Effects of ruminal LPS exposure on primary bovine ruminal epithelial cells

The objective of this study was to investigate the immunopotential of ruminal lipopolysaccharides ( LPS ) on cultured primary bovine ruminal epithelial cells ( RECs ). Primary bovine RECs were isolated from 6 yearling steers and grown in culture for 3 experiments. Experiment 1 aimed to determine the immunopotential of ruminal LPS, experiment 2 aimed to assess tolerance to chronic LPS exposure, and experiment 3 aimed to evaluate antagonistic interactions between ruminal and E.coli LPS. In experiments 1 and 2, RECs were exposed to nonpyrogenic water ( CON ), 20 μg/mL of E. coli LPS ( E.COLI ), 10 μg/mL of ruminal LPS ( RUM10 ), 20 μg/mL of ruminal LPS ( RUM20 ), and 40 μg/ mL of ruminal LPS ( RUM40 ), either continuously or intermittently. For the continuous exposure, RECs underwent a 6 h exposure, while for the intermittent exposure, the procedure was: ( 1 ) a 12 h continuous exposure to treatments followed by LPS removal for 24 h and then another 12 h of exposure ( RPT ), and ( 2 ) a 12 h continuous exposure to treatments followed by LPS removal and a recovery period of 36 h ( RCV ). In exp 3, RECs were exposed to nonpyrogenic water (0–0), 1 μg/mL E. coli LPS ( 0–1 ), 1 μg/mL ruminal LPS:1 μg/mL E. coli LPS ( 1–1 ), 10 μg/mL ruminal LPS:1 μg/mL E. coli LPS ( 10–1 ) and 50 μg/mL ruminal LPS:1 μg/mL E. coli LPS ( 50–1 ). Each experiment was done as a complete randomized block design with 6 REC donors. The REC donor was used as blocking factor. Each treatment had 2 technical replicates, and treatment responses for all data were analyzed with the MIXED procedure of SAS. For all experiments, total RNA was extracted from RECs and real-time qPCR was performed to determine the relative expression of genes for toll-like receptors ( TLR2 and TLR4 ), proinflammatory cytokines ( TNF , IL1B , and IL6 ), chemokines ( CXCL2 and CXCL8 ), growth factor-like cytokines ( CSF2 and TGFB1 ), and a lipid mediator ( PTGS2 ). In experiment 1, the targeted genes were upregulated by E.COLI, while all ruminal LPS treatments resulted in a lower transcript abundance. Regarding RPT, and RCV condition, in experiment 2, the expression of targeted genes was not affected or was at a lower abundance to E.COLI when compared with ruminal LPS treatments. Lastly, in experiment 3, all targeted genes resulted in lower or similar transcript abundance on all ruminal LPS ratios. Overall, our re-sults indicate that ruminal LPS have a limited capacity to activate the TLR4/NF-kB pathway and to induce the expression of inflammatory genes.


INTRODUCTION
Lipopolysaccharides (LPS) are outer membrane components of Gram-negative bacteria, which consisted of 3 covalently linked regions: the O-antigen, the core oligosaccharide, and the lipid A, whose structure primarily mediates the immunogenicity and the intensity of the intracellular signaling to the host (Sarmikasoglou and Faciola, 2021).Structural variations in lipid A are characterized by diversity in the acylation pattern and are associated with different immunogenicity (Sarmikasoglou and Faciola, 2022).Specifically, the hexa-acylated lipid A is found in virulent strains, whereas the penta-and tetra-acylated ones are found in non-virulent bacterial strains.This structural variation elicits strong or weak activation of the pattern recognition receptors (PRRs), especially toll like receptor 4 (TLR4) that are located on the host cell membrane.More specifically, the hexa-acylated lipid A elicits strong immune responses and the under-acylated lipid A elicits weak immune responses (Steimle et al., 2016).Additionally, under-acylated lipid As have been previously reported to antagonize the hexa-acylated ones on host cell receptor binding.Thus, LPS produced from bacteria, that express under-acylated lipid As can exhibit weak interaction with TLR4 and could maintain host homeostasis by inhibiting severe inflammation from hexa-acylated lipid A (Steimle et al., 2019).
Previous reports indicate that elevated levels of LPS in blood plasma are associated with heart failure, obesity, and metabolic diseases in humans (Fabbiano et al., 2018;Pastori et al., 2022), as well as ruminal acidosis in cattle (Nagaraja and Titgemeyer, 2007).Ruminal acidosis is a metabolic disorder that occurs when the consumption of rapidly fermentable carbohydrates replaces effective fiber, causing excessive accumulation of organic acids (volatile fatty acids and lactate) in the rumen.Ruminal pH ≤5.6 is the threshold of ruminal acidosis, where a pH 5 -5.6 indicates subacute acidosis (SARA), and pH below 5 is considered as acute acidosis.There is an excessive lysis of ruminal bacteria under SARA, increasing LPS concentration in the rumen.Under ruminal acidosis, LPS concentration has been reported to be 4-to 16-times greater than normal conditions, contributing to systemic inflammation by disrupting the epithelial barrier function, thus increasing the permeability of ruminal epithelium (Aschenbach and Gäbel, 2000;Meissner et al., 2017), and allowing the translocation of LPS into the peripheral tissue (Aschenbach et al., 2019).
Typically, there is a greater concentration of free ruminal LPS under ruminal acidosis; however, it has a wide variation (42,206 EU/mL -200,000 EU/mL) depending on the type and quality of diets fed (Gozho et al., 2007;Khafipour et al., 2009a;b;Stefanska et al., 2018) and free ruminal LPS concentration alone cannot determine ruminal acidosis.
Some PRRs, such as TLRs have been reported to be expressed in cells within the bovine ruminal tissue (Malmuthuge et al., 2012).More specifically, previous studies have found that primary rumen epithelial cells (REC) exhibit greater expression of TLR2 and TLR4, as well as, IL1B, TNF, and CXCL8, after stimulation with E. coli LPS (Zhang et al., 2016;Kent-Dennis et al., 2020), indicating the presence of TLR4 on their cell membranes.Considering that ruminal LPS is not structurally equivalent to E. coli LPS, primarily because the former exhibits under-acylated (low endotoxic) and the latter hexa-acylated (high endotoxic) lipid A structures (Sarmikasoglou et al., 2021), further investigation on the effect of ruminal microbiome derived LPS to initiate immune response to REC is warranted.We hypothesized that exposure to ruminal LPS would either not affect or would upregulate to a lower extent the expression of genes associated with the proinflammatory response and could prevent E. coli LPS to strongly stimulate TLRs activation.Therefore, we aimed to assess the changes in expression of genes related to proinflammatory responses when isolated REC were exposed to different concentrations, and frequencies of ruminal LPS, as well as investigate any potential antagonistic effects of the co-treatment between ruminal and E. coli LPS in different ratios.

