Manganese oxide nanoparticles (MnO NP (US Research Nanomaterials, Inc., Houston, TX, USA)) used in all experimental procedures were kindly provided by Dr. A. Kanthasamy (Department of Biomedical Sciences, Iowa State University, Ames, IA). Based on transmission electron microscopy the size of MnO NP were 16.8 ± 2.6 nm in diameter.

The MODE-K intestinal epithelial cell line generated from the small intestine of a C3H mouse was used in all experimental procedures described below (kind gift from Dr. C. Elson, Division of Gastroenterology and Hepatology, School of Medicine, University of Alabama at Birmingham, AL). In this study, MODE-K cells were grown in high glucose (4.5 g/L) DMEM medium containing 10% FBS, 2 mM L-glutamine, 50 units of penicillin, and 50 µg/mL streptomycin. In general, cells were inoculated into a tissue culture plate (6–96 wells) depending on the experimental requirements and were cultured overnight in a humidified atmosphere of 5% CO ₂ at 37°C. The length of each experiment and MODE-K cell density for each experiment has been provided in the methods section for each experimental procedure. For the majority of experiments, there were four treatment groups: control, 1 µg/mL E. coli LF82 lysate, 10 µg/mL MnO NP, and 10 µg/mL MnO NP plus 1 µg/mL E. coli LF82 lysate. At 24 hours before initiating experimental treatments, the culture medium was switched from high-glucose medium to low-glucose DMEM medium. Dose titration experiments were performed to identify the minimal dose of E. coli LF82 lysate and the MnO NPs that did not affect cell viability after 24 hours, and these doses were chosen to assess the interactions between the MnO NPs and E. coli LF82 lysate.

An adherent invasive E. coli (AIEC) strain LF82 that has been associated with ileal Crohn’s disease was cultivated in BHI broth at 37°C in an aerobic environment. Bacteria were harvested after overnight replication, washed with PBS, frozen, lyophilized, and stored at -20°C until used. The lyophilized bacterial cells were weighed and resuspended in sterile PBS (2 mg/mL) prior to being subjected to two freeze-thaw cycles, followed by sonication at 50 amps, 75 amps, and 100 amps for 30 seconds at each amperage. This sonication cycle was repeated three times. The lysates were sterilized by ultraviolet irradiation and sterility confirmed bacteriologically. With minor modifications, this method was used as previously described ²⁴ .

Cell viability was measured using the CellTiter 96 Aqueous Non-Radioactive Cell Proliferation (MTS assay kit from Promega Corporation, Madison, WI). MODE-K cells were inoculated into 24-well plates with 120,000 cells/mL per well one day prior to treatment. The next day, MODE-K cell monolayers were washed twice with low glucose DMEM before addition of the respective stimulants diluted in low glucose DMEM medium and cultured for an additional 48 hours. Following treatment, 20 µL of MTS solution reagent mix was added to each well and incubated at 37°C for 30 minutes. At the end of incubation, readings were taken at a wavelength of 490 nm and another reference reading for each well was taken at 670 nm to eliminate background. This method was previously described and adopted to fit our experimental procedure ²⁵ . Data shown is represented as the mean ± SEM collected from two independent experiments.

Prior to assessing cell viability, supernatants were collected from MODE-K cell cultures at 24 and 48 hours after addition of the MnO NP and/or E. coli LF82 lysate and analyzed for the presence of cytokines. A multiplex bead-based assay was used to measure the presence of IL-6, TNFα, and IFNγ. The assay was performed using Bio-Plex Pro COOH magnetic beads (Luminex Corp., Austin TX) conjugated with the following monoclonal antibodies (Fisher Scientific Co. LLC) to capture IL-6 (Cat # 14-7061-85; clone MP5-20F3), TNFα (Cat # 14-7313-81, clone AN-18), or IFNγ (Cat. # 14-7325-85; clone 1F3F3D4). Biotinylated monoclonal antibodies (Fisher Scientific Co. LLC) were used for detection of the bound (i.e., captured) IL-6 (Cat. # 13-7062-85; clone MP5-32C11), TNFα (Cat # 13-7326-85; clone MP6-Xt3), and IFNγ (Cat. # 13-7312-85; clone R4-6A2). Mean fluorescence intensity was measured using Strep-Avidin conjugated phycoerythrin (PE) (Fisher Scientific Company, LLC) and a Bio-Plex System 200 (BioRad Life Science, Hercules, CA). Standard curves were generated using commercially available cytokine standards (Peprotech, Inc., Cranbury, NJ) and Bio-Plex Manager Software Version 6.0 (Bio-Rad Laboratories, Inc., Carlsbad, CA, USA). Results are representative of two independent experiments containing six replicates per treatment.

