Mammalian/Microbial Co-cultures: The epithelial cultures were composed of CBBE-1 (ATCC) enterocytes (Passage 33–50) and HT29-MTX (Sigma Aldrich) mucus producing cells (Passage 27–48) seeded on 24-well Transwells→ with a 0.4 um pore size PET membrane (Falcon™) in a 9:1 ratio at a density of 1x105 cells per cm2. Cells were cultured with 300uL and 700uL in the apical and basolateral compartments, respectively, of 1x Advanced DMEM (Gibco) supplemented with 10% heat inactivated fetal bovine serum (Atlanta Biologicals), 1x GlutaMAX™ (Gibco), and 1% Penicillin/Streptomycin (Gibco). Dendritic cells were differentiated from human peripheral blood mononuclear cells (STEMCELL™ Technologies, isolated from whole blood according to supplier protocols) for seven days in Advanced RPMI media (Gibco) containing 35 ng/mL Interleukin-4 (Sino Biological), 10 nM retinoic acid, and 50 ng/mL GM-CSF (Gibco). After 14 days of CBBE-1/HT29-MTX epithelial culture, dendritic cells were seeded on the underside of the culture membrane at a concentration of 1x105 cells per cm2 to provide the model’s immune component, and given 2 hours to attach. The cell cultures were then incubated with 300uL and 700uL in the apical and basolateral sides, respectively, of a serum-free media formulation of 1x Advanced DMEM (Gibco) supplemented with 1x GlutaMAX™, 1% Penicillin/Streptomycin (Gibco), and 0.1% Insulin, Transferrin, Selenium cocktail (Gibco) for six additional days. Twenty-four hours prior to introducing the bacteria to the mammalian cultures, the basolateral medium was replaced with 700uL of the same serum-free DMEM-based formulation but without antibiotics, and the apical serum-free DMEM was replaced with 300uL of Dulbecco’s phosphate buffered saline + Ca2++Mg2+ (PBS, Corning) buffered with 10mM of HEPES (Gibco). Single bacterial species were grown overnight from frozen stocks in MOPS microbial media and counted via hemocytometer. The bacterial stocks were then centrifuged at 6000 rpm for 5 minutes, and the MOPS media replaced with the PBS-HEPES apical media formulation to achieve 6.7x104 bacteria per mL of apical media.
The bacteria were added to the apical compartment of the Transwells→ and given 2 hours to attach. After 2 hours, all of the apical and basolateral media was removed and replaced with PBS-HEPES apical media and serum-free, antibiotic-free DMEM, respectively. Cultures were maintained at 37 C and 5% CO2 for 48 hours. Medium in both compartments was changed every 3 hours during the day, with two 9-hour overnight periods, during the 48-hour culture period. After 48 hours, the monolayer health was examined via transepithelial electrical resistance (TEER) measurement, Lucifer Yellow permeability assay, and live/dead imaging, as described below.
Experiments to determine if N-acetyl cysteine (NAC) can improve survival of the mammalian monolayers were performed in an identical manner save for the addition of 3mM N-acetyl cysteine in the apical media starting 24 hours prior to the introduction of microbes and carrying through to the end of the experiment. Experiments conducted with and without NAC supplementation were repeated 5 times with 2 biological replicates per experiment.
Spent Media Studies
To identify damaging activity of agents other than live, active bacteria in the co-cultures, spent media studies were conducted by exposing mammalian monolayers composed of CBBE-1, HT29-MTX, and human dendritic cells prepared as described above to the collected apical media from the 2, 17, 26, and 50-hour time points of the co-culture experiments. The spent media samples were sterilized by centrifugation at 6000 rpm for 5 minutes to pellet any microbes or cellular debris, followed by filtration through 0.22 um mesh filters. The sterile spent media was then added to the apical compartment of the CBBE-1/HT29-MTX cultures prepared as described above, with the basolateral compartments containing fresh serum-free DMEM. The cultures were exposed to the spent media for 24 hours, then examined via TEER measurement, Lucifer Yellow permeability assay, and live/dead staining, as described below.
Experiments to determine if N-acetyl cysteine can improve survival of the mammalian monolayers upon exposure to spent media were performed in an identical manner using E. coli spent media supplemented with the addition of 3mM of NAC. All spent media experiments were conducted with 3 biological replicates for each group.
TEER measurements were performed using a World Precision Instruments EVOM2 epithelial volt/ohm meter and a World Precision Instruments Endohm cell culture cup. The EVOM2 was calibrated using the reference electrode before every set of measurements, and the Endohm was sterilized between measurements per the manufacturer instructions. Measurements were taken within 5 minutes of removing the cultures from the incubator to minimize the influence of temperature change.
Lucifer Yellow Permeability
Prior to the addition of the microbes or spent media, and again at the conclusion of the experiment, a 100 uM Lucifer Yellow solution in apical media was added to the apical compartment and incubated for 1–2 hours. After incubation, samples of the basolateral media were collected and their fluorescence intensities measured using a fluorescence plate reader with excitation at 428 nm and detection at 536 nm. The fluorescence intensity was then used to calculate the concentration of Lucifer Yellow using a standard curve, and the apparent permeability was calculated according to the equationwhere V is the volume of the basolateral compartment in mL, A is the area of the cell layer in cm2, Ci is the initial concentration of the Lucifer Yellow added in uM, Cf is the final concentration of Lucifer Yellow in the basolateral compartment after incubation, and T is the time of incubation in seconds.