Ethical Approval
The University of Florida Institutional Animal Use and Care Committee approved all the procedures for animal care, handling and slaughtering required for this experiment.

Ruminal Content Donors, and Collection
Ruminal content was obtained from 2 rumen-cannulated Angus crossbred steers (658 ± 79 kg BW).Before ruminal contents collection, the steers were subjected to grain-induced SARA to adapt the ruminal microbiome under SARA conditions.From d −7 to 0, steers were fed only bermudagrass hay (56.0%total digestible nutrients and 13.9% CP on a DM basis) ad libitum.From d 0 to 14, steers received 3.63 kg/d of corn grain (74.0%total starch on a DM basis) and ad libitum bermudagrass hay.One episode of ruminal pH between 5.2 to 5.6 for more than 180 min/d was used to characterize SARA induction (Owens et al., 1998;Schwaiger et al., 2013).To validate the SARA induction, we monitored the ruminal pH before (d -7 and 0) and after (d 7, 10 and 14) twice on a daily basis.The pH from the ruminal fluid on d 7, 10 and 14 before and 6 h after corn grain feeding was <5.6.The recent study did not measure the ruminal pH in shorter timeframes, however previous studies with feed-induced SARA reported that between 0 h and 6 h interval the pH remained below 5.6 both in vitro (Dai et al., 2019) and in vivo (Schwaiger et al., 2013).On d 14, ruminal contents were manually collected (7 L from each steer) and strained through 4 layers of cheesecloth into pre-warmed thermoses and promptly transported to the lab.

Ruminal Fluid Fractionation
The contents were strained again through 2-layers of cheesecloth, transferred into beakers, and immersed in ice for 15 min.The strained ruminal fluid (approximately 14 L; pool from ruminal content donors) was centrifuged (Sorvall RC-5B Refrigerated Superspeed Centrifuge, DuPont Instruments® Wilmington, DE) 3 times in succession to acquire the bacterial pellet, as previously described by Sarmikasoglou et al., (2022).Lastly, the bacterial pellets were transferred to pyrogen-free tubes, homogenized, and solubilized using Milli-Q water and stored in -80°C for later ruminal LPS extraction.

Ruminal Lipopolysaccharide Extraction
A modified hot-phenol extraction was utilized to extract LPS from ruminal bacteria obtained from the rumen-cannulated steers as described previously by Sarmikasoglou et al., (2021).Briefly, to isolate total LPS from ruminal fluid, the bacterial pellet was boiled and then treated with 90% phenol.The aqueous layer was then transported into a regenerated cellulose dialysis membrane (Fisherbrand) for further dialysis against Milli-Q until phenol was not detectable at 260 nm in Milli-Q.Dialyzed samples were then treated with 5 mM MgCl 2 followed by 20 μg/mL Dnase I (M0303s, New England Biolabs) for 2 h at 37°C to degrade contaminating DNA.After, 20 μg/mL Rnase H (T3018, New England Biolabs) was added for 2 h at 37°C, to degrade contaminating RNA and 30 mg/mL Proteinase K (Fisher BioReagents Proteinase K, Catalog No. BP1700-100) was added to degrade any proteins.The preparation was then lyophilized and crude LPS mass was determined.After lyophilization, dry samples were resuspended into Milli-Q water and centrifuged at 1110 × g for 10 min to remove any solids.The supernatant was treated with 50 mM acetic acid, 95% ethanol, transferred into ultracentrifuge tubes (Quick-Seal® Round-Top Polypropylene Tube), and then spun for 8 h at 4°C and 105,000 × g (Optima XE, Beckman Coulter Life Sciences, Indianapolis, IN).The LPS gels were resuspended in endotoxin-free water and lyophilized to determine the dry weight of pure LPS.To confirm the purity and normalization of ruminal-derived LPS, the final products were visualized with the Pierce Silver Stain Kit (Thermo Scientific) in accordance with the manufacturer's instructions.In all cases, the Pierce Silver Stain Kit indicated a purity identical to that of LPS purified from pure bacterial isolates.

REC Donor Animals and Tissue Collection
Tissues for all experiments described below were obtained from 6 yearling steers (approximately 10-mo-old) group housed in an outdoor, dry-lot pen.Steers had ad libitum access to water and pasture.Approximately 15 min post slaughtering, ruminal tissue from ventral sac was excised and washed with ice-cold Ca 2+-and Mg 2+-free phosphate buffered saline solution (PBS) containing antibiotic-antimycotic cocktail.Antibioticantimycotic cocktail was composed of 400 U/mL penicillin, 400 μg/mL streptomycin, 1 μg/mL amphotericin B (Thermo Fisher Scientific, Waltham, MA) and 240 U/mL nystatin (Sigma-Aldrich Chemicals, Burlington, MA), as final concentrations.Then, the washed ruminal tissue was submerged into the same solution and kept on ice until further REC isolation.

