Method Details

Prior to transport, mice at Charles River, Envigo, Taconic and Jackson were fed (according to online rodent model information sheets) Purina 5L79 rodent chow, Teklad Global Rodent Diet 2018S, NIH #31M Rodent Diet, and LabDiet 5K52 formulation (6% fat), respectively. Upon arrival, mice from each cohort were randomly assigned into individually ventilated cages on one rack at a housing density of 3 to 4 animals per cage and allowed to acclimate in our vivarium for at least a week undisturbed. Feed was switched to irradiated TEKLAD GLOBAL 18% protein rodent diet 2918 (Envigo) and no breeding was performed. 70% ethanol was used to disinfect surfaces and gloves between groups. Clean (but not sterile) paper towels were utilized for fecal sample collections. None of the experiments performed in this study involved treatment or pretreatment of mice with antibiotics.

Freshly voided pooled feces (approximately 200mg per cage) were collected from C57BL/6NCrl, C57BL/6NHsd, C57BL/6NTac, and C57BL/6J mice. Samples were maintained on ice and processed within 1 hour. An anaerobic chamber was not used. Feces were diluted at 1:50 (approximately 10mL) in sterile Phosphate Buffered Saline (PBS; pH = 7.4) and homogenized via vortex for 5 minutes at room temperature. Tubes were briefly centrifuged (500 x g for 1 minute) to pellet large particles. 500μl aliquots were mixed with 500μl glycerol (60%w/v) and immediately stored in −80°C.

Mice were infected upon reaching eight to twelve-week of age. To generate contamination-free S. Typhimurium culture, LB broth supplemented with appropriate antibiotics in sterile flasks was aseptically inoculated from single colonies and incubated with shaking (200 rpm at 37°C) for 16 to 20 hours. Bacteria were then harvested by centrifugation (15 min, 4000 × g, 4°C), and adjusted to a density of 1 × 10¹⁰ CFU/mL in sterile LB. To generate lower density inocula, cultures were serially diluted ten-fold. Sterile LB broth (0.1 mL) containing S. Typhimurium at the indicated densities was inoculated by oral gavage (between 6am and noon). Primary experimental outcome assessed were: animal activity, weight loss, lethal morbidity and pathogen burdens. Weights were recorded daily and feces were collected at the indicated time points (between 1 and 7 days after infection). Mice were euthanized at the indicated time points or when they became moribund. Mice that were euthanized early due to health concerns were excluded from analysis, except for experiments determining lethal morbidity after challenge. Intestinal contents (approximately 15mg/animal) were homogenized in 1mL PBS. Samples were then serially diluted in PBS and plated on selective agar in order to distinguish pathogen CFU/g. Challenged mice were considered infected when a positive stool culture was obtained at >3 days after infection.

Animals were housed in flexible front glove-box isolators (Park Bioservices, LLC). Food (irradiated 2920X Teklad diet by Envigo) and autoclaved water access was ad libitum. Bedding changes were provided weekly. The room was maintained on a 12-h light/dark cycle. Principles of the established ‘out-of-the-isolator’ gnotobiotic husbandry system ²⁸ were utilized and validated in our facility to preserve gnotobiotic status for up to 2 weeks by standard quality control (negative aerobic and anaerobic culture in brain-heart infusion broth (BHI) and 16S rRNA quantitative real-time PCR comparable to a no template control). Littermates of both sexes were removed from the isolator at 6 weeks of age and ascetically transferred to a biosafety cabinet and placed into autoclaved individually ventilated cages. Groups of 2 to 4 males or females were randomly assigned to experimental groups and housed together to facilitate social interactions and establishment of their fecal microbiota transplants. Animals were maintained in a biosafety cabinet for at least 5 days undisturbed after receiving fecal transplants and infections were carried out in the biosafety cabinet.

Previously frozen pooled fecal samples from cages of each C57BL/6 substrain were thawed on ice and delivered via oral gavage (200μl) once. Colonization was allowed to proceed for a minimum of 5 days before sample collection and infection with S. Typhimurium. Bedding changes were performed before infection, carried out under strictly aseptic conditions, and samples were collected from individual mice using autoclaved beakers before infection.

Concentrated overnight cultures were prepared as described above. Animals were colonized with fecal microbiota transplant for 5 days. Each ex-germ-free mouse received 0.1 mL of a suspension containing the indicated CFU of S. Typhimurium via oral gavage. Fresh fecal pellets were collected aseptically using autoclaved beakers and plated on agar plates containing the appropriate antibiotics.

