut microbiota of donor mice and mice receiving sterile PBS or fecal fluid prepared from fresh or frozen intestinal contents. Please click here to view a larger version of this figure.

Figure 4: Representative flow cytometry plot of IgA-coating bacteria. (A) FSS versus SSC plot. (B) Blank control. (C) APC-IgA-single staining. (D) Nucleic acid single staining. (E) IgA and nucleic acid double staining population for naive mice. Please click here to view a larger version of this figure.

Antibiotics used in the depletion procedure have different antibacterial properties. Vancomycin is specific for gram-positive bacteria³⁰. Oral doxycycline can induce significant intestinal microbiota composition changes in female C57BL/6NCrl mice³¹. Neomycin is a broad-spectrum antibiotic that targets most gut-resident bacteria³². It does not prevent intestinal inflammation, however. Broad-spectrum antibiotic cocktails are more effective than a single antibiotic^13,33,34. One study reported that triple-antibiotic treatment (amoxicillin, fosfomycin, and metronidazole) synergistically depleted the imbalanced gut microbial community in ulcerative colitis¹¹. The quadruple-antibiotic treatment successfully eliminated gut microbiota in an asthma mice model¹³ and a high fat diet mice model¹⁴. In previous studies, microbiota depletion with a single antibiotic dissolved in drinking water required more than a 6 week treatment period¹⁵. With oral gavage, the time for microbiota depletion was reduced to 2 weeks¹². However, 17 days administering an antibiotic cocktail in drinking water can lead to significant weight loss in mice due to the bitter taste of metronidazole and the subsequent decrease of water and food intake⁹. Our method, combining oral gavage and drinking water, was even more efficient in achieving satisfactory bacterial depletion in 3 days. Additionally, our antibiotic treatment regimen can avoid body weight loss in mice caused by extended an antibiotic cocktail treatment.

Avoiding bacterial contamination is an important consideration in our FMT schedule. First, it is necessary to wipe out bacteria existing in the immediate feeding environment, including cages, water, and food. Second, an aseptic environment must be maintained, including sterile tips, tubes, and surgical instruments. Test tubes and surgical instruments may contain environmental microbiota. The risk of environmental contamination can be reduced by wearing latex gloves, spraying them with 75% alcohol, and washing hands before starting a new procedure. The instruments should also be washed with 75% alcohol to avoid fecal microbiota contamination across individual mice.

Some studies have shown that coprophagic and grooming behaviors can be exploited by transient cohousing of mice, which results in the sharing of microbes across cohabitated individuals. This technique may be the simplest and most convenient method to transplant fecal microbiota³⁵. Cohousing of mice showed similar effects as FMT on body mass³⁶ and in obesity disease³⁷ and tumor models³⁸. However, the impact of cohousing on fecal microbiota is difficult to standardize, especially with F2 littermates³⁵. Moreover, this strategy is not suitable for large numbers of mice and could introduce variability. Furthermore, the time required to transplant fecal microbiota by cohousing was 4 weeks longer than that of FMT³⁹, in which the intake is identical and controllable for all mice.

FMT protocols are not standardized across studies published in the literature. Researchers use different fecal samples such as fresh and frozen stool to prepared fecal fluid and treat recipient mice with different gavage frequency and duration. Technically, fresh stool is the best choice. However, for frozen stool, no significant difference in successful transplantation rate was noted in relation to different storage periods (1 week to 6 months)^40,41. A recent study suggested that, although freezing did not significantly affect in vitro fecal bacterial viability, it reduced the colonization ability of the transplanted fecal microbiota⁴². In addition, it is notable that the genetic background⁴³ and age⁴⁴ of the recipient rodents are other important factors that significantly affect engraftment of donor microbiota. Interestingly, Ericsson et al. found that even the richness difference between the donor and the recipient’s microbiota impacts the inoculum engraftment³¹. Another study indicated the effect of transplantation frequency on bacterial colonization and community. When human microbiota is transplanted into mice, the increase of transplantation frequency from once a week to twice a week leads to different results. In this study, it was found that FMT once a week appears to be the best compromise, because it supported higher diversity in microbiota engraftment⁴⁵. This opinion was further confirmed by a study by Staley et al.⁴⁶ in which a single gavage treatment of human fecal fluid was able to establish a human microbiota-associated (HMA) animal model. Importantly, they also showed the influence of different antibiotic treatment on the engraftment efficacy of FMT. Although both “systemic” and “non-absorbable” antibiotic cocktails showed comparable effects in disrupting the indigenous microbiota to a level that permitted colonization of abundant HMA taxa, the extended antibiotic course seems more likely to contribute to consistent and extended engraftment⁴⁶. Most researchers administered 200 μL of fecal suspension (100 mg/mL) to the mice^23,45. However, transplantation duration ranged from 3 days²² to 5 weeks²³. In another study, mice receiving fecal microbiota from patients with constipation for 7 and 15 days displayed different defecation parameters⁴⁷. The transplantation duration was variable depending on the study design.

High levels of IgA-coated bacteria play an important role in the pathogenesis of inflammatory bowel disease²⁷ and arthritis²⁶. The method for measuring IgA-coated bacteria consists of pelleting bacteria by centrifugation (10,000 x g for 1 min at 24 °C), followed by pellet resuspension in 100 μL of PBS and flow cytometry²⁹. These centrifugation conditions were unsuitable for our method. To avoid loss of bacterial viability, we used a lower speed for centrifugation (800 x g). In our experiment, the IgA-coating bacteria population was hardly distinct from the negative population when staining 100 μL of fecal fluid with fluorescence-labeled antibodies for flow cytometry analysis. However, 20x dilution of fecal fluid followed with the same amount of antibody staining was sufficient to resolve this problem.

In conclusion, we established a method for rapid depletion of most dominant gut microbiota in 3 days and a procedure to use fresh or frozen donor samples in FMT. We also presented an optimized method to analyze the proportion of IgA-coated bacteria in the gut. Although our protocols cannot be directly extended to human FMT studies, they can still be helpful for studying the role of gut microbiota in the pathogenesis of many diseases in rodent models.

The authors have nothing to disclose.

This work was carried out under the sponsor of Outstanding interdisciplinary project of West China Hospital, Sichuan University (Grant Nr: ZYJC18024) and National Natural Science Foundation of China (Grant: 81770101 and 81973540).