Ruminal Epithelial Cell Isolation and Cultivation
Isolation and cultivation of RECs were done essentially as previously described (Gálfi et al., 1980;Kent-Dennis et al., 2020).Ruminal papillae were cut off at their base, chopped into small pieces and washed with Ca 2+ -and Mg 2+ -free PBS containing an antibiotic-antimycotic cocktail at a final concentration of 100 U/mL penicillin, 100 μg/mL streptomycin, and 0.25 μg/mL amphotericin B (Thermo Fisher).Papillae were subjected to serial trypsinization using a 0.6 g/ mL trypsin-EDTA solution (0.25% trypsin and 0.02% EDTA; Sigma-Aldrich Chemicals, Burlington, MA).Papillae were agitated in the trypsin-EDTA solution at 37°C, and the resulted supernatant was collected and replaced with fresh solution every 30 min.The process was repeated 6-times total, and only the 3 to 6 fractions were separately strained through sterile gauze and spun at 200 × g for 5 min at RT. Cell pellets were resuspended in PBS with Ca 2+ and Mg 2+ , centrifuged again at 200 × g for 5 min at RT and washed with PBS with Ca 2+ and Mg 2+ .The acquired cell pellets from the aforementioned fractions were pooled and resuspended in 20 mL of M199 cell culture media (Sigma Aldrich, St. Louis, MO), 15% fetal bovine serum (Gemini Bio, West Sacramento, CA), 1X GlutaMAX (Thermo Scientific, Waltham, MA), 20 mM HEPES (Sigma Aldrich, St. Louis, MO), and an antibiotic-antimycotic cocktail (100 U/mL penicillin, 100 μg/mL streptomycin, 0.25 μg/mL amphotericin B, 240 U/mL nystatin, 50 mg/L gentamycin, and 100 mg/L kanamycin as final concentrations, respectively).Cell suspensions were seeded into 60-mm cell culture dishes coated with bovine collagen and placed in an incubator with constant 37°C and 5% CO 2 humidified atmosphere.The following day, cells were washed with PBS containing Ca 2+ and Mg 2+ and fresh media was added.The M199 cell culture media was replaced with minimum essential media (MEM) containing 10% fetal bovine serum (Gemini Bio, West Sacramento, CA), 1X GlutaMAX (Thermo Scientific, Waltham, MA), and an antibiotic-antimycotic cocktail (100 U/mL penicillin, 100 μg/mL streptomycin, and 0.25 μg/mL amphotericin B, as final concentrations, respectively).The MEM media was then replaced every other day.
Before the start of the experiment, we validated that the isolated RECs are not highly contaminated with fibroblasts.As a threshold of acceptable fibroblast contamination was that REC cultures exhibit <10% of CD90-positive cells, similar to previous studies (Kisselbach et al., 2009;Kent-Dennis et al., 2020).We cultured the isolated RECs from each REC-donor for 10 d and then a subset of cells from each REC-donor was trypsinized to detach the cells and resuspend them in PBS- 0.25% BSA solution.The cells were then spun at 13,000 rpm for 5 min and resuspended again to PBS-0.25% BSA solution.After the cells were filtered through a 40 μm cell strainer (Corning®, Corning, NY) and the filtrate was centrifuged at 13,000 rpm for 5 min, the cell pellet was resuspended in 3 mL PBS-0.25%BSA.One mL was mixed with 1 μL FITC conjugated mouse anti-human CD90 (BD Pharmingen, Franklin Lakes, NJ) to a 1 μL/mL final concentration.Then, the cells were left on ice for 20 min, washed with 10 mL of PBS-0.25%BSA solution and spun at 13,000 rpm for 5 min.The cell pellet was resuspended in 1 mL PBS-0.25%BSA solution and loaded to flow cytometer.Percentage of fibroblast contamination was then determined by flow cytometry using an Attune NxT flow cytometer (Thermo Scientific, Waltham, MA) based on a minimum of 10,000 events, and data analysis was performed with FlowJo version 10.0.7 (BD Biosciences, Franklin Lakes, NJ).All of our isolated RECs (n = 6) exhibited <10% of CD90-positive cells, (Supplemental Table 1).Mice fibroblasts were used as positive control and exhibited 14% of FITC-CD90 + cells (Supplemental Table 1).

LPS Stock Preparation.
Two different sources of LPS were used for this study, one from E. coli O111:B4 (Sigma-Aldrich Co., St. Louis, MO) and one from the ruminal LPS produced by the Gram-negative bacteria of the rumen.The concentration of E. coli LPS was 20 μg/mL (equal to 200, 000 EU/mL; based on Sigma-Aldrich protocol), while the concentrations of ruminal LPS were 10 μg/mL, 20 μg/mL, and 40 μg/mL.Stock solutions were prepared for both LPS sources, and the respective doses were derived from the same stocks for the entire experiment, more specifically, 20 mg of E. coli O111:B4 dissolved in 1 mL of sterile, nonpyrogenic water and split into 100 μL aliquots.Regarding the ruminal LPS doses, we dissolved 10 mg of ruminal LPS in 1 mL of sterile, nonpyrogenic water and split into 100 μL aliquots.All stocks were filter-sterilized and stored in -20°C till further use.
For a graphical representation of experimental conditions in experiments 1 and 2, see Figure 1.In experiment 1, the continuous exposure of REC to LPS treatments lasted for 6 h.The REC were passaged to 6-well (1 well per plate for each biological replicate of isolated REC), and 12-well (2 wells per plate for each biological replicate of isolated REC) plates, for viability assessment and gene expression analyses, respectively.Following exposure to LPS treatments, 1 well for each treatment from the 6-well plate was trypsinized and cells were immediately analyzed for viability using propidium iodide staining (Thermo Fisher Scientific, Waltham, MA) to determine the percentage of dead cells with a flow cytometer.The remaining 2 wells for each treatment from the 12-well plate were lysed in 1 mL of Trizol (Thermo Fisher Scientific, Waltham, MA), and stored at −80°C until further RNA extraction.Regarding the experiment 2, the intermittent exposure of REC to LPS treatments was for either 12 h of exposure followed by removal of LPS treatments for 24 h, and another 12 h of exposure (RPT), or 12 h of exposure followed by removal of LPS, and a recovery period of 36 h (RCV).Both intermittent exposures had 2 technical replicates per treatment, with 1 plate for each biological replicate of isolated REC.On the last time point (48 h) of RPT and RCV exposures, the 2 wells for each treatment were lysed in 1 mL of Trizol (Thermo Fisher Scientific, Waltham, MA), and stored at −80°C until further RNA extraction.
Before the beginning of experiments 1 and 2, REC from each REC-donor were seeded into 60-mm cell culture dishes similarly to as described above and were cultured for 3 d (approximately 90% confluency reached).Then the REC were passaged to 6-well, and 12-well plates, for viability assessment and gene expression analyses, respectively.For the viability assessment, cells were plated in 6-well plates (1 well per plate for each biological replicate of isolated REC) at a rate of 10 5 cells/mL (2 mL of medium per well).Cells were cultured for 2 d (approximately 90% confluency reached), and then the respective LPS treatments were dosed.For the gene expression analysis, cells were plated in 12-well plates (2 wells per plate for each biological replicate of isolated REC) at a rate of 10 5 cells/mL (1 mL of medium per well).Cells were cultured for 1 d (approximately 90% confluency reached), and then the respective LPS treatments were dosed.For experiment 2 the REC were dosed when approximately 80% confluency reached to avoid cell lysis and excessive debris release due to overgrowth during the 48 h span.Lastly, the REC isolated from each REC-donor were cultured as separate cell lines.For a graphic representation of the experimental conditions of experiment 2 see Figure 1.LPS Stock Preparation.Same stocks as those prepared for experiments 1 and 2 were used for the doses of experiment 3.
Experimental design.To determine any antagonistic interactions between the different LPS sources we co-treated the isolated RECs with different ratios of ruminal to E. coli LPS.More specifically, the RECs were exposed to nonpyrogenic water (0:0), 1 μg/mL E. coli LPS (0:1), 1 μg/mL total free-ruminal LPS:1 μg/mL E. coli LPS (1:1), 10 μg/mL total free-ruminal LPS:1 μg/mL E. coli LPS (10:1) and 50 μg/mL total free-ruminal LPS:1 μg/mL E. coli LPS (50:1).The exposure to different ratios of LPS sources was continuous for 6 h, with 2 technical replicates per treatment, and 1 plate for each biological replicate of isolated REC.Following exposure to treatments, 2 wells for each treatment were lysed in 1 mL of Trizol (Thermo Fisher Scientific, Waltham, MA), and stored at −80°C until further RNA extraction.For a graphic representation of the experimental conditions of experiments 3 see Figure 1.
Before the beginning of experiment 3, the REC from each REC-donor were seeded into 60-mm cell culture dishes similarly to as described above and were cultured for 3 d (approximately 90% confluency reached).Then the REC were passaged to 12-well plates (2 wells per plate for each biological replicate of isolated REC), for the respective gene expression analysis at a rate of 10 5 cells/mL (1 mL of medium per well).Cells were cul-tured for 2 d (approximately 90% confluency reached), and then the respective LPS treatments were dosed.Lastly, the REC isolated from each REC-donor were cultured as separate cell lines.