Fecal samples were collected from individual mice prior to infection for analysis and frozen at −20 C. DNA was extracted using the MoBio PowerSoil Kit according to the manufacturer instructions, with the following two recommended modifications. (1) Samples were heated at 70°C for 10 minutes following addition of Buffer 1 to enhance disruption of gram-positive bacteria. (2) Samples were homogenized using a mini bead beater for 60 seconds to achieve greater mechanical lysis of bacterial cells. Paired end library construction was performed as previously described ¹³. Primers 515F and 806R (Supplementary Table 2) were used to amplify the V4 domain of the 16S rRNA. Both forward and reverse primers contained a unique 8 nt barcode (N), a primer pad (underlined), a linker sequence (italicized), and the Illumina adaptor sequences (bold). Each sample was barcoded with a unique forward and reverse barcode combination. PCR contained 1 Unit Kapa2G Robust Hot Start Polymerase (Kapa Biosystems), 1.5mM MgCl2, 10 μmol of each primer, 10mM dNTP’s, and 1ul of DNA. PCR conditions were: an initial incubation at 95°C for 2 min, followed by 30 cycles of 95°C for 20 s, 50°C for 20 s, 72°C for 20 s and a final extension of 72°C for 3 min. The final product was quantified on the Qubit instrument using the Qubit High Sensitivity DNA kit and individual amplicon libraries were pooled, cleaned by Ampure XP beads (Beckman Coulter), and sequenced using a 250 bp paired-end method on an Illumina MiSeq instrument in the Genome Center DNA Technologies Core, University of California, Davis. Raw paired end-sequence data were de-multiplexed and trimmed, followed by filtering for quality. Samples containing less than 1000 quality reads were removed from dataset. Quantitative Insights into Microbial Ecology (QIIME) open source software package (version 1.9) was initially used to perform sequence alignment and closed-reference operational taxonomic units (OTUs) picking against the Greengenes reference collection (version 13_8) at 97% identity. Data were later re-analyzed when QIIME 2 became available (core 2018.2 distribution). Clustering, permutational multivariate analysis of variance (PERMANOVA), beta diversity measures (principle coordinate analysis of unweighted UniFrac distances) and phylogenetic profiling were similar between QIIME 1.9 and QIIME 2. Additionally, taxonomy was also assigned with the Ribsosomal Database Project (RDP release 11) classifier using the DADA2 R package version 1.6 pipeline (https://benjjneb.github.io/dada2/tutorial.html) to resolve more sequences down to the species level.

Mice were received at 5–7 weeks of age. After allowing one week to acclimate in our vivarium, mice from Charles River Laboratories and The Jackson Laboratories were placed in clean cages (with new food and water) at a ratio of 1:1 and cohoused for 14 days. Ear punching facilitated long-term identification. In one experiment, cohoused mice were subsequently separated after two weeks. Fecal samples were flash frozen at the start of cohousing and every week thereafter. For challenge experiments, animals remained cohoused after receiving 1 × 10⁸ CFU Salmonella. All mice were sacrificed at 5 days after infection.

Ten-fold serial dilutions of previously frozen fecal transplants in glycerol or flash frozen feces were plated on MacConkey agar and incubated aerobically at 37°C overnight. Single colonies with discernable unique morphologies were picked and streaked for isolation. Colonies were initially typed based on colony appearance into lactose-fermenting red colonies (Lac⁺) and lactose non-fermenting white colonies (Lac^-). Approximately 100 colonies (30–40 colonies from each of the three vendors harboring Enterobacteriaceae) were isolated and identified biochemically using the EnteroPluri Test (BD) per the manufacturer’s directions. Isolates used for animal experiments were subjected to full-length 16S rRNA gene Sanger sequencing.

Fecal pellets from individual C57BL/6NCrl, C57BL/6NHsd, C57BL/6NTac, and C57BL/6J were plated on Difco Lactobacilli MRS Agar (10 g/L Protease Peptone No. 3, 10 g/L Beef Extract, 5 g/L Yeast Extract, 20 g/L Dextrose, 1 g/L Polysorbate 80, 2 g/L Ammonium Citrate, 5 g/L Sodium Acetate, 0.1 g/L Magnesium Sulfate, 0.05 g/L Manganese Sulfate, 2 g/L Dipotassium Phosphate, 15g/L Agar; pH 5.5). Feces were serially diluted in PBS and plates were pre-reduced for 2 days. Incubation was allowed for 3 days at 37°C in an anaerobic chamber (Shel Lab Bactron II; 5% Hydrogen, 5% Carbon Dioxide, 90% Nitrogen).

