Do Primocolonizing Bacteria Enable Bacteroides thetaiotaomicron Intestinal Colonization Independently of the Capacity To Consume Oxygen?

Aerobic bacteria are frequent primocolonizers of the human naive intestine. Their generally accepted role is to eliminate oxygen, which would allow colonization by anaerobes that subsequently dominate bacterial gut populations.

10,000 rpm and the supernatant was used as E. coli culture medium. Cultures were grown for 4 -6 hours in an anaerobic jar at 37°C. Cells were then centrifuged and resuspended in PBS buffer at 4°C to obtain 1 ml bacterial suspensions at OD600 = 1.0. Oxygen consumption was monitored over 6 minutes at 37°C with aeration using a Clark-type oxygen electrode (Liquid-Phase Oxygen electrode unit DW1, Hansatech instruments, UK) as described [4], except that readings were done upon addition of 2.5 % glucose. Oxygen consumption by E. coli WT and hemA strains in the absence or presence of heme (5 µM), or by C. scindens, was measured as follows. Bacteria were first streaked on LB-ALA for E. coli WT and hemA as above; C. scindens was grown in BHI and incubated for 48 h at 37°C in an anaerobic chamber. Single colonies were used to prepare overnight anaerobic liquid cultures at 37°C in M17-glu (for E. coli) or in BHI (C. scindens). Cultures were diluted (1/100) and grown to OD600 between 0.5 and 0.7. Cultures were then removed from the anaerobic chamber and vigorously shaken to saturate the medium with oxygen. Note that here, we monitored enzymatic capacity to eliminate oxygen; survival of C. scindens was not expected with this treatment, and was not tested. Oxygen consumption was followed using a HQ40D unit with an LDO probe (HACH Company, Loveland, Co). The LDO probe was immersed in the bacterial culture maintained at 37°C and the decrease in the amount of dissolved oxygen was followed for 6 minutes. For each sample, results were normalized to 100 % as greatest oxygen content. E. coli and C. scindens cultures used for pre-colonization were prepared respectively in M17glu at 37°C in static conditions, and in BHI medium in an anaerobic chamber. Overnight cultures were diluted 1:100 and grown to mid-exponential phase. Cultures were centrifuged and resuspended to obtain 1-2 x 10 8 colony forming units (CFU) in 100 µl 0.9 % NaCl for oral administration. B. thetaiotaomicron VPI-5482 cultures used for inoculations were prepared from overnight cultures, centrifuged, and resuspended according to OD600 to the required concentration in 0.9 % NaCl in a 100 µl volume.

Colonization of germ-free mice
At 16 h after pre-colonization, pre-colonized and control mice were inoculated with B.
thetaiotaomicron (1-2x10 3 or 2x10 4 CFU); this time is called T0. To determine CFU, we first established the correlation between B. thetaiotaomicron CFUs and OD600 by plating experiments (OD600 = 1 corresponds to 2± 1.8 x 10 8 CFU, N=5). Due to technical limits of timing for live experiments in germfree isolators, we then relied on OD600 readings to calculate CFUs for inocula in all experiments. We note that unavoidable bacterial exposure to O2 occurs just prior to administration. Two precautions were taken: first, all mouse groups were administered B. thetaiotaomicron simultaneously (one technician per animal group), and second, all experiments included the reference controls. During experiments, feces were collected just before E. coli colonization, at T0, and at specified time intervals.
In dissection experiments on co-colonized animals, animals were sacrificed 72 hours after the start of experiments. Intestinal contents were recovered for CFU determinations and microscopy.

Determination of colonization efficiency
Directly after feces sampling, and for recovery of intestinal contents, samples were transferred to an anaerobic chamber and resuspended as a 1:10 dilution in 0.9 % NaCl. Ten-fold dilutions were spotted Mann and Whitney unpaired 2-tailed tests were used to analyze statistical relevance of all data analyzed from mouse studies (GraphPad Prism; Graphpad Software Inc., USA).

Microscopy
Field emission scanning electron microscopy was performed to visualize E. coli and B. thetaiotaomicron in mouse feces. Feces were collected at T48 h, and prepared for microscopy by chemical dehydration: samples were immediately immersed in a fixative solution (2.5 % glutaraldehyde in 0.2 M sodium cacodylate buffer, pH 7.4), deposited on sterile cover-glass disks (Marienfeld, VWR, France) and stored 1 h at room temperature, then overnight at 4°C. The fixative was removed, and samples were rinsed three times for 10 min in sodium cacodylate buffer. Samples underwent progressive dehydration by soaking in a graded series of ethanol (50 % to 100 %) baths before drying under CO2. Samples were mounted on aluminum blocks (10 millimeter diameter) with conductive silver paint and sputter coated