RNA Extraction, cDNA Synthesis and Analysis of Gene Expression
Total RNA was extracted using the Quick-RNA 96 kit (Zymo Research, Irvine, CA), according to manufacturer's protocol with minor modifications adapted to the nature of our samples.Briefly 200 μL of molecular grade chloroform (Thermo Fisher Scientific, Waltham, MA) were added to 1 mL cells resuspended into Trizol and mixed gently.After 5 min, the samples were centrifuged at 15, 000 × g for 15 min at 4°C.The resulted top layer was gently removed (approx.400 μL) and treated with 400 μL molecular grade absolute ethanol (Thermo Fisher Scientific, Waltham, MA).The For experiments 1 and 3, the RECs underwent to a continuous LPS exposure for 6 h (i).For experiment 2 the RECs underwent a12 h of exposure followed by removal of LPS for 24 h and then 12 h of exposure was repeated (ii), and 12 h of exposure followed by removal of LPS and a recovery period of 36 h (iii).Created with BioRender.com.added directly to the matrix and centrifuged.Then the eluted RNA was transferred to DNase/RNase Free tubes and stored in −80°C until further integrity assessment and reverse transcription to cDNA.
The integrity of the extracted RNA was assessed by the presence of distinct 18S and 28S bands in a 1% agarose gel.The Verso cDNA synthesis kit (Thermo Scientific, Waltham, MA) was used to reverse transcribe 1 μg of RNA.After RNA quantification with spectrophotometry, samples that exhibited RNA concentration greater than 100 ng/μL were diluted in nuclease-free water to achieve final concentration of 100 ng/μL, while samples that exhibited RNA concentration lower than 100 ng/μL were not diluted.Then a stock solution of master mix was prepared with 4 μL of 5x cDNA synthesis buffer, 2 μL of dNTP mix, 1 μL of RNA primer, 1 μL of RT enhancer, 1 μL of Verso enzyme mix, and 1 μL of nuclease-free water.Ten μL of master mix and 10 μL of RNA template were added to each PCR reaction tube.Samples were then, amplified using the MultiGene OptiMax Thermal Cycler (Labnet MultiGene, Edison, NJ) according to manufacturer's protocol.The resulting cDNA was diluted to 1:5 by adding 80 μL of nuclease-free water.
Quantitative real-time PCR was performed using 2 μL of cDNA and ran in duplicate using 18 μL reaction mix consisted of 500 nM of each forward and reverse primer and iTaq Universal SYBR Green Supermix (Bio-Rad, Hercules, CA).A Bio-Rad CFX Connect light cycler (Bio-Rad, Hercules, CA) was employed with an initial denaturation step at 95°C for 30 s, followed by 40 cycles at 95°C for 5 s, specific annealing temperature for 10 s, and final extension at 60°C for 30 s. Primers (Thermo Fisher Scientific, Waltham, MA) for housekeeping genes and genes of interest were designed using the NCBI database (Table 1), and were previously reported by Herath et al. (2009), Kent-Dennis et al. (2020), and Kent-Dennis and Penner (2021).Reference gene expression was stable across experimental treatments (P > 0.05; Supplemental Figure 1).The cycle threshold (C t ) values of the genes of interest were normalized to the geometric mean of the 2 housekeeping genes.Changes in gene expression were calculated as fold change using the formula 2 −ΔΔCt (Pasternak et al., 2020), and the treatment groups were held relative to control group.

Statistical Analysis
All statistical analyses were performed as a mixed model using the MIXED procedure of SAS 9.4 (SAS Institute Inc., Cary, NC).Analysis of qRT-PCR data was performed on 2 −ΔΔCt , after log-transformation.Significance was declared at P < 0.05.
The statistical model used was where y is the observations for dependent variables, µ is overall mean, T i is fixed effect of treatments, E j is the random effect of REC donor, and ε ij is the random error.Ruminal epithelial cell donor was the blocking factor and was considered as random effect.