For one-dimensional migration assays, 300,000 MODE-K cells per well were suspended in DMEM (low glucose, 1.0 g/L) inoculated into 12 well plates and incubated at 37 °C in 5 % CO ₂ overnight. The next day a (+) shaped scratch (i.e., monolayer wound) was created in each well using a P-200 pipette tip, creating a cell free gap that, in time, would be covered by replicating/migrating MODE-K cells. Treatments were then added and cells were then incubated for 24 hours. Images were taken at 0 and 24 hours post-wounding. The four treatment groups were: medium alone (i.e., control), 1 µg/mL E. coli LF82 lysate, 10 µg/mL MnO NP, and 10 µg/mL MnO NP plus 1 µg/mL E. coli LF82 lysate. After 24 hours, experiments were terminated. Images were taken using a 5x lens on an inverted microscope Leica DMi1 (Leica Microsystems, IL, US) using the software Leica Application Suite 4.5.0 (Leica Microsystems, IL, US) from 10 representative wounded areas per treatment. Images and gap spacing were then measured and analyzed with an automation and image analysis software Metamorph 5.0.7 (Molecular Devices LLC, CA, US). (ImageJ is a freely available alternative that can be used for the same purposes as Metamorph.) Gap closure was quantified by determining the mean distance of closure (i.e., healing) of the gap. This was done by measuring the area of the cell-free space separating the two edges of the cellular monolayer at 24 hours after addition of the treatments, and then subtracting this value from starting value that was measured at the 0-hour time point. This method was previously described and adopted to fit our experimental procedure ²⁶ . Results are a representation of two independent experiments containing six replicates per treatment.

Mitochondrial superoxide generation was measured using MitoSOX ^TM Red Mitochondrial Superoxide Indicator (Thermo Scientific). The four treatment groups were: medium alone, 1 µg/mL E. coli LF82 lysate, 10 µg/mL MnO NP, and the 10 µg/mL MnO NP plus 1 µg/mL E. coli LF82 lysate. MODE-K cells were plated in 96-well plates with 8,000 cells/well one day prior to treatment and allowed to attach overnight. The next day, cells were washed twice with low glucose DMEM medium before addition of the indicated treatments suspended in low glucose DMEM medium and the cells were incubated for an additional 24 hours. Following treatment, MitoSox reagent (Thermo Scientific) was added to each plate well at a 1:1000 concentration. Relative fluorescence intensity was then acquired using a microplate reader per manufacturers instructions. Data is representative of two independent experiments containing six wells per treatment per experiment and is presented as the mean ± SEM.