Bacterial strains were grown in LB broth and pelleted. After resuspension in distilled water, tubes were boiled for 1 minute and immediately cooled on ice. 1μl of the lysed bacterial suspension containing genomic DNA was used in a PCR reaction to amplify the 16S rRNA gene using primers 63F-1387R. PCR conditions: 1μl template, 1μl each forward and reverse primers, 17μl PCR Supermix high-fidelity. Thermocycler (MJ Research PTC-200 Peltier Thermal Cycler) conditions: thirty cycles of denaturation at 95°C for 1 minute, annealing at 48°C for 1 minute, elongation at 72C for 2 minutes, followed an additional 15 minutes at 72°C. Sizes (1.5 kb) were estimated using agarose gel electrophoresis. Following the QIAquick PCR purification kit, 1μl of DNA was ligated into pCR2.1-TOPO vector. Chemically competent TOP10 E. coli cells were transformed with ligation products by heat shock (30 s at 42°C). Recombinant cells were selected on LB plates containing kanamycin and X-Gal (5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside). DNA preparations for sequencing were made with a QIAprep mini as specified by the manufacturer. Plasmids were eluted with 50 μl of water, and the products were stored at −20°C. Sizes (5.4 kb) of plasmids containing inserts were checked by agarose gel electrophoresis and concentrations (200 ng/ul) were determined on a NanoDrop 1000 spectrophotometer (Thermo Scientific). Plasmid inserts were sequenced by the College of Biological Sciences ^UCDNA Sequencing Facility using M13 primers. Two contigs were aligned for each strain using Snapgene. Evolutionary analyses were conducted in MEGA7 ²⁹. The evolutionary history was inferred by using the Maximum Likelihood method based on the Tamura-Nei model ³⁰. The analysis involved 10 nucleotide sequences. All positions containing gaps and missing data were eliminated. There were a total of 1357 positions in the final dataset.

Isolates from frozen fecal transplants of C57BL/6NCrl, C57BL/6NHsd, C57BL/6NTac, and C57BL/6J on MacConkey agar were maintained individually. Cultures were grown in LB broth overnight and 1 × 10¹⁰ CFU/ml preparations were made. In some experiments, strains were mixed in equal volumes to prepare CR-mix (E. coli Crl141, E. coli Crl142, P. mirabilis Crl143), Har-mix (E. coli Hsd145, E. cloacae Hsd146), and Tac-mix (K. oxytoca Tac148, P. vulgaris Tac149). A single dose of 100 μl was delivered via gavage to C57BL/6J mice. Colonization levels were determined by plating feces on MacConkey agar. E. coli Nissle 1917 and the E. coli Nissle 1917 cydAB mutant were transformed with plasmids (pCAL61 or pCAL62) prior to inoculation of mice to facilitate recovery from the feces.

For supplementary figure panels 9a-c, an overnight liquid culture of S. Typhimurium was diluted in PBS to a concentration of 1 × 10³ CFU/mL and 100 μl aliquots were added to 900 μl of mouse fecal homogenate in glass tubes with loose-fitting caps. Feces was processed as follows: Approximately 100 to 300 mg of feces were collected from one cage of mice pooled. Samples were diluted 1:100 in PBS and homogenized by vortex for 10 minutes at room temperature. Large particles were pelleted by centrifugation at 500g for 1 minute. Half of the sample was filtered through a 0.45 μm low-protein binding membrane. Fecal homogenates were stored at 4°C for up to 1 hour prior to adding S. Typhimurium. Tubes were then incubated for 24 hours at 37°C while shaking under atmospheric oxygen. For supplementary figure panels 9d-e, pooled feces from C57BL/6J mice was diluted 1:100 in PBS, homogenized, briefly centrifuged, stored at 4°C for up to 1 hour prior to adding S. Typhimurium to achieve a final concentration of 1 × 10³ CFU/mL fecal homogenate. 900 μl aliquots of Jackson feces containing S. Typhimurium were distributed to glass tubes with loose-fitting caps. Overnight cultures of each commensal Enterobacteriaceae strain were diluted in PBS to a concentration of 1 × 10⁴ CFU/mL and 100 μl was added to tubes containing 900 μl of S. Typhimurium in fecal homogenate. For anaerobic conditions, a duplicate set of tubes was prepared and moved into the anaerobic chamber (Shel Lab Bactron II; 5% Hydrogen, 5% Carbon Dioxide, 90% Nitrogen) immediately and incubated without shaking at 37°C. CFU/ml were determined by selective plating of serial dilutions.

The E. coli Nissle cydAB mutant was constructed as described previously ¹⁹. Complementation was achieved by introducing cydAB on a plasmid under expression of the native promoter as follows. Primer pair “EcNcomp Fwd” and “EcNcomp Rev” (Supplementary Table 2) was used to amplify the coding sequence of E. coli Nissle 1917 cydAB operon plus 500bp upstream and downstream. The resulting PCR product was visualized via agarose gel electrophoresis and purified using Zymoclean Gel DNA Recovery kit (Zymo Research). The fragment was then cloned into the EcoRV restriction site of the low copy number plasmid pWSK129 ³¹ to generate the plasmid pBMM17 via Gibson Assembly Master Mix (NEB).