Experiment 1:
Effect of Ruminal and E. coli LPS on Cytotoxicity and Expression of Genes Associated with Proinflammatory Response of RECs under continuous exposure Cytotoxicity of RECs.No treatment effects were observed on the percentage of dead RECs (P = 0.83; Figure 2).Our cytotoxicity results indicate that all of the LPS treatments exhibit a low level of cytotoxic activity (average = 4.43%), and do not differ with the CON.Therefore, our LPS treatments do not affect the viability of our RECs thus the gene expression analyses below are based on the same number of viable RECs.
Expression of Genes Associated with Proinflammatory Response of RECs.The immunostimulatory potency of ruminal LPS to RECs was assessed in gene expression level by exposing primary RECs to 3-levels of ruminal LPS.Lipopolysaccharides from E. coli were used as positive control.
Toll-like receptor genes (TLR2, TLR4) exhibited similar expression patterns.More specifically, the gene expression of TLR2 increased (P < 0.01), when each of the LPS treatment was compared with CON.In addition, TLR2 gene expression was lower (P < 0.01), when RUM10, and RUM20, were compared with E.COLI, respectively.Lastly, no effects were detected among the LPS treatments (Figure 3A).The gene expression of TLR4 was greater (P < 0.01), when each of the LPS treatments, were compared with CON.In addition, TLR4 gene expression was greater (P < 0.01), when E.COLI LPS was compared with each of the ruminal LPS treatments.No effects were detected among the LPS treatments (Figure 3B).
Regarding the gene expression of proinflammatory cytokine genes, we targeted the expression of genes encoding the TNF, IL1B, and IL6.More specifically, the gene expression of TNF increased (P < 0.01), when each of the LPS treatments were compared with CON.Also, TNF gene expression was greater (P < 0.01), when E.COLI LPS was compared with each of the ruminal LPS treatments.Lastly, no differences were detected among the LPS treatments (Figure 3C).The In addition, IL1B gene expression was greater (P < 0.01), when E.COLI LPS was compared with each of the ruminal LPS treatments.Lastly, IL1B gene expression was greater (P < 0.01), when RUM20 and RUM40 were compared with RUM10, respectively (Figure 3D).Considering the gene expression of IL6, an increase (P < 0.01), was observed when each of the LPS treatments were compared with CON.No effects were detected among LPS treatments (Figure 3E).
The gene expression of CXCL2, was greater (P < 0.01), when each LPS treatment was compared with CON.Also, CXCL2 gene expression exhibited an increase (P < 0.01) when E.COLI LPS was compared with RUM10.Lastly, the gene expression of CXCL2 increased (P < 0.01), when RUM20 was compared with RUM10 (Figure 3F).An increase was observed in the expression of CXCL8, when E.COLI LPS (P < 0.01), RUM20 (P < 0.01), and RUM40 (P < 0.01), were compared with CON, respectively.Each of the ruminal LPS treatments exhibited a decrease (P < 0.01), when compared with E.COLI.No effects were detected between ruminal LPS treatments (Figure 3G).
Growth factor-like cytokine genes (CSF2, and TGFB1), and the lipid mediator of inflammation gene (PTGS2) expression were quantified.More specifically, the gene expression of CSF2, increased (P < 0.01) when     each of the LPS treatments were compared with CON.In addition, CSF2 gene expression increased (P < 0.01), when E.COLI LPS was compared with RUM10.Lastly, the gene expression of CSF2 increased (P < 0.01), when RUM20 was compared with RUM10 (Figure 3H).Regarding the gene expression of TGFB1, no effects (P = 0.15) were detected between each LPS treatments and CON, as well as between LPS treatments (Figure 3I).
An increase (P < 0.01) observed in the expression of PTGS2, when E.COLI LPS, and RUM40, were compared with CON, respectively.Furthermore, PTGS2 gene expression exhibited an increase (P < 0.01) when E.COLI LPS was compared with RUM10 and RUM20 treatments, respectively.Lastly, the gene expression of PTGS2 exhibited an increase (P < 0.01) when RUM40, was compared with RUM10 and RUM20 treatments, respectively (Figure 3J).

Effect of Ruminal and E. coli LPS and Expression of Genes Associated with Proinflammatory Response of RECs under intermittent manner.
To evaluate any potential tolerance effects of primary RECs to ruminal LPS we determined if the exposure to ruminal LPS, under a repeated and a recovery manner induces a sustained proinflammatory response.

Expression of Genes Associated with Proinflammatory Response of RECs under RPT exposure to LPS sources. The potential tolerance of
RECs to LPS stimulation under a repeated manner was assessed in gene expression level by exposing primary RECs to 3-levels of ruminal LPS.Lipopolysaccharides from E. coli were used as positive control.
Gene expression of TLR2 was not affected by any of the LPS treatments (P = 0.79), while the gene expression of TLR4 was increased (P < 0.01), when compared with CON and E.COLI.Lastly, the gene expression of TLR4 was increased (P < 0.01) when E.COLI was compared with each of the ruminal LPS treatments (Figure 4A and B).
Regarding the gene expression of proinflammatory cytokine genes, we targeted the expression of genes encoding the TNF, IL1B, and IL6.More specifically, the gene expression of TNF, increased (P < 0.01) in E.COLI, when compared with CON.Also, TNF gene expression exhibited an increase (P < 0.01) when E.COLI LPS was compared with each of the ruminal LPS treatments.No effects were detected among LPS treatments (Figure 4C).The gene expression of IL1B, increased (P < 0.01) in each of the LPS treatments, when compared with CON.In addition, IL1B gene expression exhibited an increase (P < 0.01) when E.COLI LPS was compared with each of the ruminal LPS treat-ments.No effects were detected among LPS treatments (Figure 4D).Considering the gene expression of IL6, an increase (P < 0.01) observed when E.COLI, was compared with CON.In addition, IL6 gene expression exhibited an increase (P < 0.01) when E.COLI LPS was compared with each of the ruminal LPS treatments.No effects were detected among LPS treatments (Figure 4E).
About the chemokine genes we selected to assess the gene expression of genes encoding the CXCL2, and CXCL8.The gene expression of CXCL2, exhibited an increase (P < 0.01) when each LPS treatment, was compared with CON.Furthermore, the E.COLI had a greater (P < 0.01) CXCL2 gene expression, when compared with each of the ruminal LPS treatments.No effects were detected among LPS treatments (Figure 4F).Regarding the gene expression of CXCL8, an increase (P < 0.01) was observed when each of the LPS treatments were compared with CON.Furthermore, CXCL8 gene expression increased (P < 0.01) when E.COLI LPS was compared with each of the ruminal LPS treatments.No effects were detected between LPS treatments (Figure 4G).
Growth factor-like cytokine genes (CSF2, and TGFB1), and the lipid mediator of inflammation gene (PTGS2) expression were quantified.Concerning the gene expression of CSF2 (P = 0.27), and TGFB1 (P = 0.74), no effects were detected among LPS treatments and CON, as well as between LPS treatments (Figure 4H and 4I).Regarding the gene expression of PTGS2, increased (P = 0.01) was observed when each of the LPS treatments were compared with CON.No effects were detected between LPS treatments (Figure 4J).