Live/dead staining was carried out according to the protocol for the Invitrogen LIVE/DEAD Viability/Cytotoxicity Kit for mammalian cells. In brief, an aqueous solution of 4 uM ethidium homodimer-1 and 2 uM calcein-AM was prepared and added to the apical compartment after the cultures were washed twice with PBS. After incubation for 15 minutes at 37 C, the stain was removed, and the cultures were washed once. Fresh PBS was then added to the apical compartment to prevent drying out as the samples were imaged.
Oxygen tension in both apical and basolateral compartments was monitored using Lucid Scientific’s ruthenium based optical probes and logging software21. The two probes were fixed 3mm from the apical surface and at the same depth as the basolateral surface of the cell culture insert membrane, respectively. Oxygen tension measurements were continuously collected in 5.1 second intervals during the 9 hour overnight periods.
The bacterial numbers in the co-cultures were measured via spotting of the apical and basolateral media on fastidious anaerobe agar plates incubated in an anaerobic chamber and counting the colonies that arose. The timepoints chosen for microbe spotting were 2 hours, 17 hours, 26 hours, 41 hours, and 50 hours after bacterial inoculation, as those corresponded, respectively, to immediately after the inoculation period, the end of the first overnight period, experimental midpoint, end of the second overnight period, and experimental conclusion. At the conclusion of experiments, following the collection of the apical media, the monolayer was scraped from the Transwell→ membrane and homogenized in 300 uL of sterile apical media. This homogenate was then plated on agar to determine the concentration of bacteria that attached to the mammalian cells over the course of the experiment. All samples were diluted in sterile PBS to ensure plates could be accurately counted.
Lipopolysaccharide Measurement: The concentration of LPS in the apical compartment was measured using a Limulus amebocyte lysate (LAL) chromogenic endotoxin quantification kit (Pierce). The assay was performed according to the manufacturer’s procedure. In brief, 50 uL of centrifuged and filtered sample media and known standards were added to a 96 well plate kept at 37 C. These samples were incubated with 50 uL of LAL reagent for 10 minutes and 100 uL of chromogenic solution for 6 minutes at 37 C. After incubation, 100 uL of stop solution (25% acetic acid) was added to each well, and the absorbance was then measured at 405 nm. The concentration of LPS was determined from a standard curve using LPS isolated from E. coli O111:B4 as a standard. LPS measurements were conducted using media samples collected at the same timepoints as described above for analysis of bacterial growth.
Reactive Oxygen Species Measurement: Reactive oxygen species (ROS) measurements were performed via a chemiluminescent Acridan assay modified from the procedure described by Uy et al.22. Briefly, 50 uL of the apical media samples and 50 uL of PBS were combined in a 96 well plate. Next, solutions A and B from the G.E. Healthcare Amersham ECL Plus Western Blotting Detection Reagent kit were mixed in a 40:1 ratio as specified by the manufacturer; 50 uL of this mixed detection solution was then added to each sample, and the plate was incubated in the dark for 5 minutes at room temperature. The plate’s luminescence was read on a Biotek Synergy 2 plate reader, and the concentration of ROS determined using a standard curve obtained by measuring the luminescence of hydrogen peroxide in an excess of horse radish peroxidase (Sigma) incubated with the detection solution as described by Zhu et al23. Measurements were taken within 45 minutes of media collection to minimize ROS decomposition. The effect of antioxidants on the concentration of ROS in the collected media was explored by adding 3, 5, and 10 mM of N-acetyl cysteine to the collected media sample-PBS mixture 30 minutes prior to adding the detection solution. ROS analyses were conducted using media samples collected at the same timepoints as described above for analysis of bacterial growth.
The concentration of nitrites was measured in the collected media as a surrogate for reactive nitrogen species (RNS) concentration using a Griess reagent kit from Abcam→. The measurements were carried out according to the manufacturer’s instructions. Briefly, medium samples were combined with the Griess reaction mix (Griess Reagent I and II, nitrite assay buffer) and incubated for 10 minutes before measuring the absorbance of the samples at 540 nm. Concentrations of nitrite were calculated according to a standard curve generated from the included nitrite solution. Nitrite concentration measurements were conducted using media samples collected at the same timepoints as described above for analysis of bacterial growth.
Lipid Peroxidation Assay
Lipid peroxidation is an indicator of ROS-mediated cell damage, and was measured using a kit from Abcam→ for the detection of malondialdehyde (MDA), a commonly used biomarker for lipid peroxidation in cell media24. In brief, apical media samples and standards were placed in a 96-well plate and incubated for 30 minutes at room temperature with the MDA color reagent, and then incubated for another 60 minutes with the reaction solution. The 96-well plate was read at 695 nm absorbance, and the concentration of MDA in the samples was calculated using a standard curve generated from the included MDA solution. Lipid peroxidation analyses were conducted using media samples collected at the same timepoints as described above for analysis of bacterial growth.
Statistical analysis was conducted via ANOVA and post-hoc multiple comparisons via Tukey’s honestly significant difference criterion for analyzing differences between groups at the same timepoint or within a group at more than two timepoints. Paired sample t-tests were used to determine differences within the same group at two timepoints. All statistical calculations were carried out in MATLAB®.