Mitochondrial respiration was measured using a Seahorse XF24e Extracellular Flux analyzer (Seahorse Bioscience, North Billerica, MA). The Seahorse XF24e Extracellular Flux analyzer is a high-throughput instrument that yields real-time measurements of cellular respiration rates. In our studies, we were interested in assessing the effects of four treatments on mitochondrial respiration and their effects on oxygen consumption rates (OCR). MODE-K cells were seeded at a density of 300,000/mL into a T-25 culture flask and incubated overnight in 5% CO ₂ at 37.8ºC. The next day, cells were treated with respective treatments for 24 hours. Stimulated cells were recovered from the plates by scraping and added to the Seahorse plates using CellTac at 60,000 cells per well and incubated for 1 hour to allow the cells to adhere. At this time, the mito-stressor reagents oligomycin (1 µM), FCCP (2 µM) and rotenone/antimycin A (0.5 µM) were loaded into the corresponding injection ports of the Seahorse FluxPak cartridge and placed in the Seahorse analyzer to be equilibrated. After equilibration, the plate containing treated cells was then placed in the Seahorse analyzer covered with the Flux Pak cartridge. The analyzer was then programmed to measure the basal OCR readouts at five specified time intervals before progressing to inject the mito-stressors every three cycles of measuring OCR. Measuring mitochondrial respiration with the Seahorse analyzer allowed the measurement of the effect of each treatment at the three critical stages of cellular respiration. Data shown is the mean ± SEM and are representative of five independent wells.

All statistical data analyses were performed using Prism 4.0 (GraphPad Software, San Diego, CA). Data were analyzed using one-way ANOVA with the Tukey post-hoc test for comparing all treatment groups with that of the control. Multiple comparisons test was also performed to analyze differences between treatment groups. Differences with p < 0.05 were considered significant. All data is representative of two independent experiments containing six wells per treatment.

The induction of reactive oxygen species (ROS) may negatively impact the crucial barrier function of the intestinal epithelium. Reactive oxygen species are increased in the inflamed mucosa of IBD patients and may contribute to loss of intestinal barrier function ²⁷ . Furthermore, during inflammation, oxidative stress can lead to the induction of cell injury and apoptosis of intestinal epithelial cells ²⁸ .

To assess the induction of mitochondrial ROS in the presence of MnO NPs, the MitoSOX assay was used. After 24 hours of incubation with the indicated treatments, measurements were taken and data was normalized as percent control ( Figure 1). Although there was no significant difference, there was approximately a 15% or 25% increase in mitochondrial superoxide generation in treatment groups receiving MnO NP or MnO NP + bacterial lysate, respectively ( Figure 1). While there was a slight additive effect, this data suggests that mitochondrial superoxide generation by the MODE-K cells was primarily driven by the presence of MnO NP as opposed to the E. coli LF82 lysate.

Damage and impairment of the intestinal epithelial barrier may result in an increased permeability and absorption of toxic and phlogistic factors from the intestinal lumen leading to inflammation and disequilibrium of epithelial homeostasis ²⁹ . This inflammatory response can lead to the induction of cellular injury and apoptosis of intestinal epithelial cells ^{28, 29} . To assess the secretion of inflammatory cytokines from the MODE-K cells, culture supernatants were harvested 24 hours after the addition of the treatments. IL-6 was the only cytokine detected (i.e., no TNFα or IFNγ) and this was associated with exposure to the E. coli LF82 lysate as opposed to the MnO NP alone. However, a significant in IL-6 secretion was detected when the MODE-K cells were treated with both the E. coli LF82 lysate + MnO NP ( p < 0.05) ( Figure 2A). Similar results were also obtained in culture supernatants collected after 48 hours of incubation ( Figure 2B); however, there was a threefold increase in the secretion of IL-6 at 48 hours when the MODE-K cells were exposed to both stimulants.

Wound healing is a crucial function of the intestinal epithelium as it can be injured by toxic luminal substances, inflammation, interactions with microbes, oxidative stress, and pharmaceuticals ^{27, 30} . Upon injury, the intestinal epithelium undergoes a wound healing process which is dependent on the balance of three cellular events; restitution, proliferation, and differentiation of immature epithelial cells adjacent to the wounded area ³¹ . To assess the effects of MnO NP on the wound healing process (i.e., replication and migration), a scratch wound assay using MODE-K cell monolayers was performed. Using a one-way ANOVA, a significant effect was observed between treatment groups and controls (p < 0.05) ( Figures 3A and 3B).