Expression of Genes Associated with Proinflammatory Response of RECs under RCV exposure to LPS sources.
To evaluate the ability of primary RECs to sustain the expression of genes associated with proinflammatory response after a continuous exposure to LPS sources, we pursued a 12 h continuous exposure of RECs to treatments followed by removal of LPS and a recovery period of 36 h.
Tolerogenic receptor genes (TLR2, TLR4) exhibited similar expression patterns.More specifically, no effects were detected on the gene expression of TLR2 (P = 0.19) and TLR4 (P = 0.34) for any of the treatments (Figure 5A and 5B).Regarding the gene expression of proinflammatory cytokine genes, we targeted the expression of genes encoding the TNF, IL1B, and IL6.More specifically, the gene expression of TNF increased (P = 0.02) when E.COLI was compared with CON.Also, TNF gene expression increased (P = 0.02), when E.COLI LPS was compared with RUM10 and RUM40, respectively.No effects were detected between LPS treatments (Figure 5C).The gene expression of IL1B, increased (P < 0.01) when E.COLI LPS, and RUM20, were compared with CON, respectively.In addition, IL1B gene expression increased (P < 0.01) when E.COLI LPS was compared with each of the ruminal LPS treatments.No effects were detected among LPS treatments (Figure 5D).Considering the gene expression of IL6, increased (P < 0.01) was observed when E.COLI LPS, and RUM40, were compared with CON, respectively.In addition, IL6 gene expression exhibited an increase (P < 0.01) when E.COLI LPS was compared with RUM10 and RUM20 treatments, respectively.(Figure 5E).
About the chemokine genes we selected to assess the gene expression of genes encoding the CXCL2, and CXCL8.The gene expression of CXCL2, exhibited an increase (P < 0.01) when E.COLI, and RUM40 were compared with CON, respectively.Furthermore, the E.COLI had a greater (P < 0.01) CXCL2 gene expression, when compared with each of the ruminal LPS treatments (Figure 5F).Regarding the gene expression of CXCL8, an increase (P < 0.01) was observed when E.COLI LPS, was compared with CON.Furthermore, CXCL8 gene expression exhibited an increase (P < 0.01) when E.COLI LPS was compared with each of the ruminal LPS treatments (Figure 5G).
Growth factor-like cytokine genes (CSF2, and TGFB1), and the lipid mediator of inflammation gene (PTGS2) expression were quantified.More specifically, the gene expression of CSF2, increased (P < 0.01) when E.COLI, RUM20 and RUM40 were compared with CON, respectively.In addition, CSF2 gene expression exhibited an increase (P < 0.01) when RUM10 was compared with RUM40 (Figure 5H).Lastly, no effects were detected on the gene expression of TGFB1 (P = 0.99) and PTGS2 (P = 0.13) for any of the treatments (Figure 5I and 5J).

Effect of co-treatment Ruminal and E. coli LPS on Expression of Genes Associated with Proinflammatory
Response of RECs under continuous manner.
Expression of Genes Associated with Proinflammatory Response of RECs under co-treatment exposure to LPS sources.
The potential antagonistic interaction between ruminal and E. coli LPS to function at RECs was assessed in gene expression level by co-treating primary RECs with different ratios (0:1, 1:1, 10:1, and 50:1) of ruminal to E. coli LPS.

