ABSTRACT: Germ-free mice were colonized as neonates with either a simplified human infant microbiota consortium consisting of four Bifidobacterium species, or with a complex, conventional murine microbiota. Cultures were combined in equal ratios based on OD600 readings (volume equal to approximately OD600 1.0 for each of the species), pelleted, then re-suspended in 3 mL sterile PBS. Colonization of males and females with bifidobacteria was achieved by administering 0.2 mL of this treatment directly to the stomach of each dam via oral gavage on alternating days post-partum. The treatment was also administered with the drinking water. Pups were administered 0.02 mL of the treatment mixture starting at P1, and on alternating days until P20. Dilution plating of the treatment mixture on MRS plates at several timepoints during the experiment confirmed that the dams received approximately 1.1x1010 total CFUs (viable bacterial cells) per gavage. The pups received a smaller volume of the gavage mixture, approximate to 1.1x109 total CFUs per treatment. Conventionalized (colonization of germ-free mice with normal complex murine microbiota) males and females were derived by administering fecal slurries from specific pathogen free mice, which harbor a complex microbial community, via oral gavage to the dams on P1, and by adding used bedding (containing fecal pellets) to the cages on P1 and P3. Germ-free control mice received gavages of sterile 0.1 M phosphate buffered saline (PBS; Gibco) in the same manner as the experimental mice. We examined the cerebellum, cortex, and hippocampus of both groups of colonized mice in addition to germ-free control mice at P4, P10, and P20. For preparation of RNA for gene expression analysis, the mice were individually anesthetized using isoflurane until no reflex was observed. The mice were rapidly decapitated, and the brains were dissected to isolate the cerebellum, hippocampus, and cortex. The tissues were flash frozen and stored separately at -80°C until RNA isolation. RNA was isolated from the cerebellum, cortex, and hippocampus of male mice from each developmental timepoint via Trizol. The 260/280 ratio was measured for each sample using a Nanodrop spectrophotometer, and after this step ranged between 1.8 and 2.0. RNA samples were stored at −80°C until use. Two µg of RNA from 5 samples (male mice only) were pooled into one representative sample per treatment, time point, and brain region (27 pooled samples total, n=5 mice per pooled sample). Pooled RNA was purified with the Zymo RNA clean and concentrate kit according to the manufacture’s protocol with one modification. Zymo-Spin IIICG columns were used instead of Zymo-Spin IC Columns in order to purify larger quantities of RNA (~100 µg of RNA). A Bioanalyzer (Agilent) was used to verify that the purity and concentration of pooled and cleaned RNA was of sufficient quality and quantity prior to use with the synaptic plasticity qPCR array. Using the “Mouse Synaptic Plasticity RT² Profiler PCR Array” (PAMM-126Z; Qiagen), 96 genes were profiled on the 27 samples (1 pooled sample as described above for each of the 3 brain regions, at each of the 3 time points, in each of the 3 treatment groups). The manufacturer’s protocol was followed for sample preparation. Briefly, 500 ng of each pooled sample was treated with DNase to eliminate genomic DNA contamination, followed by conversion to cDNA using the RT2 First Strand Kit. The cDNA was added to the kit mastermix on ice, and distributed to the 384-well plate containing 4 replicate primer assays for the 84 pathway-specific genes. Reactions using SYBR Green chemistry (Qiagen SYBR Green qPCR Mastermix, Cat. no. 330529) were performed in a 384 well block ViiATM 7 Real-Time PCR system (Applied Biosystems, Foster City, USA). The threshold values were set consistently across all runs in the same analysis (threshold manually set to 0.5 and lower limit of detection set to 35).