Post-hoc analysis indicated a significantly (p < 0.05) slower wound healing response when the control cells (medium alone) was compared to MODE-K cells treated with either 10 µg/mL MnO NP alone or 10 µg/mL MnO NP + 1 µg/mL E. coli LF82 lysate. In contrast, the addition of the E. coli LF82 lysate alone to the cultures had no effect on the replication and migration of the MODE-K cells. However, analysis revealed a significant (p < 0.05) decrease in wound healing when cells were treated with 10 µg/mL MnO NP + 1 µg/mL E. coli LF82 lysate compared to the 10 µg/mL MnO NP alone treatment group suggesting a synergistic effect on wound healing.

Mitochondrial function plays a critical role in maintaining intestinal stemness and homeostasis ³² . As mitochondrial dysfunction is associated with a loss of proliferative capacity by intestinal epithelial cells, the effects on crucial factors of mitochondrial respiration were measured using a mitostress test. Utilizing the Seahorse XFe24 Analyzer, changes in basal respiration, maximal respiration, proton leak, ATP production, and spare respiration capacity were measured ( Figure 4).

Basal respiration is a measurement that shows the energetic demand of the cell under baseline conditions. Analysis showed a significant (p < 0.05) elevation of basal respiration in both treatments groups containing MnO NP when compared to medium and E. coli LF82 lysate alone, with the highest increase induced by the MnO NP alone ( Figure 5).

Maximal mitochondrial respiration, which is a measurement of the maximum rate of respiration that cells can achieve when operating at maximum capacity was alsao measured. Significantly (p < 0.05) lower levels of maximal respiration were seen in cells treated with the MnO NP alone and an even more severe affect was observed in the MnO NP + E. coli LF82 lysate treated group ( Figure 6). While not significant, MODE-K cells treated with both MnO NP + E. coli LF82 lysate, experienced a greater decrease in maximal respiration indicating that the presence of bacterial lysate further impacted mitochondrial respiration.

Another parameter measured in the mitostress test (MST) is mitochondrial proton leak. Mitochondrial proton leak can have a major impact on mitochondrial oxidative phosphorylation coupling efficiency and production of reactive oxygen species. The presence of MnO NP caused a significant increase (p < 0.05) in mitochondrial proton leak either alone or in combination with the E. coli LF82 lysate ( Figure 7).

ATP production was also measured during the MST and it was observed that MnO NP induced higher oxygen consumption rates, as evidenced by the significant (p < 0.05) increase in ATP production when compared to control and MODE-K cells treated with the E. coli LF82 lysate alone. In addition, the presence of the bacterial lysate appeared to attenuate the metabolic activity induced MnO NP alone ( Figure 8). Any treatment group containing MnO NP experienced an increase in ATP production suggesting this to be a MnO NP-mediated effect.

The final parameter measured in the MST was spare respiratory capacity, which measures the capability of the cell to respond to an energetic demand, which can be an indicator of the cell’s fitness or flexibility. When compared to control MODE-K cells, any group treated with MnO NP had a significant (p < 0.05) decrease in the spare respiratory capacity suggesting that a decrease in respiratory spare capacity was mediated by the presence of the MnO NPs independent of the presence of E. coli LF82 lysate ( Figure 9).

Mitochondrial dysfunction is associated with a loss of proliferative capacity in intestinal epithelial cells ³² . As the MTS assay takes into account three parameters (cytotoxic, cytostatic, and antiproliferative effects), any decrease in the presence of viable cells may be a result from either one or a combination of these three parameters and would provide crucial insight about cell viability and health. After 48 hours of incubation, a significant decrease was observed in the MnO + E. coli treated MODE-K cells ( p < 0.05) ( Figure 10) which is consistent with the wound healing data shown in Figure 3. There was no significant impact measured by the MTS when the cells were treated with either the MnO NP or E. coli LF82 lysate alone.

Open Science Framework: Cytotoxic effects of manganese oxide nanoparticles in combination with microbial components on intestinal epithelial cells, https://doi.org/10.17605/OSF.IO/2RKEA ⁵¹

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).