Recognition of ruminal LPS by RECs
Highly fermentable diets provide more potential energy for maintenance and productive traits to cattle; however, concurrently these diets increase the potential of organic acid accumulation in the rumen (Aschenbach et al., 2011), as well as establish hyperosmotic conditions at the ruminal epithelial barrier (Owens et al., 1998;Nagaraja and Titgemeyer, 2007).Additionally, as the fermentation rate in the rumen increases, it results into greater bacterial lysis rate and therefore greater levels of ruminal LPS from the Gram-negative bacteria (Dai et al., 2019).The aforementioned conditions increase the risk of SARA development and could further disrupt the rumen epithelial barriers, thus allowing ruminal LPS to translocate across ruminal epithelium and affect epithelial function (Plaizier et al., 2018).Recognition of ruminal LPS occurs by the tolerogenic receptors that are present on the cell surface of RECs (Dionissopoulos et al., 2012).Presence of tolerogenic receptors (TLR2 and TLR4) have been reported in ruminal epithelial tissue (Chen et al., 2012;Dionissopoulos et al., 2012;Liu et al., 2013).
For experiments 1 and 2, a range of ruminal LPS concentrations were selected to reflect the concentrations reported in the literature, while a concentration of 20 μg/mL E. coli LPS equivalent to 200, 000 EU/ mL was selected as positive control (Plaizier et al., 2012;Kent-Dennis et al., 2020).In experiment 1, RECs were exposed for 6 h to different LPS sources, under a continuous manner.The gene expression of tolerogenic receptors were similar to previous studies using with RECs.More specifically, E.COLI treatment upregulated the tolerogenic receptors encoding for TLR2 and TLR4 genes (Kent-Dennis et al., 2020).This observation further supports the presence of those receptors on RECs surface and indicates that 20 μg/mL is an adequate dose for their stimulation.Additionally, we observed that both TLR2 and TLR4 gene expression were upregulated in the presence of ruminal LPS, indicating the capacity of ruminal LPS to stimulate them and further activate the NF-κB pathway.Overall, those findings support previous studies conducted with fecal LPS on the human gut microbiome that found an absence of direct signaling through these LPS-sensing signaling pathways (Vatanen et al., 2016;d'Hennezel et al., 2017).
In experiment 2, the RECs underwent 2 different challenges (RPT and RCV) of intermittent LPS exposures.Previous studies dosed 1,000 and 50,000 EU/mL of E. coli LPS on RECs in a RPT and RCV manner and found that the TLR2 gene upregulated in RPT and was not affected during RCV challenge.The same study also reported that the TLR4 gene expression exhibited no difference on the non-treated group during either RPT or RCV exposures (Kent-Dennis et al., 2020).In our study, the gene encoding for TLR2 exhibited no difference on their expressions during both RPT and RCV, which agrees with the aforementioned study, while the gene encoding for TLR4 was upregulated during RPT and not affected during RCV.Interestingly the expression of the same genes was not affected by the ruminal LPS treatments.Those findings indicate that, similarly to intestinal epithelial cells (Lotz et al., 2007;Lee et al., 2008), RECs potentially develop tolerance to immunostimulatory LPS, such as E. coli LPS.More specifically, data suggest that under strong stimulation of innate immune response, RECs develop and sustain tolerance during future exposures to immunostimulatory LPS.However, a weak stimulation (e.g., ruminal LPS) does have the potential to result in the development of a greater tolerance to ruminal LPS in future exposures.This finding indicates that ruminal LPS exhibit a protective interaction with ruminal epithelium, by mitigating its sensing by RECs tolerogenic receptors.
In experiment 3, RECs were co-treated with 1 μg/mL E. coli LPS and increasing doses of ruminal LPS for 6 h continuously to assess their ability to interfere with the immune stimulation at the gene level.Previous studies co-treated human peripheral blood mononuclear cells (hPBMC), with 1 ng/mL E. coli LPS and increasing doses (10, 100, 1000-fold) of human fecal LPS for 20 h continuously.They found that inhibition of IL6 and IL1B expression were achieve at ratios ≥10-1 and 1-1 fecal LPS to E.coli LPS, respectively, thus suggesting that fecal LPS is a potent inhibitor of TLR4 stimulation (d'Hennezel et al., 2017).To our knowledge, our study is the first to pursue a co-treatment challenge between E. coli LPS and ruminal LPS on RECs and assess the immunomodulatory properties of ruminal LPS.We found that the TLR2 gene was upregulated in 0-1, and 50-1 ratios, respectively; however, ratios of 1-1, and 10-1 exhibited lower gene expression compared with 0-1, which indicates that ruminal LPS inhibits the TLR2 stimulation, and that excessive amount of ruminal LPS would potentially have similar properties to E. coli LPS.The TLR4 gene expression of the 1-1, 10-1, and 50-1 ratios inhibited TLR4 stimulation, which corroborates with the previous aforementioned studies conducted with hPBMC (d'Hennezel et al., 2017).Overall, it appears that ruminal LPS has the ability to interfere with E. coli LPS, because it is composed mostly of under-acylated lipid As that could bind with greater affinity to the tolerogenic receptors of RECs, resulting into weaker activation (Steimle et al., 2019).
In experiment 1, primary RECs were exposed for 6 h continuously to a range of ruminal LPS concentrations to further evaluate the expression of genes associated with proinflammatory response, while E. coli LPS was used as positive control.Gene expression of TNF, IL1B, and IL6 were upregulated by the ruminal LPS treatments.
Similarly, the expression of TNF, IL1B, and IL6 were upregulated in the presence of E. coli LPS (Kent-Dennis et al., 2020); however, the upregulation was greater for TNF and IL1B compared with ruminal LPS treatments.The ability of REC to express proinflammatory genes has been previously demonstrated (Kent-Dennis et al., 2020); however, it seems that in presence of ruminal LPS, RECs are affected at a lower extent as indicated by the RNA quantity of those proinflammatory genes.This outcome suggests that ruminal LPS is a weak immunostimulant that initiates the development of inflammatory responses.
In experiment 2, RECs underwent 2 different challenges (RPT and RCV) of intermittent LPS exposures.Gene expression of TNF, IL1B, and IL6 were not affected or were upregulated at a lower extent by the ruminal LPS treatments, compared with E.COLI, during either RPT or RCV challenges.Also, the expression of TNF, and IL1B was upregulated in the presence of E. coli LPS, during both RPT and RCV, similarly to previous studies (Kent-Dennis et al., 2020).Also, IL6 expression was upregulated in the presence of E. coli LPS, during both RPT and RCV.These data indicate that ruminal LPS is a weak immunostimulant of RECs and has the ability to initiate the development of a tolerance to ruminal LPS during future exposures.
Additionally, the different effects on IL6 gene expression indicate a dual immunological role of IL6 in RECs.Previous studies have highlighted that IL6 could act as pro-and anti-inflammatory stimulant.Concerning its pro-inflammatory function, IL6 stimulates antibody production and effector T-cell development, while its anti-inflammatory functions are related to exerting the induction of acute phase proteins synthesis, such as fibrinogen and serum amyloid A (Xing et al., 1998;Tanaka et al., 2014).Thus reducing the expression of proinflammatory cytokines, such as TNF and IL1B, and upregulation of their antagonists, such as IL-1 receptor antagonist protein, and TNF soluble receptor (Tanaka et al., 2014).Previously, to our knowledge, no published work focused on the role of IL6 on RECs; however, our data showed that RECs have the capacity to express IL6.Also, IL6 in RECs seems to exert a specific suppressive role to TNF, potentially by stimulating the production of anti-inflammatory molecules, such as TNF soluble receptors.Future studies should focus on elucidating the role of IL6 in RECs and investigate which anti-inflammatory molecules promote their expression.
In experiment 3, RECs were co-treated with ratios from different LPS sources.Gene expression of TNF, and IL1B was lower, while gene expression of IL6 was similar to 0-1 dose compared with the ratios of ≥1 μg/ mL ruminal LPS.These data further support the development of an antagonistic interaction between E. coli and ruminal LPS.Lastly, as expected the 0-1 dose that consisted of E.coli LPS alone stimulated the expression of TNF, IL1B, and IL6, in alignment with previous studies (Kent-Dennis et al., 2020).
Chemokines exhibit antimicrobial properties and are involved in immunoregulatory and inflammatory processes.More specifically, CXCL2 and CXCL8 are chemokines that play a major role in the recruitment of leukocytes under inflammation conditions.More specifically, both of those chemokines are associated with the recruitment and infiltration of neutrophils, while CXCL2 has been also reported as a mucosal lymphocyte recruiter (Harada et al., 1994;Ohtsuka et al., 2001).
Under a continuous exposure to different LPS sources (Experiment 1), gene expression of CXCL2 and CXCL8, was upregulated by the ruminal LPS treatments, while they were upregulated to a greater extent in the presence of E.coli LPS, similarly to previous studies (Kent-Dennis et al., 2020).This outcome suggests that in the rumen tissue, there are RECs that encode chemokine genes in the presence of immunostimulants, that further promote the recruitment of neutrophils at the site of infection and stimulate neutrophil oxidative burst and granule release to eliminate the inflammatory response.Furthermore, in cases of weak immuno- stimulants, such as ruminal LPS, the chemokine gene expression remains at lower levels; therefore, it appears that ruminal LPS do not have the potential to induce excessive inflammatory response in cattle.Those findings align well with the observed expression pattern of TLR2, TLR4, TNF, IL1B, and IL6 indicating a consistent transcript abundance.
In experiment 2, RECs underwent 2 different challenges (RPT and RCV) of intermittent LPS exposures.Gene expression of CXCL2 and CXCL8 were upregulated or not affected by the ruminal LPS treatments, during either RPT or RCV challenges.Additionally, during both challenges, CXCL2 and CXCL8 gene expression were upregulated to a greater extent to the remaining treatments in effect of E. coli LPS, which is a finding that aligns well with data from previous studies (Kent-Dennis et al., 2020).Those findings further indicate that RECs are able to develop a sustained tolerance to ruminal LPS.
Results from the co-treatment challenge (Experiment 3) showed that ratios of 1-1, and 10-1 upregulated the expression of both CXCL2 and CXCL8.Those data suggest that the presence of LPS, independently of the ratio between ruminal and E. coli LPS, induces the expression of CXCL2 and CXCL8; however, the presence of LPS at greater levels (i.e., 50-1) mitigate that stimulatory potential.
Growth factor-like cytokines have been previously reported to be produced by intestinal epithelial cells during acute inflammation, and are associated with promoting and regulating immune cells functions (Hooper, 2015).To expand more on that aspect, we evaluated the expression of CSF2 and TGFB1 in REC.
The expression of CSF2 under inflammatory conditions, has been reported in different cell types, such as intestinal epithelial cells (Egea et al., 2013).After its production, CSF2 can induce differentiation and survival in myeloid cells, such as mononuclear phagocytes (Mortha et al., 2014).A previous study, conducted with colonic epithelial cells, found that the production of CSF2 is associated with the restoration of epithelial cells and recovery after colitis induction (Egea et al., 2013).Regarding the expression of CSF2 in REC, previous reports have shown that similarly to intestinal epithelial cells, and in agreement with our findings, CSF2 is expressed in REC after exposure to strong immunostimulatory LPS sources (Kent-Dennis et al., 2020).More specifically, our data from the continuous exposure to LPS sources (Experiments 1 and 3) indicate that doses of 1 -20 μg/mL of E.coli LPS can stimulate gene expression of gene encoding for CSF2.However, in the presence of ruminal LPS, either in single or co-treatment, the gene expression of CSF2 was not affected or lower than the E. coli treatments.In experi-ment 2, RECs underwent 2 different challenges (RPT and RCV) in which the LPS treatments did not alter (RPT challenge), while upregulated (RCV challenge) the expression of CSF2.The findings from E.coli LPS during RCV challenge are aligned with previous studies in RECs (Kent-Dennis et al., 2020), and indicate that RECs exhibit sustained tolerance.
In regard to the expression of TGFB1, we found no differences in gene expression in any of the challenges conducted.From the literature, activation and production of TGFB1 has been previously reported in endotoxin stimulated murine peritoneal macrophages (Nunes et al., 1995); however, no upregulation was reported in RECs (Kent-Dennis et al., 2020).Our findings are in agreement with previous REC studies and in contrast to studies with macrophages, suggesting that TGFB1 is not upregulated in the presence of either weak or strong immunostimulants, thus RECs seem to not have the capability to stimulate gene expression and potentially the production of TGFB1.Further research should focus on elucidating the pattern of TGFB1 expression in RECs, under intense inflammation challenge.In the presence of ruminal LPS, the production of TGFB1 does not seem to be stimulated.
Lipid mediators of inflammation are derived from omega-6 polyunsaturated fatty acids, such as prostaglandins, and constitute vigorous enhancers of innate and adaptive immune response that are commonly present in several inflammatory disorders (Rogerio et al., 2015).To expand more on that aspect, we evaluated the expression of PTGS2 in RECs.
The PTGS2 is a key enzyme in the initial step of prostaglandin synthesis in vivo, because it catalyzes arachidonic acid to prostaglandins (Rumzhum and Ammit, 2016).In cases of sustained immune stimulation, an overproduction of PTGS2 and consequently elevated levels of prostaglandins have been observed (Hida et al., 2000).Overproduction of PTGS2 is commonly observed in carcinogenesis initiation and progression via promoting angiogenesis, metastasis, and immunosuppression (Dohadwala et al., 2001;Mao et al., 2003).Gene expression of PTGS2 under inflammation has been reported in RECs; therefore, it is evident that RECs have the capability to produce PTGS2 (Kent-Dennis et al., 2020;Kent-Dennis and Penner, 2021).
In our study, gene expression of PTGS2 was upregulated in the presence of E. coli LPS, a finding that comes in agreement with previous studies that exposed primary RECs to 10,000 and 50,000 EU/ mL of E. coli LPS (Kent-Dennis et al., 2020;Kent-Dennis and Penner, 2021).Those findings indicate that hexa-acylated LPS synthesized from species like E. coli has the potential to induce strong stimulation of the immune response in RECs, which later stimulate the activation of genes such as PTGS2 that further reduces inflammation.On the other hand, the gene expression of PTGS2 was not affected in the presence of lower doses of ruminal LPS (<50μg/mL), which further validates its weak immunopotential.Lastly, our findings in experiment 3, indicate that potential antagonistic interactions exist between hexa-and under-acylated lipid As, in RECs.

CONCLUSIONS
Our findings demonstrate that in vitro, ruminal LPS would have weak immunopotential, and would tend to prevent, rather than exacerbate, inflammation, however further validation in vivo should be qualified.Furthermore, our findings demonstrate that repeated exposure of RECs to ruminal LPS, results in the development of a sustained tolerance, thus further protecting the cells against septic shock in cases of chronic exposure to LPS.Overall, our results demonstrate that ruminal LPS have a limited capacity to induce the expression of inflammatory genes.Collectively, these findings shed light on the mechanisms that link different LPS sources and inflammation response in ruminal epithelial cells.
Sarmikasoglou et al.: RUMINAL LPS AND RUMINAL EPITHELIAL CELLS Experiment 3: Effect of Co-Treatment Ruminal and E. coli LPS on Expression of Genes Associated with Proinflammatory Response of RECs under Continuous Manner.
Figure 1.Schematic diagram of experimental conditions of experiments 1 -3.For experiments 1 and 3, the RECs underwent to a continuous LPS exposure for 6 h (i).For experiment 2 the RECs underwent a12 h of exposure followed by removal of LPS for 24 h and then 12 h of exposure was repeated (ii), and 12 h of exposure followed by removal of LPS and a recovery period of 36 h (iii).Created with BioRender.com.
Sarmikasoglou et al.: RUMINAL LPS AND RUMINAL EPITHELIAL CELLSgene expression of IL1B was greater (P < 0.01), when each of the LPS treatments were compared with CON.
Sarmikasoglou et al.: RUMINAL LPS AND RUMINAL EPITHELIAL CELLS Sarmikasoglou et al.: RUMINAL LPS AND RUMINAL EPITHELIAL CELLS Sarmikasoglou et al.: RUMINAL LPS AND RUMINAL EPITHELIAL CELLS
Sarmikasoglou et al.: RUMINAL LPS AND RUMINAL EPITHELIAL CELLS Sarmikasoglou et al.: RUMINAL LPS AND RUMINAL EPITHELIAL CELLS Sarmikasoglou et al.: RUMINAL LPS AND RUMINAL EPITHELIAL CELLS